DO NOT USE THIS SOFTWARE UNTIL YOU HAVE READ THIS LICENSE AGREEMENT.

BY
USING THE SOFTWARE (OR AUTHORIZING ANY OTHER PERSON TO DO SO), YOU AGREE TO
ACCEPT AND ABIDE BY ALL THE TERMS OF THIS LICENSE AGREEMENT. THE LICENSE
AGREEMENT REFERS TO YOU AS "CUSTOMER".

1 DEFINITIONS

LICENSED SOFTWARE shall mean solely the object code of the following versions of SCO’s UNIX oper-
ating system software: Versions V5, V6, and V7 adapted for PDP-11 computer systems.

SCO INTELLECTUAL PROPERTY RIGHTS shall mean patent, copyright and trade secret rights of The
Santa Cruz Operation, Inc. ("SCO") in the LICENSED SOFTWARE.

2 LICENSE GRANT

SCO grants to CUSTOMER a worldwide, non-exclusive, royalty-free license under SCO INTELLEC-
TUAL PROPERTY RIGHTS to reproduce, modify, and use the LICENSED SOFTWARE solely for non-
commercial uses, and to distribute the LICENSED SOFTWARE to a third party who is also bound by the
terms and conditions of this License Agreement. For this purpose, any distribution of a copy of
LICENSED SOFTWARE shall be accompanied by a verbatim copy of this License Agreement. CUS-
TOMER shall not reverse compile, reverse assemble, or otherwise reverse engineer the LICENSED SOFT-
WARE or any portion thereof, or remove or alter any copyright or other proprietary notices appearing in the
LICENSED SOFTWARE.

3 "AS IS" DELIVERY

CUSTOMER acknowledges that it accepts the LICENSED SOFTWARE "AS IS", and that SCO is under
no obligation to supply any support, training, consulting, or other services for the LICENSED SOFT-
WARE.

4 WARRANTY DISCLAIMER/LIMITATION OF LIABILITY

SCO DISCLAIMS ALL WARRANTIES WITH REGARD TO ANY PORTION OF THE LICENSED
SOFTWARE, INCLUDING BUT NOT LIMITED TO ALL IMPLIED WARRANTIES OF MER-
CHANTABILITY AND FITNESS, AND ANY WARRANTY THAT THE LICENSED SOFTWARE
DOES NOT VIOLATE ANY PROPRIETARY RIGHT OF A THIRD PARTY. IN NO EVENT SHALL
SCO BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY
DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
AN ACTION OF CONTRACT, NEGLIGENCE, OR ANY OTHER THEORY ARISING OUT OF OR IN
CONNECTION WITH THE USE OR PERFORMANCE OF ANY PORTION OF THE LICENSED SOFT-
WARE.

5 INDEMNITY FROM CUSTOMER

CUSTOMER will hold SCO harmless against all liabilities, demands, damages, expenses, or losses arising
out of use by CUSTOMER of LICENSED SOFTWARE.

6 TERMINATION

6.1 This Agreement shall be effective unless and until terminated.

6.2 Either party may terminate this Agreement at any time upon 30 days written notice. In addition, if
CUSTOMER shall fail to perform or observe any of the terms and conditions to be performed or observed
by it under this License Agreement, SCO may in its sole discretion thereafter elect, by written notice, to
 -2-

immediately terminate this License Agreement, together with all rights granted herein to CUSTOMER.

6.3 Termination of this Agreement shall not release CUSTOMER from any liability which shall have
accrued in favor of SCO at the time such termination becomes effective, nor from CUSTOMER’s obliga-
tion not to reverse engineer any portion of the LICENSED SOFTWARE.

6.4 In the event of any termination of this Agreement for any reason, CUSTOMER shall immediately
destroy all whole or partial copies of the LICENSED SOFTWARE in CUSTOMER’s possession, custody,
or control. In addition, CUSTOMER shall cease to use the LICENSED SOFTWARE as of the date of ter-
mination.

7 GENERAL TERMS

7.1 This License Agreement shall be governed by the substantive laws of the State of California.

7.2 CUSTOMER shall not assign any rights or delegate any of its obligations under this License Agreement
without the written consent of SCO.

7.3 CUSTOMER assures SCO that, with respect to any LICENSED SOFTWARE obtained under this
Agreement, CUSTOMER will comply with all applicable governmental export laws and regulations.

7.4 Waiver of any breach hereunder by CUSTOMER may be effected only by a writing signed by SCO and
shall not constitute a waiver of any other breach.

7.5 CUSTOMER understands and agrees that this License Agreement is the complete and exclusive state-
ment of the understanding between the parties with respect to its subject matter, and supersedes all commu-
nications and understanding between the parties relating to such subject matter.
-

UNIX PROGRAMMER’S MANUAL

Sixth Edition

K. Thompson
D. M. Ritchie

May, 1975
-

This manual was set by a Graphic Systems phototypeset-

ter driven by the troff formatting program operating un-
der the UNIX system. The text of the manual was pre-
pared using the ed text editor.
-

PREFACE
to the Sixth Edition

We are grateful to L. L. Cherry, R. C. Haight, S. C. Johnson, B. W. Kernighan, M. E. Lesk, and E. N. Pin-

son for their contributions to the system software, and to L. E. McMahon for software and for his contribu-
tions to this manual. We are particularly appreciative of the invaluable technical, editorial, and administra-
tive efforts of J. F. Ossanna, M. D. McIlroy, and R. Morris. They all contributed greatly to the stock of
UNIX software and to this manual. Their inventiveness, thoughtful criticism, and ungrudging support in-
creased immeasurably not only whatever success the UNIX system enjoys, but also our own enjoyment in its
creation.

1
-

INTRODUCTION TO THIS MANUAL

This manual gives descriptions of the publicly available features of UNIX. It provides neither a general
overview − see ‘‘The UNIX Time-sharing System’’ (Comm. ACM 17 7, July 1974, pp. 365-375) for that −
nor details of the implementation of the system, which remain to be disclosed.
Within the area it surveys, this manual attempts to be as complete and timely as possible. A conscious de-
cision was made to describe each program in exactly the state it was in at the time its manual section was
prepared. In particular, the desire to describe something as it should be, not as it is, was resisted. In-
evitably, this means that many sections will soon be out of date.
This manual is divided into eight sections:
I. Commands
II. System calls
III. Subroutines
IV. Special files
V. File formats and conventions
VI. User-maintained programs
VII. User-maintained subroutines
VIII. Maintenance
Commands are programs intended to be invoked directly by the user, in contradistinction to subroutines,
which are intended to be called by the user’s programs. Commands generally reside in directory /bin (for
bin ary programs). Some programs also reside in / usr/ bin, to save space in /bin. These directories are
searched automatically by the command interpreter.
System calls are entries into the UNIX supervisor. In assembly language, they are coded with the use of the
opcode sys, a synonym for the trap instruction. In this edition, the C language interface routines to the sys-
tem calls have been incorporated in section II.
A small assortment of subroutines is available; they are described in section III. The binary form of most
of them is kept in the system library / lib/ liba.a. The subroutines available from C and from Fortran are al-
so included; they reside in / lib/ libc.a and / lib/ libf.a respectively.
The special files section IV discusses the characteristics of each system ‘‘file’’ which actually refers to an
I/O device. The names in this section refer to the DEC device names for the hardware, instead of the names
of the special files themselves.
The file formats and conventions section V documents the structure of particular kinds of files; for example,
the form of the output of the loader and assembler is given. Excluded are files used by only one command,
for example the assembler’s intermediate files.
User-maintained programs and subroutines (sections VI and VII) are not considered part of the UNIX sys-
tem, and the principal reason for listing them is to indicate their existence without necessarily giving a com-
plete description. The authors of the individual programs should be consulted for more information.
Section VIII discusses commands which are not intended for use by the ordinary user, in some cases be-
cause they disclose information in which he is presumably not interested, and in others because they per-
form privileged functions.

Each section consists of a number of independent entries of a page or so each. The name of the entry is in
the upper corners of its pages, its preparation date in the upper middle. Entries within each section are al-
phabetized. The page numbers of each entry start at 1. (The earlier hope for frequent, partial updates of the
manual is clearly in vain, but in any event it is not feasible to maintain consecutive page numbering in a
document like this.)

2
-

All entries are based on a common format, not all of whose subsections will always appear.
The name section repeats the entry name and gives a very short description of its purpose.
The synopsis summarizes the use of the program being described. A few conventions are used,
particularly in the Commands section:
Boldface words are considered literals, and are typed just as they appear.
Square brackets ( [ ] ) around an argument indicate that the argument is optional. When
an argument is given as ‘‘name’’, it always refers to a file name.
Ellipses ‘‘. . .’’ are used to show that the previous argument-prototype may be repeated.
A final convention is used by the commands themselves. An argument beginning with a
minus sign ‘‘−’’ is often taken to mean some sort of flag argument even if it appears in a
position where a file name could appear. Therefore, it is unwise to have files whose
names begin with ‘‘−’’.
The description section discusses in detail the subject at hand.
The files section gives the names of files which are built into the program.
A see also section gives pointers to related information.
A diagnostics section discusses the diagnostic indications which may be produced. Messages
which are intended to be self-explanatory are not listed.
The bugs section gives known bugs and sometimes deficiencies. Occasionally also the suggested
fix is described.
At the beginning of this document is a table of contents, organized by section and alphabetically within
each section. There is also a permuted index derived from the table of contents. Within each index entry,
the title of the writeup to which it refers is followed by the appropriate section number in parentheses. This
fact is important because there is considerable name duplication among the sections, arising principally
from commands which exist only to exercise a particular system call.

This manual was prepared using the UNIX text editor ed and the formatting program troff.

3
-

HOW TO GET STARTED

This section provides the basic information you need to get started on UNIX: how to log in and log out, how
to communicate through your terminal, and how to run a program. See ‘‘UNIX for Beginners’’ by Brian W.
Kernighan for a more complete introduction to the system.
Logging in. You must call UNIX from an appropriate terminal. UNIX supports ASCII terminals typified by
the TTY 37, the GE Terminet 300, the Dasi 300, and various graphical terminals. You must also have a
valid user name, which may be obtained, together with the telephone number, from the system administra-
tors. The same telephone number serves terminals operating at all the standard speeds. After a data con-
nection is established, the login procedure depends on what kind of terminal you are using.
300-baud terminals: Such terminals include the GE Terminet 300, most display terminals, Exe-
cuport, TI, GSI, and certain Anderson-Jacobson terminals. These terminals generally have a speed
switch which should be set at ‘‘300’’ (or ‘‘30’’ for 30 characters per second) and a half/full duplex
switch which should be set at full-duplex. (This switch will often have to be changed since many
other systems require half-duplex). When a connection is established, the system types ‘‘login:’’;
you type your user name, followed by the ‘‘return’’ key. If you have a password, the system asks
for it and turns off the printer on the terminal so the password will not appear. After you have
logged in, the ‘‘return’’, ‘‘new line’’, or ‘‘linefeed’’ keys will give exactly the same results.
TTY 37 terminal: When you have established a data connection, the system types out a few
garbage characters (the ‘‘login:’’ message at the wrong speed). Depress the ‘‘break’’ (or ‘‘inter-
rupt’’) key; this is a speed-independent signal to UNIX that a 150-baud terminal is in use. The sys-
tem then will type ‘‘login:,’’ this time at the correct speed; you respond with your user name.
From the TTY 37 terminal, and any other which has the ‘‘new-line’’ function (combined carriage
return and linefeed), terminate each line you type with the ‘‘new-line’’ key (not the ‘‘return’’ key).
For all these terminals, it is important that you type your name in lower-case if possible; if you type upper-
case letters, UNIX will assume that your terminal cannot generate lower-case letters and will translate all
subsequent upper-case letters to lower case.
The evidence that you have successfully logged in is that the Shell program will type a ‘‘%’’ to you. (The
Shell is described below under ‘‘How to run a program.’’)
For more information, consult getty (VIII), which discusses the login sequence in more detail, and tty (IV),
which discusses typewriter I/O.
Logging out. There are three ways to log out:
You can simply hang up the phone.
You can log out by typing an end-of-file indication (EOT character, control ‘‘d’’) to the Shell. The
Shell will terminate and the ‘‘login: ’’ message will appear again.
You can also log in directly as another user by giving a login command (I).
How to communicate through your terminal. When you type to UNIX, a gnome deep in the system is gath-
ering your characters and saving them in a secret place. The characters will not be given to a program until
you type a return (or new-line), as described above in Logging in.
UNIX typewriter I/O is full-duplex. It has full read-ahead, which means that you can type at any time, even
while a program is typing at you. Of course, if you type during output, the output will have the input char-
acters interspersed. However, whatever you type will be saved up and interpreted in correct sequence.
There is a limit to the amount of read-ahead, but it is generous and not likely to be exceeded unless the sys-
tem is in trouble. When the read-ahead limit is exceeded, the system throws away all the saved characters.
On a typewriter input line, the character ‘‘@’’ kills all the characters typed before it, so typing mistakes can
be repaired on a single line. Also, the character ‘‘#’’ erases the last character typed. Successive uses of

4
-

‘‘#’’ erase characters back to, but not beyond, the beginning of the line. ‘‘@’’ and ‘‘#’’ can be transmitted
to a program by preceding them with ‘‘\’’. (So, to erase ‘‘\’’, you need two ‘‘#’’s).
The ASCII ‘‘delete’’ (a.k.a. ‘‘rubout’’) character is not passed to programs but instead generates an interrupt
signal. This signal generally causes whatever program you are running to terminate. It is typically used to
stop a long printout that you don’t want. However, programs can arrange either to ignore this signal alto-
gether, or to be notified when it happens (instead of being terminated). The editor, for example, catches in-
terrupts and stops what it is doing, instead of terminating, so that an interrupt can be used to halt an editor
printout without losing the file being edited.
The quit signal is generated by typing the ASCII FS character. It not only causes a running program to ter-
minate but also generates a file with the core image of the terminated process. Quit is useful for debugging.
Besides adapting to the speed of the terminal, UNIX tries to be intelligent about whether you have a terminal
with the new-line function or whether it must be simulated with carriage-return and line-feed. In the latter
case, all input carriage returns are turned to new-line characters (the standard line delimiter) and both a car-
riage return and a line feed are echoed to the terminal. If you get into the wrong mode, the stty command
(I) will rescue you.
Tab characters are used freely in UNIX source programs. If your terminal does not have the tab function,
you can arrange to have them turned into spaces during output, and echoed as spaces during input. The
system assumes that tabs are set every eight columns. Again, the stty command (I) will set or reset this
mode. Also, there is a file which, if printed on TTY 37 or TermiNet 300 terminals, will set the tab stops cor-
rectly (tabs (V)).
Section tty (IV) discusses typewriter I/O more fully.
How to run a program; the Shell. When you have successfully logged into UNIX, a program called the
Shell is listening to your terminal. The Shell reads typed-in lines, splits them up into a command name and
arguments, and executes the command. A command is simply an executable program. The Shell looks first
in your current directory (see next section) for a program with the given name, and if none is there, then in
a system directory. There is nothing special about system-provided commands except that they are kept in
a directory where the Shell can find them.
The command name is always the first word on an input line; it and its arguments are separated from one
another by spaces.
When a program terminates, the Shell will ordinarily regain control and type a ‘‘%’’ at you to indicate that
it is ready for another command.
The Shell has many other capabilities, which are described in detail in section sh (I).
The current directory. UNIX has a file system arranged in a hierarchy of directories. When the system ad-
ministrator gave you a user name, he also created a directory for you (ordinarily with the same name as
your user name). When you log in, any file name you type is by default in this directory. Since you are the
owner of this directory, you have full permissions to read, write, alter, or destroy its contents. Permissions
to have your will with other directories and files will have been granted or denied to you by their owners.
As a matter of observed fact, few UNIX users protect their files from destruction, let alone perusal, by other
users.
To change the current directory (but not the set of permissions you were endowed with at login) use chdir
(I).
Path names. To refer to files not in the current directory, you must use a path name. Full path names begin
with ‘‘/’’, the name of the root directory of the whole file system. After the slash comes the name of each
directory containing the next sub-directory (followed by a ‘‘/’’) until finally the file name is reached. E.g.:
/ usr/ lem/ filex refers to the file filex in the directory lem; lem is itself a subdirectory of usr; usr springs di-
rectly from the root directory.
If your current directory has subdirectories, the path names of files therein begin with the name of the sub-

5
-

directory (no prefixed ‘‘/’’).

Without important exception, a path name may be used anywhere a file name is required.
Important commands which modify the contents of files are cp (I), mv (I), and rm (I), which respectively
copy, move (i.e. rename) and remove files. To find out the status of files or directories, use ls (I). See mkdir
(I) for making directories; rmdir (I) for destroying them.
For a fuller discussion of the file system, see ‘‘The UNIX Time-Sharing System,’’ by the present authors. It
may also be useful to glance through section II of this manual, which discusses system calls, even if you
don’t intend to deal with the system at that level.
Writing a program. To enter the text of a source program into a UNIX file, use ed (I). The three principal
languages in UNIX are assembly language (see as (I)), Fortran (see fc (I)), and C (see cc (I)). After the pro-
gram text has been entered through the editor and written on a file, you can give the file to the appropriate
language processor as an argument. The output of the language processor will be left on a file in the cur-
rent directory named ‘‘a.out’’. (If the output is precious, use mv to move it to a less exposed name soon.) If
you wrote in assembly language, you will probably need to load the program with library subroutines; see
ld (I). The other two language processors call the loader automatically.
When you have finally gone through this entire process without provoking any diagnostics, the resulting
program can be run by giving its name to the Shell in response to the ‘‘%’’ prompt.
Next, you will need cdb (I) or db (I) to examine the remains of your program. The former is useful for C
programs, the latter for assembly-language. No debugger is much help for Fortran.
Your programs can receive arguments from the command line just as system programs do. See exec (II).
Text processing. Almost all text is entered through the editor. The commands most often used to write text
on a terminal are: cat, pr, roff, nroff, and troff, all in section I.
The cat command simply dumps ASCII text on the terminal, with no processing at all. The pr command
paginates the text, supplies headings, and has a facility for multi-column output. Troff and nroff are elabo-
rate text formatting programs, and require careful forethought in entering both the text and the formatting
commands into the input file. Troff drives a Graphic Systems phototypesetter; it was used to produce this
manual. Nroff produces output on a typewriter terminal. Roff (I) is a somewhat less elaborate text format-
ting program, and requires somewhat less forethought.
Surprises. Certain commands provide inter-user communication. Even if you do not plan to use them, it
would be well to learn something about them, because someone else may aim them at you.
To communicate with another user currently logged in, write (I) is used; mail (I) will leave a message
whose presence will be announced to another user when he next logs in. The write-ups in the manual also
suggest how to respond to the two commands if you are a target.
When you log in, a message-of-the-day may greet you before the first ‘‘%’’.

6
-

AR ( I ) 3/15/72 AR ( I )

NAME
ar − archive and library maintainer
SYNOPSIS
ar key afile name ...
DESCRIPTION
Ar maintains groups of files combined into a single archive file. Its main use is to create and update library
files as used by the loader. It can be used, though, for any similar purpose.
Key is one character from the set drtux, optionally concatenated with v. Afile is the archive file. The names
are constituent files in the archive file. The meanings of the key characters are:
d means delete the named files from the archive file.
r means replace the named files in the archive file. If the archive file does not exist, r creates it. If the
named files are not in the archive file, they are appended.
t prints a table of contents of the archive file. If no names are given, all files in the archive are tabled. If
names are given, only those files are tabled.
u is similar to r except that only those files that have been modified are replaced. If no names are given, all
files in the archive that have been modified are replaced by the modified version.
x extracts the named files. If no names are given, all files in the archive are extracted. In neither case does
x alter the archive file.
v means verbose. Under the verbose option, ar gives a file-by-file description of the making of a new
archive file from the old archive and the constituent files. The following abbreviations are used:
c copy
a append
d delete
r replace
x extract
FILES
/tmp/vtm? temporary
SEE ALSO
ld (I), archive (V)
BUGS
Option tv should be implemented as a table with more information.
There should be a way to specify the placement of a new file in an archive. Currently, it is placed at the
end.
Since ar has not been rewritten to deal properly with the new file system modes, extracted files have mode
666.
For the same reason, only the first 8 characters of file names are significant.
If the same file is mentioned twice in an argument list, it may be put in the archive twice.

-1-
-

AS ( I ) 1/15/73 AS ( I )

NAME
as − assembler
SYNOPSIS
as [ − ] name ...
DESCRIPTION
As assembles the concatenation of the named files. If the optional first argument − is used, all undefined
symbols in the assembly are treated as global.
The output of the assembly is left on the file a.out. It is executable if no errors occurred during the assem-
bly, and if there were no unresolved external references.
FILES
/lib/as2 pass 2 of the assembler
/tmp/atm[1-3]? temporary
a.out object
SEE ALSO
ld (I), nm (I), db (I), a.out (V), ‘UNIX Assembler Manual’.
DIAGNOSTICS
When an input file cannot be read, its name followed by a question mark is typed and assembly ceases.
When syntactic or semantic errors occur, a single-character diagnostic is typed out together with the line
number and the file name in which it occurred. Errors in pass 1 cause cancellation of pass 2. The possible
errors are:
) Parentheses error
] Parentheses error
< String not terminated properly
* Indirection used illegally
. Illegal assignment to ‘.’
A Error in address
B Branch instruction is odd or too remote
E Error in expression
F Error in local (‘f’ or ‘b’) type symbol
G Garbage (unknown) character
I End of file inside an if
M Multiply defined symbol as label
O Word quantity assembled at odd address
P ‘.’ different in pass 1 and 2
R Relocation error
U Undefined symbol
X Syntax error
BUGS
Symbol table overflow is not checked. x errors can cause incorrect line numbers in following diagnostics.

-1-
-

BAS ( I ) 5/15/74 BAS ( I )

NAME
bas − basic
SYNOPSIS
bas [ file ]
DESCRIPTION
Bas is a dialect of Basic. If a file argument is provided, the file is used for input before the console is read.
Bas accepts lines of the form:
statement
integer statement
Integer numbered statements (known as internal statements) are stored for later execution. They are stored
in sorted ascending order. Non-numbered statements are immediately executed. The result of an immedi-
ate expression statement (that does not have ‘=’ as its highest operator) is printed.
Statements have the following syntax:
expression
The expression is executed for its side effects (assignment or function call) or for printing as de-
scribed above.
comment ...
This statement is ignored. It is used to interject commentary in a program.
done
Return to system level.
draw expression expression expression
A line is drawn on the Tektronix 611 display ‘/dev/vt0’ from the current display position to the XY
co-ordinates specified by the first two expressions. The scale is zero to one in both X and Y direc-
tions. If the third expression is zero, the line is invisible. The current display position is set to the
end point.
display list
The list of expressions and strings is concatenated and displayed (i.e. printed) on the 611 starting at
the current display position. The current display position is not changed.
dump
The name and current value of every variable is printed.
edit
The UNIX editor, ed, is invoked with the file argument. After the editor exits, this file is recompiled.
erase
The 611 screen is erased.
for name = expression expression statement
for name = expression expression
...
next
The for statement repetitively executes a statement (first form) or a group of statements (second
form) under control of a named variable. The variable takes on the value of the first expression, then
is incremented by one on each loop, not to exceed the value of the second expression.
goto expression
The expression is evaluated, truncated to an integer and execution goes to the corresponding integer
numbered statment. If executed from immediate mode, the internal statements are compiled first.
if expression statement
if expression
...
[ else
... ]

-1-
-

BAS ( I ) 5/15/74 BAS ( I )

fi
The statement (first form) or group of statements (second form) is executed if the expression evalu-
ates to non-zero. In the second form, an optional else allows for a group of statements to be executed
when the first group is not.
list [expression [expression]]
is used to print out the stored internal statements. If no arguments are given, all internal statements
are printed. If one argument is given, only that internal statement is listed. If two arguments are giv-
en, all internal statements inclusively between the arguments are printed.
print list
The list of expressions and strings are concatenated and printed. (A string is delimited by " charac-
ters.)
prompt list
Prompt is the same as print except that no newline character is printed.
return [expression]
The expression is evaluated and the result is passed back as the value of a function call. If no expres-
sion is given, zero is returned.
run
The internal statements are compiled. The symbol table is re-initialized. The random number gener-
ator is reset. Control is passed to the lowest numbered internal statement.
save [expression [expression]]
Save is like list except that the output is written on the file argument. If no argument is given on the
command, b.out is used.
Expressions have the following syntax:
name
A name is used to specify a variable. Names are composed of a letter followed by letters and digits.
The first four characters of a name are significant.
number
A number is used to represent a constant value. A number is written in Fortran style, and contains
digits, an optional decimal point, and possibly a scale factor consisting of an e followed by a possibly
signed exponent.
( expression )
Parentheses are used to alter normal order of evaluation.
expression
The result is the negation of the expression.
expression operator expression
Common functions of two arguments are abbreviated by the two arguments separated by an operator
denoting the function. A complete list of operators is given below.
expression ( [expression [ , expression] ... ] )
Functions of an arbitrary number of arguments can be called by an expression followed by the argu-
ments in parentheses separated by commas. The expression evaluates to the line number of the entry
of the function in the internally stored statements. This causes the internal statements to be com-
piled. If the expression evaluates negative, a builtin function is called. The list of builtin functions
appears below.
name [ expression [ , expression ] ... ]
Each expression is truncated to an integer and used as a specifier for the name. The result is syntacti-
cally identical to a name. a[1,2] is the same as a[1][2]. The truncated expressions are restricted to
values between 0 and 32767.
The following is the list of operators:

-2-
-

BAS ( I ) 5/15/74 BAS ( I )

=
= is the assignment operator. The left operand must be a name or an array element. The result is the
right operand. Assignment binds right to left, all other operators bind left to right.
& |
& (logical and) has result zero if either of its arguments are zero. It has result one if both its argu-
ments are non-zero. | (logical or) has result zero if both of its arguments are zero. It has result one if
either of its arguments are non-zero.
< <= > >= == <>
The relational operators (< less than, <= less than or equal, > greater than, >= greater than or equal,
== equal to, <> not equal to) return one if their arguments are in the specified relation. They return
zero otherwise. Relational operators at the same level extend as follows: a>b>c is the same as
a>b&b>c.
+−
Add and subtract.
*/
Multiply and divide.
ˆ
Exponentiation.
The following is a list of builtin functions:
arg(i)
is the value of the i -th actual parameter on the current level of function call.
exp(x)
is the exponential function of x.
log(x)
is the natural logarithm of x.
sqr(x)
is the square root of x.
sin(x)
is the sine of x (radians).
cos(x)
is the cosine of x (radians).
atn(x)
is the arctangent of x. Its value is between −π/2 and π/2.
rnd( )
is a uniformly distributed random number between zero and one.
expr( )
is the only form of program input. A line is read from the input and evaluated as an expression. The
resultant value is returned.
abs(x)
is the absolute value of x.
int(x)
returns x truncated (towards 0) to an integer.
FILES
/tmp/btm? temporary
b.out save file
DIAGNOSTICS
Syntax errors cause the incorrect line to be typed with an underscore where the parse failed. All other diag-
nostics are self explanatory.

-3-
-

BAS ( I ) 5/15/74 BAS ( I )

BUGS
Has been known to give core images.

-4-
-

BC ( I ) 2/20/75 BC ( I )

NAME
bc − arbitrary precision interactive language
SYNOPSIS
bc [ −l ] [ file ... ]
DESCRIPTION
Bc is an interactive processor for a language which resembles C but provides unlimited precision arith-
metic. It takes input from any files given, then reads the standard input. The ‘−l’ argument stands for the
name of a library of mathematical subroutines which contains sine (named ‘s’), cosine (‘c’), arctangent
(‘a’), natural logarithm (‘l’), and exponential (‘e’). The syntax for bc programs is as follows; E means ex-
pression, S means statement.
Comments
are enclosed in /* and */.
Names
letters a−z
array elements: letter[E]
The words ‘ibase’, ‘obase’, and ‘scale’
Other operands
arbitrarily long numbers with optional sign and decimal point.
(E)
sqrt ( E )
<letter> ( E , ... , E )
Operators
+ − * / % ˆ
++ −− (prefix and postfix; apply to names)
== <= >= != < >
= =+ =− =* =/ =% =ˆ
Statements
E
{ S ; ... ; S }
if ( E ) S
while ( E ) S
for ( E ; E ; E ) S
null statement
break
quit
Function definitions are exemplified by
define <letter> ( <letter> ,..., <letter> ) {
auto <letter>, ... , <letter>
S; ... S
return ( E )
}
All function arguments are passed by value.
The value of a statement that is an expression is printed unless the main operator is an assignment. Either
semicolons or newlines may separate statements. Assignment to scale influences the number of digits to be
retained on arithmetic operations. Assignments to ibase or obase set the input and output number radix re-
spectively.
The same letter may be used as an array name, a function name, and a simple variable simultaneously. ‘Au-
to’ variables are saved and restored during function calls. All other variables are global to the program.
When using arrays as function arguments or defining them as automatic variables empty square brackets
must follow the array name.

-1-
-

BC ( I ) 2/20/75 BC ( I )

For example
scale = 20
define e(x){
auto a, b, c, i, s
a=1
b=1
s=1
for(i=1; 1==1; i++){
a = a*x
b = b*i
c = a/b
if(c == 0) return(s)
s = s+c
}
}
defines a function to compute an approximate value of the exponential function and
for(i=1; i<=10; i++) e(i)
prints approximate values of the exponential function of the first ten integers.
FILES
/usr/lib/lib.b mathematical library
SEE ALSO
dc (I), C Reference Manual, ‘‘BC − An Arbitrary Precision Desk-Calculator Language.’’
BUGS
No &&, | | yet.
for statement must have all three E’s
quit is interpreted when read, not when executed.

-2-
-

CAT ( I ) 1/15/73 CAT ( I )

NAME
cat − concatenate and print
SYNOPSIS
cat file ...
DESCRIPTION
Cat reads each file in sequence and writes it on the standard output. Thus
cat file
prints the file, and
cat file1 file2 >file3
concatenates the first two files and places the result on the third.
If no input file is given, or if the argument ‘−’ is encountered, cat reads from the standard input file.

SEE ALSO
pr(I), cp(I)
DIAGNOSTICS
none; if a file cannot be found it is ignored.
BUGS
cat x y >x and cat x y >y cause strange results.

-1-
-

CC ( I ) 5/15/74 CC ( I )

NAME
cc − C compiler
SYNOPSIS
cc [ −c ] [ −p ] [ −f ] [ −O ] [ −S ] [ −P ] file ...
DESCRIPTION
Cc is the UNIX C compiler. It accepts three types of arguments:
Arguments whose names end with ‘.c’ are taken to be C source programs; they are compiled, and each ob-
ject program is left on the file whose name is that of the source with ‘.o’ substituted for ‘.c’. The ‘.o’ file is
normally deleted, however, if a single C program is compiled and loaded all at one go.
The following flags are interpreted by cc. See ld (I) for load-time flags.
−c Suppress the loading phase of the compilation, and force an object file to be produced even if only
one program is compiled.
−p Arrange for the compiler to produce code which counts the number of times each routine is called;
also, if loading takes place, replace the standard startup routine by one which automatically calls the
monitor subroutine (III) at the start and arranges to write out a mon.out file at normal termination of
execution of the object program. An execution profile can then be generated by use of prof (I).
−f In systems without hardware floating-point, use a version of the C compiler which handles floating-
point constants and loads the object program with the floating-point interpreter. Do not use if the
hardware is present.
−O Invoke an object-code optimizer.
−S Compile the named C programs, and leave the assembler-language output on corresponding files
suffixed ‘.s’.
−P Run only the macro preprocessor on the named C programs, and leave the output on corresponding
files suffixed ‘.i’.
Other arguments are taken to be either loader flag arguments, or C-compatible object programs, typically
produced by an earlier cc run, or perhaps libraries of C-compatible routines. These programs, together with
the results of any compilations specified, are loaded (in the order given) to produce an executable program
with name a.out.
FILES
file.c input file
file.o object file
a.out loaded output
/tmp/ctm? temporary
/lib/c[01] compiler
/lib/fc[01] floating-point compiler
/lib/c2 optional optimizer
/lib/crt0.o runtime startoff
/lib/mcrt0.o runtime startoff of profiling
/lib/fcrt0.o runtime startoff for floating-point interpretation
/lib/libc.a C library; see section III.
/lib/liba.a Assembler library used by some routines in libc.a
SEE ALSO
‘‘Programming in C— a tutorial,’’ C Reference Manual, monitor (III), prof (I), cdb (I), ld (I).
DIAGNOSTICS
The diagnostics produced by C itself are intended to be self-explanatory. Occasional messages may be pro-
duced by the assembler or loader. Of these, the most mystifying are from the assembler, in particular ‘‘m,’’
which means a multiply-defined external symbol (function or data).

-1-
-

CC ( I ) 5/15/74 CC ( I )

BUGS

-2-
-

CHDIR ( I ) 5/15/74 CHDIR ( I )

NAME
chdir − change working directory
SYNOPSIS
chdir directory
DESCRIPTION
Directory becomes the new working directory. The process must have execute (search) permission in direc-
tory.
Because a new process is created to execute each command, chdir would be ineffective if it were written as
a normal command. It is therefore recognized and executed by the Shell.
SEE ALSO
sh (I), pwd (I)
BUGS

-1-
-

CHMOD ( I ) 2/8/75 CHMOD ( I )

NAME
chmod − change mode
SYNOPSIS
chmod octal file ...
DESCRIPTION
The octal mode replaces the mode of each of the files. The mode is constructed from the OR of the follow-
ing modes:
4000 set user ID on execution
2000 set group ID on execution
1000 sticky bit for shared, pure-procedure programs (see below)
0400 read by owner
0200 write by owner
0100 execute (search in directory) by owner
0070 read, write, execute (search) by group
0007 read, write, execute (search) by others
Only the owner of a file (or the super-user) may change its mode.
If an executable file is set up for sharing (‘‘−n’’ option of ld (I) ), then mode 1000 prevents the system from
abandoning the swap-space image of the program-text portion of the file when its last user terminates.
Thus when the next user of the file executes it, the text need not be read from the file system but can simply
be swapped in, saving time. Ability to set this bit is restricted to the super-user since swap space is con-
sumed by the images; it is only worth while for heavily used commands.

SEE ALSO
ls (I), chmod (II)
BUGS

-1-
-

CMP ( I ) 5/15/74 CMP ( I )

NAME
cmp − compare two files
SYNOPSIS
cmp [ −l ] [ −s ] file1 file2
DESCRIPTION
The two files are compared. (If file1 is ‘−’, the standard input is used.) Under default options, cmp makes
no comment if the files are the same; if they differ, it announces the byte and line number at which the dif-
ference occurred. If one file is an initial subsequence of the other, that fact is noted. Moreover, return code
0 is yielded for identical files, 1 for different files, and 2 for an inaccessible or missing argument.
Options:
−l Print the byte number (decimal) and the differing bytes (octal) for each difference.
−s Print nothing for differing files; return codes only.
SEE ALSO
diff (I), comm (I)
BUGS

-1-
-

COMM ( I ) 5/15/74 COMM ( I )

NAME
comm − print lines common to two files
SYNOPSIS
comm [ − [ 123 ] ] file1 file2
DESCRIPTION
Comm reads file1 and file2, which should be in sort, and produces a three column output: lines only in file1;
lines only in file2; and lines in both files. The filename ‘−’ means the standard input.
Flags 1, 2, or 3 suppress printing of the corresponding column. Thus comm −12 prints only the lines com-
mon to the two files; comm −23 prints only lines in the first file but not in the second; comm −123 is a no-
op.

SEE ALSO
cmp (I), diff (I)
BUGS

-1-
-

CP ( I ) 2/8/75 CP ( I )

NAME
cp − copy
SYNOPSIS
cp file1 file2
DESCRIPTION
The first file is copied onto the second. The mode and owner of the target file are preserved if it already ex-
isted; the mode of the source file is used otherwise.
If file2 is a directory, then the target file is a file in that directory with the file-name of file1.
It is forbidden to copy a file onto itself.
SEE ALSO
cat (I), pr (I), mv (I)
BUGS

-1-
-

CREF ( I ) 2/5/73 CREF ( I )

NAME
cref − make cross reference listing
SYNOPSIS
cref [ −acilostux123 ] name ...
DESCRIPTION
Cref makes a cross reference listing of program files in assembler or C format. The files named as argu-
ments in the command line are searched for symbols in the appropriate syntax.
The output report is in four columns:
(1) (2) (3) (4)
symbol file see text as it appears in file
below
Cref uses either an ignore file or an only file. If the −i option is given, the next argument is taken to be an
ignore file; if the −o option is given, the next argument is taken to be an only file. Ignore and only files are
lists of symbols separated by new lines. All symbols in an ignore file are ignored in columns (1) and (3) of
the output. If an only file is given, only symbols in that file appear in column (1). At most one of −i and −o
may be used. The default setting is −i. Assembler predefined symbols or C keywords are ignored.
The −s option causes current symbols to be put in column 3. In the assembler, the current symbol is the
most recent name symbol; in C, the current function name. The −l option causes the line number within the
file to be put in column 3.
The −t option causes the next available argument to be used as the name of the intermediate temporary file
(instead of /tmp/crt??). The file is created and is not removed at the end of the process.
Options:
a assembler format (default)
c C format input
i use ignore file (see above)
l put line number in col. 3 (instead of current symbol)
o use only file (see above)
s current symbol in col. 3 (default)
t user supplied temporary file
u print only symbols that occur exactly once
x print only C external symbols
1 sort output on column 1 (default)
2 sort output on column 2
3 sort output on column 3

FILES
/tmp/crt?? temporaries
/usr/lib/aign default assembler ignore file
/usr/lib/cign default C ignore file
/usr/bin/crpost post processor
/usr/bin/upost post processor for −u option
/bin/sort used to sort temporaries

SEE ALSO
as (I), cc (I)
BUGS

-1-
-

DATE ( I ) 11/1/74 DATE ( I )

NAME
date − print and set the date
SYNOPSIS
date [ s ] [ mmddhhmm[yy] ]
DESCRIPTION
If no argument is given, the current date and time are printed. If an argument is given, the current date is
set. The first mm is the month number; dd is the day number in the month; hh is the hour number (24 hour
system); the second mm is the minute number; yy is the last 2 digits of the year number and is optional. For
example:
date 10080045
sets the date to Oct 8, 12:45 AM. The current year is the default if no year is mentioned. The system oper-
ates in GMT. Date takes care of the conversion to and from local standard and daylight time.
If the argument is ‘‘s,’’ date calls the network file store via the TIU interface (if present) and sets the clock
to the time thereby obtained.
DIAGNOSTICS
‘‘No permission’’ if you aren’t the super-user and you try to change the date; ‘‘bad conversion’’ if the date
set is syntactically incorrect.
FILES
/dev/tiu/d0
BUGS

-1-
-

DB ( I ) 8/20/73 DB ( I )

NAME
db − debug
SYNOPSIS
db [ core [ namelist ] ] [ − ]
DESCRIPTION
Unlike many debugging packages (including DEC’s ODT, on which db is loosely based), db is not loaded
as part of the core image which it is used to examine; instead it examines files. Typically, the file will be ei-
ther a core image produced after a fault or the binary output of the assembler. Core is the file being de-
bugged; if omitted core is assumed. Namelist is a file containing a symbol table. If it is omitted, the sym-
bol table is obtained from the file being debugged, or if not there from a.out. If no appropriate name list
file can be found, db can still be used but some of its symbolic facilities become unavailable.
For the meaning of the optional third argument, see the last paragraph below.
The format for most db requests is an address followed by a one character command. Addresses are ex-
pressions built up as follows:
1. A name has the value assigned to it when the input file was assembled. It may be relocatable or not
depending on the use of the name during the assembly.
2. An octal number is an absolute quantity with the appropriate value.
3. A decimal number immediately followed by ‘.’ is an absolute quantity with the appropriate value.
4. An octal number immediately followed by r is a relocatable quantity with the appropriate value.
5. The symbol . indicates the current pointer of db. The current pointer is set by many db requests.
6. A * before an expression forms an expression whose value is the number in the word addressed by the
first expression. A * alone is equivalent to ‘*.’.
7. Expressions separated by + or blank are expressions with value equal to the sum of the components.
At most one of the components may be relocatable.
8. Expressions separated by − form an expression with value equal to the difference to the components.
If the right component is relocatable, the left component must be relocatable.
9. Expressions are evaluated left to right.
Names for registers are built in:
r0 ... r5
sp
pc
fr0 ... fr5
These may be examined. Their values are deduced from the contents of the stack in a core image file.
They are meaningless in a file that is not a core image.
If no address is given for a command, the current address (also specified by ‘‘.’’) is assumed. In general,
‘‘.’’ points to the last word or byte printed by db.
There are db commands for examining locations interpreted as numbers, machine instructions, ASCII char-
acters, and addresses. For numbers and characters, either bytes or words may be examined. The following
commands are used to examine the specified file.
/ The addressed word is printed in octal.
\ The addressed byte is printed in octal.
" The addressed word is printed as two ASCII characters.
´ The addressed byte is printed as an ASCII character.
` The addressed word is printed in decimal.

-1-
-

DB ( I ) 8/20/73 DB ( I )

? The addressed word is interpreted as a machine instruction and a symbolic form of the instruction, in-
cluding symbolic addresses, is printed. Often, the result will appear exactly as it was written in the
source program.
& The addressed word is interpreted as a symbolic address and is printed as the name of the symbol
whose value is closest to the addressed word, possibly followed by a signed offset.
<nl>(i. e., the character ‘‘new line’’) This command advances the current location counter ‘‘.’’ and prints
the resulting location in the mode last specified by one of the above requests.
ˆ This character decrements ‘‘.’’ and prints the resulting location in the mode last selected one of the
above requests. It is a converse to <nl>.
% Exit.
Odd addresses to word-oriented commands are rounded down. The incrementing and decrementing of ‘‘.’’
done by the <nl> and ˆ requests is by one or two depending on whether the last command was word or byte
oriented.
The address portion of any of the above commands may be followed by a comma and then by an expres-
sion. In this case that number of sequential words or bytes specified by the expression is printed. ‘‘.’’ is ad-
vanced so that it points at the last thing printed.
There are two commands to interpret the value of expressions.
= When preceded by an expression, the value of the expression is typed in octal. When not preceded by
an expression, the value of ‘‘.’’ is indicated. This command does not change the value of ‘‘.’’.
: An attempt is made to print the given expression as a symbolic address. If the expression is relocat-
able, that symbol is found whose value is nearest that of the expression, and the symbol is typed, fol-
lowed by a sign and the appropriate offset. If the value of the expression is absolute, a symbol with
exactly the indicated value is sought and printed if found; if no matching symbol is discovered, the oc-
tal value of the expression is given.
The following command may be used to patch the file being debugged.
! This command must be preceded by an expression. The value of the expression is stored at the loca-
tion addressed by the current value of ‘‘.’’. The opcodes do not appear in the symbol table, so the user
must assemble them by hand.
The following command is used after a fault has caused a core image file to be produced.
$ causes the fault type and the contents of the general registers and several other registers to be printed
both in octal and symbolic format. The values are as they were at the time of the fault.
For some purposes, it is important to know how addresses typed by the user correspond with locations in
the file being debugged. The mapping algorithm employed by db is non-trivial for two reasons: First, in an
a.out file, there is a 20(8) byte header which will not appear when the file is loaded into core for execution.
Therefore, apparent location 0 should correspond with actual file offset 20. Second, addresses in core im-
ages do not correspond with the addresses used by the program because in a core image there is a header
containing the system’s per-process data for the dumped process, and also because the stack is stored con-
tiguously with the text and data part of the core image rather than at the highest possible locations. Db
obeys the following rules:
If exactly one argument is given, and if it appears to be an a.out file, the 20-byte header is skipped during
addressing, i.e., 20 is added to all addresses typed. As a consequence, the header can be examined begin-
ning at location −20.
If exactly one argument is given and if the file does not appear to be an a.out file, no mapping is done.
If zero or two arguments are given, the mapping appropriate to a core image file is employed. This means
that locations above the program break and below the stack effectively do not exist (and are not, in fact,
recorded in the core file). Locations above the user’s stack pointer are mapped, in looking at the core file,
to the place where they are really stored. The per-process data kept by the system, which is stored in the
first part of the core file, cannot currently be examined (except by $).

-2-
-

DB ( I ) 8/20/73 DB ( I )

If one wants to examine a file which has an associated name list, but is not a core image file, the last argu-
ment ‘‘−’’ can be used (actually the only purpose of the last argument is to make the number of arguments
not equal to two). This feature is used most frequently in examining the memory file /dev/mem.
SEE ALSO
as (I), core (V), a.out (V), od (I)
DIAGNOSTICS
‘‘File not found’’ if the first argument cannot be read; otherwise ‘‘?’’.
BUGS
There should be some way to examine the registers and other per-process data in a core image; also there
should be some way of specifying double-precision addresses. It does not know yet about shared text seg-
ments.

-3-
-

DC ( I ) 2/8/75 DC ( I )

NAME
dc − desk calculator
SYNOPSIS
dc [ file ]
DESCRIPTION
Dc is an arbitrary precision arithmetic package. Ordinarily it operates on decimal integers, but one may
specify an input base, output base, and a number of fractional digits to be maintained. The overall structure
of dc is a stacking (reverse Polish) calculator. If an argument is given, input is taken from that file until its
end, then from the standard input. The following constructions are recognized:
number
The value of the number is pushed on the stack. A number is an unbroken string of the digits 0-9.
It may be preceded by an underscore to input a negative number. Numbers may contain decimal
points.
+ − * % ˆ
The top two values on the stack are added (+), subtracted (−), multiplied (*), divided (/), remain-
dered (%), or exponentiated (ˆ). The two entries are popped off the stack; the result is pushed on the
stack in their place. Any fractional part of an exponent is ignored.
sx The top of the stack is popped and stored into a register named x, where x may be any character. If
the s is capitalized, x is treated as a stack and the value is pushed on it.
lx The value in register x is pushed on the stack. The register x is not altered. All registers start with
zero value. If the l is capitalized, register x is treated as a stack and its top value is popped onto the
main stack.
d The top value on the stack is duplicated.
p The top value on the stack is printed. The top value remains unchanged.
f All values on the stack and in registers are printed.
q exits the program. If executing a string, the recursion level is popped by two. If q is capitalized, the
top value on the stack is popped and the string execution level is popped by that value.
x treats the top element of the stack as a character string and executes it as a string of dc commands.
[ ... ] puts the bracketed ascii string onto the top of the stack.
<x >x =x
The top two elements of the stack are popped and compared. Register x is executed if they obey the
stated relation.
v replaces the top element on the stack by its square root. Any existing fractional part of the argument
is taken into account, but otherwise the scale factor is ignored.
! interprets the rest of the line as a UNIX command.
c All values on the stack are popped.
i The top value on the stack is popped and used as the number radix for further input.
o The top value on the stack is popped and used as the number radix for further output.
k the top of the stack is popped, and that value is used as a non-negative scale factor: the appropriate
number of places are printed on output, and maintained during multiplication, division, and expo-
nentiation. The interaction of scale factor, input base, and output base will be reasonable if all are
changed together.
z The stack level is pushed onto the stack.
? A line of input is taken from the input source (usually the console) and executed.

-1-
-

DC ( I ) 2/8/75 DC ( I )

An example which prints the first ten values of n! is

[la1+dsa*pla10>y]sy
0sa1
lyx
SEE ALSO
bc (I), which is a preprocessor for dc providing infix notation and a C-like syntax which implements func-
tions and reasonable control structures for programs.
DIAGNOSTICS
(x) ? for unrecognized character x.
(x) ? for not enough elements on the stack to do what was asked by command x.
‘Out of space’ when the free list is exhausted (too many digits).
‘Out of headers’ for too many numbers being kept around.
‘Out of pushdown’ for too many items on the stack.
‘Nesting Depth’ for too many levels of nested execution.
BUGS

-2-
-

DD ( I ) 5/15/74 DD ( I )

NAME
dd − convert and copy a file
SYNOPSIS
dd [option=value] ...
DESCRIPTION
Dd copies the specified input file to the specified output with possible conversions. The standard input and
output are used by default. The input and output block size may be specified to take advantage of raw
physical I/O.
option values
if= input file name; standard input is default
of= output file name; standard output is default
ibs= input block size (default 512)
obs= output block size (default 512)
bs= set both input and output block size, superseding ibs and obs; also, if no conversion is
specified, it is particularly efficient since no copy need be done
cbs=n conversion buffer size
skip=n skip n input records before starting copy
count=n copy only n input records
conv=ascii convert EBCDIC to ASCII
ebcdic convert ASCII to EBCDIC
lcase map alphabetics to lower case
ucase map alphabetics to upper case
swab swap every pair of bytes
noerror do not stop processing on an error
sync pad every input record to ibs
... , ... several comma-separated conversions
Where sizes are specified, a number of bytes is expected. A number may end with k, b or w to specify
multiplication by 1024, 512, or 2 respectively. Also a pair of numbers may be separated by x to indicate a
product.
Cbs is used only if ascii or ebcdic conversion is specified. In the former case cbs characters are placed into
the conversion buffer, converted to ASCII, and trailing blanks trimmed and new-line added before sending
the line to the output. In the latter case ASCII characters are read into the conversion buffer, converted to
EBCDIC, and blanks added to make up an output record of size cbs.
After completion, dd reports the number of whole and partial input and output blocks.
For example, to read an EBCDIC tape blocked ten 80-byte EBCDIC card images per record into the ASCII
file x:
dd if=/dev/rmt0 of=x ibs=800 cbs=80 conv=ascii,lcase
Note the use of raw magtape. Dd is especially suited to I/O on the raw physical devices because it allows
reading and writing in arbitrary record sizes.
SEE ALSO
cp (I)
BUGS
The ASCII/EBCDIC conversion tables are taken from the 256 character standard in the CACM Nov, 1968.
It is not clear how this relates to real life.
Newlines are inserted only on conversion to ASCII; padding is done only on conversion to EBCDIC. There
should be separate options.

-1-
-

DIFF ( I ) 5/15/74 DIFF ( I )

NAME
diff − differential file comparator
SYNOPSIS
diff [ − ] name1 name2
DESCRIPTION
Diff tells what lines must be changed in two files to bring them into agreement. The normal output contains
lines of these forms:
n1 a n3,n4
n1,n2 d n3
n1,n2 c n3,n4
These lines resemble ed commands to convert file name1 into file name2. The numbers after the letters per-
tain to file name2. In fact, by exchanging ‘a’ for ‘d’ and reading backward one may ascertain equally how
to convert file name2 into name1. As in ed, identical pairs where n1 = n2 or n3 = n4 are abbreviated as a
single number.
Following each of these lines come all the lines that are affected in the first file flagged by ‘*’, then all the
lines that are affected in the second file flagged by ‘.’.
Under the − option, the output of diff is a script of a, c and d commands for the editor ed, which will change
the contents of the first file into the contents of the second. In this connection, the following shell program
may help maintain multiple versions of a file. Only an ancestral file ($1) and a chain of version-to-version
ed scripts ($2,$3,...) made by diff need be on hand. A ‘latest version’ appears on the standard output.
(cat $2 ... $9; echo "1,$p") | ed − $1
Except for occasional ‘jackpots’, diff finds a smallest sufficient set of file differences.
SEE ALSO
cmp (I), comm (I), ed (I)
DIAGNOSTICS
‘jackpot’ − To speed things up, the program uses hashing. You have stumbled on a case where there is a
chance that this has resulted in a difference being called where none actually existed. Sometimes reversing
the order of files will make a jackpot go away.
BUGS
Editing scripts produced under the − option are naive about creating lines consisting of a single ‘.’.

-1-
-

DSW ( I ) 3/15/72 DSW ( I )

NAME
dsw − delete interactively
SYNOPSIS
dsw [ directory ]
DESCRIPTION
For each file in the given directory (‘.’ if not specified) dsw types its name. If y is typed, the file is deleted;
if x, dsw exits; if new-line, the file is not deleted; if anything else, dsw asks again.
SEE ALSO
rm (I)
BUGS
The name dsw is a carryover from the ancient past. Its etymology is amusing.

-1-
-

DU ( I ) 1/20/73 DU ( I )

NAME
du − summarize disk usage
SYNOPSIS
du [ −s ] [ −a ] [ name ... ]
DESCRIPTION
Du gives the number of blocks contained in all files and (recursively) directories within each specified di-
rectory or file name. If name is missing, ‘.’ is used.
The optional argument −s causes only the grand total to be given. The optional argument −a causes an en-
try to be generated for each file. Absence of either causes an entry to be generated for each directory only.
A file which has two links to it is only counted once.
BUGS
Non-directories given as arguments (not under −a option) are not listed.
Removable file systems do not work correctly since i-numbers may be repeated while the corresponding
files are distinct. Du should maintain an i-number list per root directory encountered.

-1-
-

ECHO ( I ) 3/15/72 ECHO ( I )

NAME
echo − echo arguments
SYNOPSIS
echo [ arg ... ]
DESCRIPTION
Echo writes its arguments in order as a line on the standard output file. It is mainly useful for producing di-
agnostics in command files.
BUGS

-1-
-

ED ( I ) 1/15/73 ED ( I )

NAME
ed − text editor
SYNOPSIS
ed [ − ] [ name ]
DESCRIPTION
Ed is the standard text editor.
If a name argument is given, ed simulates an e command (see below) on the named file; that is to say, the
file is read into ed’s buffer so that it can be edited. The optional − suppresses the printing of character
counts by e, r, and w commands.
Ed operates on a copy of any file it is editing; changes made in the copy have no effect on the file until a w
(write) command is given. The copy of the text being edited resides in a temporary file called the buffer.
There is only one buffer.
Commands to ed have a simple and regular structure: zero or more addresses followed by a single character
command, possibly followed by parameters to the command. These addresses specify one or more lines in
the buffer. Every command which requires addresses has default addresses, so that the addresses can often
be omitted.
In general, only one command may appear on a line. Certain commands allow the input of text. This text
is placed in the appropriate place in the buffer. While ed is accepting text, it is said to be in input mode. In
this mode, no commands are recognized; all input is merely collected. Input mode is left by typing a period
‘.’ alone at the beginning of a line.
Ed supports a limited form of regular expression notation. A regular expression specifies a set of strings of
characters. A member of this set of strings is said to be matched by the regular expression. The regular ex-
pressions allowed by ed are constructed as follows:
1. An ordinary character (not one of those discussed below) is a regular expression and matches that char-
acter.
2. A circumflex ‘ˆ’ at the beginning of a regular expression matches the empty string at the beginning of a
line.
3. A currency symbol ‘$’ at the end of a regular expression matches the null character at the end of a line.
4. A period ‘.’ matches any character except a new-line character.
5. A regular expression followed by an asterisk ‘*’ matches any number of adjacent occurrences (includ-
ing zero) of the regular expression it follows.
6. A string of characters enclosed in square brackets ‘[ ]’ matches any character in the string but no others.
If, however, the first character of the string is a circumflex ‘ˆ’ the regular expression matches any char-
acter except new-line and the characters in the string.
7. The concatenation of regular expressions is a regular expression which matches the concatenation of the
strings matched by the components of the regular expression.
8. A regular expression enclosed between the sequences ‘\(’ and ‘\)’is identical to the unadorned expres-
sion; the construction has side effects discussed under the s command.
9. The null regular expression standing alone is equivalent to the last regular expression encountered.
Regular expressions are used in addresses to specify lines and in one command (see s below) to specify a
portion of a line which is to be replaced. If it is desired to use one of the regular expression metacharacters
as an ordinary character, that character may be preceded by ‘\’. This also applies to the character bounding
the regular expression (often ‘/’) and to ‘\’ itself.
To understand addressing in ed it is necessary to know that at any time there is a current line. Generally
speaking, the current line is the last line affected by a command; however, the exact effect on the current
line is discussed under the description of the command. Addresses are constructed as follows.

-1-
-

ED ( I ) 1/15/73 ED ( I )

1. The character ‘.’ addresses the current line.

2. The character ‘$’ addresses the last line of the buffer.
3. A decimal number n addresses the n-th line of the buffer.
4. ‘´x’ addresses the line marked with the mark name character x, which must be a lower-case letter.
Lines are marked with the k command described below.
5. A regular expression enclosed in slashes ‘/’ addresses the first line found by searching toward the
end of the buffer and stopping at the first line containing a string matching the regular expression. If
necessary the search wraps around to the beginning of the buffer.
6. A regular expression enclosed in queries ‘?’ addresses the first line found by searching toward the
beginning of the buffer and stopping at the first line containing a string matching the regular expres-
sion. If necessary the search wraps around to the end of the buffer.
7. An address followed by a plus sign ‘+’ or a minus sign ‘−’ followed by a decimal number specifies
that address plus (resp. minus) the indicated number of lines. The plus sign may be omitted.
8. If an address begins with ‘+’ or ‘−’ the addition or subtraction is taken with respect to the current
line; e.g. ‘−5’ is understood to mean ‘.−5’.
9. If an address ends with ‘+’ or ‘−’, then 1 is added (resp. subtracted). As a consequence of this rule
and rule 8, the address ‘−’ refers to the line before the current line. Moreover, trailing ‘+’ and ‘−’
characters have cumulative effect, so ‘−−’ refers to the current line less 2.
10. To maintain compatibility with earlier version of the editor, the character ‘ˆ’ in addresses is entirely
equivalent to ‘−’.
Commands may require zero, one, or two addresses. Commands which require no addresses regard the
presence of an address as an error. Commands which accept one or two addresses assume default addresses
when insufficient are given. If more addresses are given than such a command requires, the last one or two
(depending on what is accepted) are used.
Addresses are separated from each other typically by a comma ‘,’. They may also be separated by a semi-
colon ‘;’. In this case the current line ‘.’ is set to the previous address before the next address is interpreted.
This feature can be used to determine the starting line for forward and backward searches (‘/’, ‘?’) . The
second address of any two-address sequence must correspond to a line following the line corresponding to
the first address.
In the following list of ed commands, the default addresses are shown in parentheses. The parentheses are
not part of the address, but are used to show that the given addresses are the default.
As mentioned, it is generally illegal for more than one command to appear on a line. However, any com-
mand may be suffixed by ‘p’ or by ‘l’, in which case the current line is either printed or listed respectively
in the way discussed below.
( . )a
<text>
.
The append command reads the given text and appends it after the addressed line. ‘.’ is left on the
last line input, if there were any, otherwise at the addressed line. Address ‘0’ is legal for this com-
mand; text is placed at the beginning of the buffer.
( . , . )c
<text>
.
The change command deletes the addressed lines, then accepts input text which replaces these lines.
‘.’ is left at the last line input; if there were none, it is left at the first line not deleted.
(.,.) d
The delete command deletes the addressed lines from the buffer. The line originally after the last line
deleted becomes the current line; if the lines deleted were originally at the end, the new last line be-
comes the current line.

-2-
-

ED ( I ) 1/15/73 ED ( I )

e filename
The edit command causes the entire contents of the buffer to be deleted, and then the named file to be
read in. ‘.’ is set to the last line of the buffer. The number of characters read is typed. ‘filename’ is
remembered for possible use as a default file name in a subsequent r or w command.
f filename
The filename command prints the currently remembered file name. If ‘filename’ is given, the cur-
rently remembered file name is changed to ‘filename’.
(1,$) g/regular expression/command list
In the global command, the first step is to mark every line which matches the given regular expres-
sion. Then for every such line, the given command list is executed with ‘.’ initially set to that line. A
single command or the first of multiple commands appears on the same line with the global com-
mand. All lines of a multi-line list except the last line must be ended with ‘\’. A, i, and c commands
and associated input are permitted; the ‘.’ terminating input mode may be omitted if it would be on
the last line of the command list. The (global) commands, g, and v, are not permitted in the com-
mand list.
( . )i
<text>
.
This command inserts the given text before the addressed line. ‘.’ is left at the last line input; if there
were none, at the addressed line. This command differs from the a command only in the placement
of the text.
( . ) kx
The mark command marks the addressed line with name x, which must be a lower-case letter. The
address form ‘´x’ then addresses this line.
( . , . )l
The list command prints the addressed lines in an unambiguous way: non-graphic characters are
printed in octal, and long lines are folded. An l command may follow any other on the same line.
( . , . ) ma
The move command repositions the addressed lines after the line addressed by a. The last of the
moved lines becomes the current line.
( . , . )p
The print command prints the addressed lines. ‘.’ is left at the last line printed. The p command
may be placed on the same line after any command.
q
The quit command causes ed to exit. No automatic write of a file is done.
($) r filename
The read command reads in the given file after the addressed line. If no file name is given, the re-
membered file name, if any, is used (see e and f commands) . The remembered file name is not
changed unless ‘filename’ is the very first file name mentioned. Address ‘0’ is legal for r and causes
the file to be read at the beginning of the buffer. If the read is successful, the number of characters
read is typed. ‘.’ is left at the last line read in from the file.
( . , . ) s/regular expression/replacement/ or,
( . , . ) s/regular expression/replacement/g
The substitute command searches each addressed line for an occurrence of the specified regular ex-
pression. On each line in which a match is found, all matched strings are replaced by the replace-
ment specified, if the global replacement indicator ‘g’ appears after the command. If the global indi-
cator does not appear, only the first occurrence of the matched string is replaced. It is an error for the
substitution to fail on all addressed lines. Any character other than space or new-line may be used in-
stead of ‘/’ to delimit the regular expression and the replacement. ‘.’ is left at the last line substituted.
An ampersand ‘&’ appearing in the replacement is replaced by the string matching the regular ex-
pression. The special meaning of ‘&’ in this context may be suppressed by preceding it by ‘\’. As a
more general feature, the characters ‘\n’, where n is a digit, are replaced by the text matched by the n-

-3-
-

ED ( I ) 1/15/73 ED ( I )

th regular subexpression enclosed between ‘\(’ and ‘\)’. When nested, parenthesized subexpressions
are present, n is determined by counting occurrences of ‘\(’ starting from the left.
Lines may be split by substituting new-line characters into them. The new-line in the replacement
string must be escaped by preceding it by ‘\’.
(.,.)ta
This command acts just like the m command, except that a copy of the addressed lines is placed after
address a (which may be 0). ‘.’ is left on the last line of the copy.
(1,$) v/regular expression/command list
This command is the same as the global command except that the command list is executed with ‘.’
initially set to every line except those matching the regular expression.
(1,$) w filename
The write command writes the addressed lines onto the given file. If the file does not exist, it is creat-
ed mode 666 (readable and writeable by everyone) . The remembered file name is not changed unless
‘filename’ is the very first file name mentioned. If no file name is given, the remembered file name,
if any, is used (see e and f commands) . ‘.’ is unchanged. If the command is successful, the number
of characters written is typed.
($) =
The line number of the addressed line is typed. ‘.’ is unchanged by this command.
!UNIX command
The remainder of the line after the ‘!’ is sent to UNIX to be interpreted as a command. ‘.’ is un-
changed.
( .+1 ) <newline>
An address alone on a line causes the addressed line to be printed. A blank line alone is equivalent to
‘.+1p’; it is useful for stepping through text.
If an interrupt signal (ASCII DEL) is sent, ed prints a ‘?’ and returns to its command level.
Some size limitations: 512 characters per line, 256 characters per global command list, 64 characters per
file name, and 128K characters in the temporary file. The limit on the number of lines depends on the
amount of core: each line takes 1 word.
FILES
/tmp/#, temporary; ‘#’ is the process number (in octal).
DIAGNOSTICS
‘?’ for errors in commands; ‘TMP’ for temporary file overflow.
SEE ALSO
A Tutorial Introduction to the ED Text Editor (B. W. Kernighan)
BUGS
The s command causes all marks to be lost on lines changed.

-4-
-

EQN ( I ) 2/22/74 EQN ( I )

NAME
eqn − typeset mathematics
SYNOPSIS
eqn [ file ] ...
DESCRIPTION
Eqn is a troff (I) preprocessor for typesetting mathematics on the Graphics Systems phototypesetter. Usage
is almost always
eqn file ... | troff
If no files are specified, eqn reads from the standard input. A line beginning with ‘‘.EQ’’ marks the start of
an equation; the end of an equation is marked by a line beginning with ‘‘.EN’’. Neither of these lines is al-
tered or defined by eqn, so you can define them yourself to get centering, numbering, etc. All other lines
are treated as comments, and passed through untouched.
Spaces, tabs, newlines, braces, double quotes, tilde and circumflex are the only delimiters. Braces ‘‘{}’’ are
used for grouping. Use tildes ‘‘˜’’ to get extra spaces in an equation.
Subscripts and superscripts are produced with the keywords sub and sup. Thus x sub i makes x i , a sub i
a
sup 2 produces a2i , and e sup {x sup 2 + y sup 2} gives e x +y . Fractions are made with over. a over b is
2 2

b
1
and 1 over sqrt {ax sup 2 +bx+c} is . sqrt makes square roots.

√ ax 2 + bx + c
n
The keywords from and to introduce lower and upper limits on arbitrary things: lim Σ xi is made with lim
n−>∞ 0
from {n-> inf} sum from 0 to n x sub i. Left and right brackets, braces, etc., of the right height are made
 y2 
with left and right: left [ x sup 2 + y sup 2 over alpha right ] ˜=˜1 produces  x 2 +  = 1. The right
 α 
clause is optional.
a
Vertical piles of things are made with pile, lpile, cpile, and rpile: pile {a above b above c} produces b.
c
There can be an arbitrary number of elements in a pile. lpile left-justifies, pile and cpile center, with differ-
ent vertical spacing, and rpile right justifies.
Diacritical marks are made with dot, dotdot, hat, bar: x dot = f(t) bar is ẋ = f (t). Default sizes and fonts
can be changed with size n and various of roman, italic, and bold.

∫
Keywords like sum (Σ) int ( ) inf (∞) and shorthands like >= (≥) −> (− >), != (≠), are recognized. Spell
out Greek letters in the desired case, as in alpha, GAMMA. Mathematical words like sin, cos, log are made
Roman automatically. Troff (I) four-character escapes like \(bs () can be used anywhere. Strings enclosed
in double quotes "..." are passed through untouched.
SEE ALSO
A System for Typesetting Mathematics (Computer Science Technical Report #17, Bell Laboratories, 1974.)
TROFF Users’ Manual (internal memorandum)
TROFF Made Trivial (internal memorandum)
troff (I), neqn (I)
BUGS
Undoubtedly. Watch out for small or large point sizes − it’s tuned too well for size 10. Be cautious if in-
serting horizontal or vertical motions, and of backslashes in general.

-1-
-

EXIT ( I ) 3/15/72 EXIT ( I )

NAME
exit − terminate command file
SYNOPSIS
exit
DESCRIPTION
Exit performs a seek to the end of its standard input file. Thus, if it is invoked inside a file of commands,
upon return from exit the shell will discover an end-of-file and terminate.
SEE ALSO
if (I), goto (I), sh (I)
BUGS

-1-
-

FC ( I ) 8/20/73 FC ( I )

NAME
fc − Fortran compiler
SYNOPSIS
fc [ −c ] sfile1.f ... ofile1 ...
DESCRIPTION
Fc is the UNIX Fortran compiler. It accepts three types of arguments:
Arguments whose names end with ‘.f’ are assumed to be Fortran source program units; they are compiled,
and the object program is left on the file sfile1.o (i.e. the file whose name is that of the source with ‘.o’ sub-
stituted for ‘.f’).
Other arguments (except for −c) are assumed to be either loader flags, or object programs, typically pro-
duced by an earlier fc run, or perhaps libraries of Fortran-compatible routines. These programs, together
with the results of any compilations specified, are loaded (in the order given) to produce an executable pro-
gram with name a.out.
The −c argument suppresses the loading phase, as does any syntax error in any of the routines being com-
piled.
The following is a list of differences between fc and ANSI standard Fortran (also see the BUGS section):
1. Arbitrary combination of types is allowed in expressions. Not all combinations are expected to be
supported at runtime. All of the normal conversions involving integer, real, double precision and com-
plex are allowed.
2. Two forms of ‘‘implicit’’ statements are recognized: implicit integer /i−n/ or implicit integer (i−n).
3. The types doublecomplex, logical*1, integer*1, integer*2, integer*4 (same as integer), real*4 (real),
and real*8 (double precision) are supported.
4. & as the first character of a line signals a continuation card.
5. c as the first character of a line signals a comment.
6. All keywords are recognized in lower case.
7. The notion of ‘column 7’ is not implemented.
8. G-format input is free form− leading blanks are ignored, the first blank after the start of the number
terminates the field.
9. A comma in any numeric or logical input field terminates the field.
10. There is no carriage control on output.
11. A sequence of n characters in double quotes ‘"’ is equivalent to n h followed by those characters.
12. In data statements, a hollerith string may initialize an array or a sequence of array elements.
13. The number of storage units requested by a binary read must be identical to the number contained in
the record being read.
14. If the first character in an input file is ‘‘#’’, a preprocessor identical to the C preprocessor is called,
which implements ‘‘#define’’ and ‘‘#include’’ preprocessor statements. (See the C reference manual
for details.) The preprocessor does not recognize Hollerith strings written with n h.
In I/O statements, only unit numbers 0-19 are supported. Unit number n refers to file fortnn; (e.g. unit 9 is
file ‘fort09’). For input, the file must exist; for output, it will be created. Unit 5 is permanently associated
with the standard input file; unit 6 with the standard output file. Also see setfil (III) for a way to associate
unit numbers with named files.
FILES
a.out loaded output
f.tmp[123] temporary (deleted)
/usr/fort/fc1 compiler proper
/lib/fr0.o runtime startoff

-1-
-

FC ( I ) 8/20/73 FC ( I )

/lib/filib.a interpreter library

/lib/libf.a builtin functions, etc.
/lib/liba.a system library
SEE ALSO
rc (I), which announces a more pleasant Fortran dialect; the ANSI standard; ld (I) for loader flags. For
some subroutines, try ierror, getarg, setfil (III).
DIAGNOSTICS
Compile-time diagnostics are given in English, accompanied if possible with the offending line number and
source line with an underscore where the error occurred. Runtime diagnostics are given by number as fol-
lows:
1 invalid log argument
2 bad arg count to amod
3 bad arg count to atan2
4 excessive argument to cabs
5 exp too large in cexp
6 bad arg count to cmplx
7 bad arg count to dim
8 excessive argument to exp
9 bad arg count to idim
10 bad arg count to isign
11 bad arg count to mod
12 bad arg count to sign
13 illegal argument to sqrt
14 assigned/computed goto out of range
15 subscript out of range
16 real**real overflow
17 (negative real)**real
100 illegal I/O unit number
101 inconsistent use of I/O unit
102 cannot create output file
103 cannot open input file
104 EOF on input file
105 illegal character in format
106 format does not begin with (
107 no conversion in format but non-empty list
108 excessive parenthesis depth in format
109 illegal format specification
110 illegal character in input field
111 end of format in hollerith specification
112 bad argument to setfil
120 bad argument to ierror
999 unimplemented input conversion
BUGS
The following is a list of those features not yet implemented:
arithmetic statement functions
scale factors on input
Backspace statement.

-2-
-

FILE ( I ) 1/16/75 FILE ( I )

NAME
file − determine file type
SYNOPSIS
file file ...
DESCRIPTION
File performs a series of tests on each argument in an attempt to classify it. If an argument appears to be
ascii, file examines the first 512 bytes and tries to guess its language.
BUGS

-1-
-

FIND ( I ) 5/15/74 FIND ( I )

NAME
find − find files
SYNOPSIS
find pathname expression
DESCRIPTION
Find recursively descends the directory hierarchy from pathname seeking files that match a boolean expres-
sion written in the primaries given below. In the descriptions, the argument n is used as a decimal integer
where +n means more than n, −n means less than n and n means exactly n.
−name filename True if the filename argument matches the current file name. Normal Shell argument
syntax may be used if escaped (watch out for ‘[’, ‘?’ and ‘*’).
−perm onum True if the file permission flags exactly match the octal number onum (see chmod(I)).
If onum is prefixed by a minus sign, more flag bits (017777, see stat(II)) become signif-
icant and the flags are compared: (flags&onum)==onum.
−type c True if the type of the file is c, where c is b, c, d or f for block special file, character
special file, directory or plain file.
−links n True if the file has n links.
−user uname True if the file belongs to the user uname.
−group gname As it is for −user so shall it be for −group (someday).
−size n True if the file is n blocks long (512 bytes per block).
−atime n True if the file has been accessed in n days.
−mtime n True if the file has been modified in n days.
−exec command True if the executed command returns exit status zero (most commands do). The end of
the command is punctuated by an escaped semicolon. A command argument ‘{}’ is re-
placed by the current pathname.
−ok command Like −exec except that the generated command line is printed with a question mark
first, and is executed only if the user responds y.
−print Always true; causes the current pathname to be printed.
The primaries may be combined with these operators (ordered by precedence):
! prefix not
−a infix and, second operand evaluated only if first is true
−o infix or, second operand evaluated only if first is false
( expression ) parentheses for grouping. (Must be escaped.)
To remove files named ‘a.out’ and ‘*.o’ not accessed for a week:
find / "(" −name a.out −o −name "*.o" ")" −a −atime +7 −a −exec rm {} ";"
FILES
/etc/passwd
SEE ALSO
sh (I), if(I), file system (V)
BUGS
There is no way to check device type.
Syntax should be reconciled with if.

-1-
-

GOTO ( I ) 3/15/72 GOTO ( I )

NAME
goto − command transfer
SYNOPSIS
goto label
DESCRIPTION
Goto is allowed only when the Shell is taking commands from a file. The file is searched from the begin-
ning for a line beginning with ‘:’ followed by one or more spaces followed by the label. If such a line is
found, the goto command returns. Since the read pointer in the command file points to the line after the la-
bel, the effect is to cause the Shell to transfer to the labelled line.
‘:’ is a do-nothing command that is ignored by the Shell and only serves to place a label.
SEE ALSO
sh (I)
BUGS

-1-
-

GREP ( I ) 5/15/74 GREP ( I )

NAME
grep − search a file for a pattern
SYNOPSIS
grep [ −v ] [ −b ] [ −c ] [ −n ] expression [ file ] ...
DESCRIPTION
Grep searches the input files (standard input default) for lines matching the regular expression. Normally,
each line found is copied to the standard output. If the −v flag is used, all lines but those matching are
printed. If the −c flag is used, only a count of matching lines is printed. If the −n flag is used, each line is
preceded its relative line number in the file. If the −b flag is used, each line is preceded by the block num-
ber on which it was found. This is sometimes useful in locating disk block numbers by context.
In all cases the file name is shown if there is more than one input file.
For a complete description of the regular expression, see ed (I). Care should be taken when using the char-
acters $ * [ ˆ | ( ) and \ in the regular expression as they are also meaningful to the Shell. It is generally nec-
essary to enclose the entire expression argument in quotes.
SEE ALSO
ed (I), sh (I)
BUGS
Lines are limited to 256 characters; longer lines are truncated.

-1-
-

IF ( I ) 5/2/74 IF ( I )

NAME
if − conditional command
SYNOPSIS
if expr command [ arg ... ]
DESCRIPTION
If evaluates the expression expr, and if its value is true, executes the given command with the given argu-
ments.
The following primitives are used to construct the expr:
−r file true if the file exists and is readable.
−w file true if the file exists and is writable.
s1 = s2 true if the strings s1 and s2 are equal.
s1 != s2 true if the strings s1 and s2 are not equal.
{ command } The bracketed command is executed to obtain the exit status. Status zero is considered
true. The command must not be another if.
These primaries may be combined with the following operators:
! unary negation operator
−a binary and operator
−o binary or operator
( expr ) parentheses for grouping.
−a has higher precedence than −o. Notice that all the operators and flags are separate arguments to if and
hence must be surrounded by spaces. Notice also that parentheses are meaningful to the Shell and must be
escaped.
SEE ALSO
sh (I), find (I)
BUGS

-1-
-

KILL ( I ) 2/8/75 KILL ( I )

NAME
kill − terminate a process
SYNOPSIS
kill [ −signo ] processid ...
DESCRIPTION
Kills the specified processes. The process number of each asynchronous process started with ‘&’ is report-
ed by the Shell. Process numbers can also be found by using ps (I).
If process number 0 is used, then all processes belonging to the current user and associated with the same
control typewriter are killed.
The killed process must belong to the current user unless he is the super-user.
If a signal number preceded by ‘‘−’’ is given as first argument, that signal is sent instead of kill (see signal
(II)).
SEE ALSO
ps (I), sh (I), signal (II)
BUGS

-1-
-

LD ( I ) 8/16/73 LD ( I )

NAME
ld − link editor
SYNOPSIS
ld [ −sulxrdni ] name ...
DESCRIPTION
Ld combines several object programs into one; resolves external references; and searches libraries. In the
simplest case the names of several object programs are given, and ld combines them, producing an object
module which can be either executed or become the input for a further ld run. (In the latter case, the −r op-
tion must be given to preserve the relocation bits.) The output of ld is left on a.out. This file is made ex-
ecutable only if no errors occurred during the load.
The argument routines are concatenated in the order specified. The entry point of the output is the begin-
ning of the first routine.
If any argument is a library, it is searched exactly once at the point it is encountered in the argument list.
Only those routines defining an unresolved external reference are loaded. If a routine from a library refer-
ences another routine in the library, the referenced routine must appear after the referencing routine in the
library. Thus the order of programs within libraries is important.
Ld understands several flag arguments which are written preceded by a ‘−’. Except for −l, they should ap-
pear before the file names.
−s ‘squash’ the output, that is, remove the symbol table and relocation bits to save space (but impair the
usefulness of the debugger). This information can also be removed by strip.
−u take the following argument as a symbol and enter it as undefined in the symbol table. This is useful
for loading wholly from a library, since initially the symbol table is empty and an unresolved reference
is needed to force the loading of the first routine.
−l This option is an abbreviation for a library name. −l alone stands for ‘/lib/liba.a’, which is the stan-
dard system library for assembly language programs. −lx stands for ‘/lib/libx.a’ where x is any charac-
ter. A library is searched when its name is encountered, so the placement of a −l is significant.
−x do not preserve local (non-.globl) symbols in the output symbol table; only enter external symbols.
This option saves some space in the output file.
−X Save local symbols except for those whose names begin with ‘L’. This option is used by cc to discard
internally generated labels while retaining symbols local to routines.
−r generate relocation bits in the output file so that it can be the subject of another ld run. This flag also
prevents final definitions from being given to common symbols, and suppresses the ‘undefined sym-
bol’ diagnostics.
−d force definition of common storage even if the −r flag is present.
−n Arrange that when the output file is executed, the text portion will be read-only and shared among all
users executing the file. This involves moving the data areas up the the first possible 4K word bound-
ary following the end of the text.
−i When the output file is executed, the program text and data areas will live in separate address spaces.
The only difference between this option and −n is that here the data starts at location 0.
FILES
/lib/lib?.a libraries
a.out output file
SEE ALSO
as (I), ar (I)
BUGS

-1-
-

LN ( I ) 3/15/72 LN ( I )

NAME
ln − make a link
SYNOPSIS
ln name1 [ name2 ]
DESCRIPTION
A link is a directory entry referring to a file; the same file (together with its size, all its protection informa-
tion, etc) may have several links to it. There is no way to distinguish a link to a file from its original direc-
tory entry; any changes in the file are effective independently of the name by which the file is known.
Ln creates a link to an existing file name1. If name2 is given, the link has that name; otherwise it is placed
in the current directory and its name is the last component of name1.
It is forbidden to link to a directory or to link across file systems.
SEE ALSO
rm (I)
BUGS
There is nothing particularly wrong with ln, but tp doesn’t understand about links and makes one copy for
each name by which a file is known; thus if the tape is extracted several copies are restored and the infor-
mation that links were involved is lost.

-1-
-

LOGIN ( I ) 3/15/72 LOGIN ( I )

NAME
login − sign onto UNIX
SYNOPSIS
login [ username ]
DESCRIPTION
The login command is used when a user initially signs onto UNIX, or it may be used at any time to change
from one user to another. The latter case is the one summarized above and described here. See ‘How to
Get Started’ for how to dial up initially.
If login is invoked without an argument, it asks for a user name, and, if appropriate, a password. Echoing is
turned off (if possible) during the typing of the password, so it will not appear on the written record of the
session.
After a successful login, accounting files are updated and the user is informed of the existence of and mes-
sage-of-the-day files. Login initializes the user and group IDs and the working directory, then executes a
command interpreter (usually sh (I)) according to specifications found in a password file.
Login is recognized by the Shell and executed directly (without forking).
FILES
/etc/utmp accounting
/usr/adm/wtmp accounting
.mail mail
/etc/motd message-of-the-day
/etc/passwd password file
SEE ALSO
init (VIII), getty (VIII), mail (I), passwd (I), passwd (V)
DIAGNOSTICS
‘login incorrect,’ if the name or the password is bad. ‘No Shell,’, ‘cannot open password file,’ ‘no directo-
ry’: consult a UNIX programming counselor.
BUGS

-1-
-

LS ( I ) 3/20/74 LS ( I )

NAME
ls − list contents of directory
SYNOPSIS
ls [ −ltasdruifg ] name ...
DESCRIPTION
For each directory argument, ls lists the contents of the directory; for each file argument, ls repeats its name
and any other information requested. The output is sorted alphabetically by default. When no argument is
given, the current directory is listed. When several arguments are given, the arguments are first sorted ap-
propriately, but file arguments appear before directories and their contents. There are several options:
−l list in long format, giving mode, number of links, owner, size in bytes, and time of last modification
for each file. (See below.) If the file is a special file the size field will instead contain the major and
minor device numbers.
−t sort by time modified (latest first) instead of by name, as is normal
−a list all entries; usually those beginning with ‘.’ are suppressed
−s give size in blocks for each entry
−d if argument is a directory, list only its name, not its contents (mostly used with −l to get status on di-
rectory)
−r reverse the order of sort to get reverse alphabetic or oldest first as appropriate
−u use time of last access instead of last modification for sorting (−t) or printing (−l)
−i print i-number in first column of the report for each file listed
−f force each argument to be interpreted as a directory and list the name found in each slot. This option
turns off −l, −t, −s, and −r, and turns on −a; the order is the order in which entries appear in the direc-
tory.
−g Give group ID instead of owner ID in long listing.
The mode printed under the −l option contains 11 characters which are interpreted as follows: the first char-
acter is
d if the entry is a directory;
b if the entry is a block-type special file;
c if the entry is a character-type special file;
− if the entry is a plain file.
The next 9 characters are interpreted as three sets of three bits each. The first set refers to owner permis-
sions; the next to permissions to others in the same user-group; and the last to all others. Within each set
the three characters indicate permission respectively to read, to write, or to execute the file as a program.
For a directory, ‘execute’ permission is interpreted to mean permission to search the directory for a speci-
fied file. The permissions are indicated as follows:
r if the file is readable
w if the file is writable
x if the file is executable
− if the indicated permission is not granted
The group-execute permission character is given as s if the file has set-group-ID mode; likewise the user-
execute permission character is given as s if the file has set-user-ID mode.
The last character of the mode is normally blank but is printed as ‘‘t’’ if the 1000 bit of the mode is on. See
chmod (I) for the current meaning of this mode.
FILES
/etc/passwd to get user ID’s for ls −l.

-1-
-

LS ( I ) 3/20/74 LS ( I )

BUGS

-2-
-

MAIL ( I ) 2/21/75 MAIL ( I )

NAME
mail − send mail to designated users
SYNOPSIS
mail [ −yn ] [ person ... ]
DESCRIPTION
Mail with no argument searches for a file called prints it if it is nonempty, then asks if it should be saved. If
the answer is y, the mail is added to mbox. Finally is truncated to zero length. To leave the mailbox un-
touched, hit ‘delete.’ The question can be answered on the command line with the argument ‘−y’ or ‘−n’.
When persons are named, mail takes the standard input up to an end of file and adds it to each person’s file.
The message is preceded by the sender’s name and a postmark.
A person is either a user name recognized by login (I), in which case the mail is sent to the default working
directory of that user; or the path name of a directory, in which case in that directory is used.
When a user logs in he is informed of the presence of mail. No mail will be received from a sender to
whom is inaccessible or unwritable.
FILES
/etc/passwd to identify sender and locate persons
/etc/utmp to identify sender
.mail input mail
mbox saved mail
/tmp/m# temp file
SEE ALSO
write (I)
BUGS

-1-
-

MAN ( I ) 8/20/73 MAN ( I )

NAME
man − run off section of UNIX manual
SYNOPSIS
man [ section ] [ title ... ]
DESCRIPTION
Man is a shell command file which locates and prints one or more sections of this manual. Section is the
section number of the manual, as an Arabic not Roman numeral, and is optional. Title is one or more sec-
tion names; these names bear a generally simple relation to the page captions in the manual. If the section
is missing, 1 is assumed. For example,
man man
would reproduce this page.
FILES
/usr/man/man?/*
BUGS
The manual is supposed to be reproducible either on the phototypesetter or on a typewriter. However, on a
typewriter some information is necessarily lost.

-1-
-

MESG ( I ) 3/15/72 MESG ( I )

NAME
mesg − permit or deny messages
SYNOPSIS
mesg [ n ] [ y ]
DESCRIPTION
Mesg with argument n forbids messages via write by revoking non-user write permission on the user’s
typewriter. Mesg with argument y reinstates permission. All by itself, mesg reverses the current permis-
sion. In all cases the previous state is reported.
FILES
/dev/tty?
SEE ALSO
write (I)
DIAGNOSTICS
‘?’ if the standard input file is not a typewriter
BUGS

-1-
-

MKDIR ( I ) 3/15/72 MKDIR ( I )

NAME
mkdir − make a directory
SYNOPSIS
mkdir dirname ...
DESCRIPTION
Mkdir creates specified directories in mode 777. The standard entries ‘.’ and ‘..’ are made automatically.
SEE ALSO
rmdir (I)
BUGS

-1-
-

MV ( I ) 8/20/73 MV ( I )

NAME
mv − move or rename a file
SYNOPSIS
mv name1 name2
DESCRIPTION
Mv changes the name of name1 to name2. If name2 is a directory, name1 is moved to that directory with
its original file-name. Directories may only be moved within the same parent directory (just renamed).
If name2 already exists, it is removed before name1 is renamed. If name2 has a mode which forbids writ-
ing, mv prints the mode and reads the standard input to obtain a line; if the line begins with y, the move
takes place; if not, mv exits.
If name2 would lie on a different file system, so that a simple rename is impossible, mv copies the file and
deletes the original.
BUGS
It should take a −f flag, like rm, to suppress the question if the target exists and is not writable.

-1-
-

NEQN ( I ) 4/30/74 NEQN ( I )

NAME
neqn − typeset mathematics on terminal
SYNOPSIS
neqn [ file ] ...
DESCRIPTION
Neqn is an nroff (I) preprocessor. The input language is the same as that of eqn (I). Normal usage is almost
always
neqn file ... | nroff
Output is meant for terminals with forward and reverse capabilities, such as the Model 37 teletype or GSI
terminal.
If no arguments are specified, neqn reads the standard input, so it may be used as a filter.
SEE ALSO
eqn (I), gsi (VI)
BUGS
Because of some interactions with nroff there may not always be enough space left before and after lines
containing equations.

-1-
-

NEWGRP ( I ) 4/8/75 NEWGRP ( I )

NAME
newgrp − log in to a new group
SYNOPSIS
newgrp group
DESCRIPTION
Newgrp changes the group identification of its caller, analogously to login. The same person remains
logged in, and the current directory is unchanged, but calculations of access permissions to files are per-
formed with respect to the new group ID.
A password is demanded if the group has a password and the user himself does not.
When most users log in, they are members of the group named ‘other.’
FILES
/etc/group, /etc/passwd
SEE ALSO
login (I), group (V)
BUGS

-1-
-

NICE ( I ) 2/8/75 NICE ( I )

NAME
nice − run a command at low priority
SYNOPSIS
nice [ −number ] command [ arguments ]
DESCRIPTION
Nice executes command with low scheduling priority. If a numerical argument is given, that priority (in the
range 1-20) is used; if not, priority 4 is used.
The super-user may run commands with priority higher than normal by using a negative priority, e.g.
‘−−10’.
SEE ALSO
nohup (I), nice (II)
BUGS

-1-
-

NM ( I ) 8/20/73 NM ( I )

NAME
nm − print name list
SYNOPSIS
nm [ −cnrupg ] [ name ]
DESCRIPTION
Nm prints the symbol table from the output file of an assembler or loader run. Each symbol name is pre-
ceded by its value (blanks if undefined) and one of the letters U (undefined) A (absolute) T (text segment
symbol), D (data segment symbol), B (bss segment symbol), or C (common symbol). If the symbol is local
(non-external) the type letter is in lower case. The output is sorted alphabetically.
If no file is given, the symbols in a.out are listed.
Options are:
−c list only C-style external symbols, that is those beginning with underscore ‘ ’.
−g print only global (external) symbols
−n sort by value instead of by name
−p don’t sort; print in symbol-table order
−r sort in reverse order
−u print only undefined symbols.
FILES
a.out
BUGS

-1-
-

NOHUP ( I ) 11/1/73 NOHUP ( I )

NAME
nohup − run a command immune to hangups
SYNOPSIS
nohup command [ arguments ]
DESCRIPTION
Nohup executes command with hangups, quits and interrupts all ignored.
SEE ALSO
nice (I), signal (II)
BUGS

-1-
-

NROFF ( I ) 4/15/75 NROFF ( I )

NAME
nroff − format text
SYNOPSIS
nroff [ +n ] [ −n ] [ −nn ] [ −ran ] [ −mx ] [ −s ] [ −h ] [ −q ] files
DESCRIPTION
Nroff formats text according to control lines embedded in the text files. Nroff reads the standard input if no
file arguments are given. An argument of just ‘‘−’’ refers to the standard input. The non-file option argu-
ments are interpreted as follows:
+n Output commences at the first page whose page number is n or larger.
−n Printing stops after page n.
−nn First generated (not necessarily printed) page is given number n; simulates ‘‘.pn n’’.
−ran Set number register to the value n.
−mname Prepends a standard macro file; simulates ‘‘.so /usr/lib/tmac.name’’.
−s Stop prior to each page to permit paper loading. Printing is restarted by typing a ‘newline’
character.
−h Spaces are replaced where possible with tabs to speed up output (or reduce the size of the out-
put file).
−q Prompt names for insertions are not printed and the bell character is sent instead; the insertion
is not echoed.

FILES
/usr/lib/suftab suffix hyphenation tables
/tmp/rtm? temporary
/usr/lib/tmac.* standard macro files
SEE ALSO
NROFF User’s Manual (internal memorandum).
neqn (I), col (I)
BUGS

-1-
-

OD ( I ) 1/15/73 OD ( I )

NAME
od − octal dump
SYNOPSIS
od [ −abcdho ] [ file ] [ [ + ] offset[ . ][ b ] ]
DESCRIPTION
Od dumps file in one or more formats as selected by the first argument. If the first argument is missing −o
is default. The meanings of the format argument characters are:
a interprets words as PDP-11 instructions and dis-assembles the operation code. Unknown operation
codes print as ???.
b interprets bytes in octal.
c interprets bytes in ascii. Unknown ascii characters are printed as \?.
d interprets words in decimal.
h interprets words in hex.
o interprets words in octal.

The file argument specifies which file is to be dumped. If no file argument is specified, the standard input is
used. Thus od can be used as a filter.
The offset argument specifies the offset in the file where dumping is to commence. This argument is nor-
mally interpreted as octal bytes. If ‘.’ is appended, the offset is interpreted in decimal. If ‘b’ is appended,
the offset is interpreted in blocks. (A block is 512 bytes.) If the file argument is omitted, the offset argu-
ment must be preceded by ‘+’.
Dumping continues until end-of-file.
SEE ALSO
db (I)
BUGS

-1-
-

OPR ( I ) 7/17/74 OPR ( I )

NAME
opr − off line print
SYNOPSIS
opr [ −destination ] [ −crm ] [ name ... ]
DESCRIPTION
Opr causes the named files to be printed off line at the specified destination. If no names appear the stan-
dard input is assumed.
At the mother system the following destinations are recognized. The default destination is mh.
lp Local line printer.
mh GCOS at Murray Hill Comp Center. GCOS identification must be registered in the UNIX password
file (see passwd (V)).
sp Spider network printer.
xx The two-character code xx is taken to be a Murray Hill GCOS station id. Useful codes are ‘r1’ for
quality print and ‘q1’ for quality print with special ribbon.
Opr uses spooling daemons that do the job when facilities become available. Flag −r causes the named
files to be removed when spooled. Flag −c causes copies to be made so as to insulate the daemons from
any intervening changes to the files.
Flag −m causes mail to be sent when UNIX is finished transmitting the file. For GCOS jobs the mail in-
cludes the snumb.
FILES
/etc/passwd personal ident cards
/lib/dpr dataphone spooler
/etc/dpd dataphone daemon
/usr/dpd/* spool area
/lib/lpr line printer spooler
/etc/lpd line printer daemon
/usr/lpd/* spool area
/lib/npr spider network spooler
SEE ALSO
fsend (I), dpd (VIII), lpd (VIII)
BUGS
Line printer spooler doesn’t handle flags.
Spider network spooler doesn’t spool.

-1-
-

PASSWD ( I ) 9/1/72 PASSWD ( I )

NAME
passwd − change login password
SYNOPSIS
passwd name password
DESCRIPTION
The password becomes associated with the given login name. This can only be done by corresponding user
or by the super-user. An explicit null argument ("") for the password argument removes any password.
FILES
/etc/passwd
SEE ALSO
login (I), passwd (V), crypt (III)
BUGS

-1-
-

PFE ( I ) 11/1/73 PFE ( I )

NAME
pfe − print floating exception
SYNOPSIS
pfe
DESCRIPTION
Pfe examines the floating point exception register and prints a diagnostic for the last floating point excep-
tion.
SEE ALSO
signal (II)
BUGS
Since the system does not save the exception register in a core image file, the message refers to the last er-
ror encountered by anyone. Floating exceptions are therefore volatile.

-1-
-

PR ( I ) 3/20/74 PR ( I )

NAME
pr − print file
SYNOPSIS
pr [ −h header ] [ −n ] [ +n ] [ −wn ] [ −ln ] [ −t ] [ −sc ] [ −m ] name . . .
DESCRIPTION
Pr produces a printed listing of one or more files. The output is separated into pages headed by a date, the
name of the file or a specified header, and the page number. If there are no file arguments, pr prints its stan-
dard input, and is thus usable as a filter.
Options apply to all following files but may be reset between files:
−n produce n-column output
+n begin printing with page n
−h treat the next argument as a header to be used instead of the file name
−wn for purposes of multi-column output, take the width of the page to be n characters instead of the de-
fault 72
−ln take the length of the page to be n lines instead of the default 66
−t do not print the 5-line header or the 5-line trailer normally supplied for each page
−sc separate columns by the single character c instead of by the appropriate amount of white space. A
missing c is taken to be a tab.
−m print all files simultaneously, each in one column
Interconsole messages via write(I) are forbidden during a pr.
FILES
/dev/tty? to suspend messages.
SEE ALSO
cat (I), cp (I)
DIAGNOSTICS
none; files not found are ignored
BUGS

-1-
-

PROF ( I ) 3/12/73 PROF ( I )

NAME
prof − display profile data
SYNOPSIS
prof [ −v ] [ −a ] [ −l ] [ file ]
DESCRIPTION
Prof interprets the file mon.out produced by the monitor subroutine. Under default modes, the symbol table
in the named object file (a.out default) is read and correlated with the mon.out profile file. For each exter-
nal symbol, the percentage of time spent executing between that symbol and the next is printed (in decreas-
ing order), together with the number of times that routine was called and the number of milliseconds per
call.
If the −a option is used, all symbols are reported rather than just external symbols. If the −l option is used,
the output is listed by symbol value rather than decreasing percentage. If the −v option is used, all printing
is suppressed and a profile plot is produced on the 611 display.
In order for the number of calls to a routine to be tallied, the −p option of cc must have been given when the
file containing the routine was compiled. This option also arranges for the mon.out file to be produced au-
tomatically.
FILES
mon.out for profile
a.out for namelist
/dev/vt0 for plotting
SEE ALSO
monitor (III), profil (II), cc (I)
BUGS
Beware of quantization errors.

-1-
-

PS ( I ) 3/20/74 PS ( I )

NAME
ps − process status
SYNOPSIS
ps [ aklx ] [ namelist ]
DESCRIPTION
Ps prints certain indicia about active processes. The a flag asks for information about all processes with
typewriters (ordinarily only one’s own processes are displayed); x asks even about processes with no type-
writer; l asks for a long listing. Ordinarily only the typewriter number (if not one’s own), the process num-
ber, and an approximation to the command line are given. If the k flag is specified, the file /usr/sys/core is
used in place of /dev/mem. This is used for postmortem system debugging. If a second argument is given,
it is taken to be the file containing the system’s namelist.
The long listing is columnar and contains
The name of the process’s control typewriter.
Flags associated with the process. 01: in core; 02: system process; 04: locked in code (e.g. for physi-
cal I/O); 10: being swapped; 20: being traced by another process.
The state of the process. 0: nonexistent; S: sleeping; W: waiting; R: running; Z: terminated; T:
stopped.
The user ID of the process owner.
The process ID of the process; as in certain cults it is possible to kill a process if you know its true
name.
The priority of the process; high numbers mean low priority.
The size in blocks of the core image of the process.
The event for which the process is waiting or sleeping; if blank, the process is running.
The command and its arguments.
Ps makes an educated guess as to the file name and arguments given when the process was created by ex-
amining core memory or the swap area. The method is inherently somewhat unreliable and in any event a
process is entitled to destroy this information, so the names cannot be counted on too much.
FILES
/unix system namelist
/dev/mem core memory
/usr/sys/core alternate core file
/dev searched to find swap device and typewriter names
SEE ALSO
kill (I)
BUGS

-1-
-

PWD ( I ) 5/15/74 PWD ( I )

NAME
pwd − working directory name
SYNOPSIS
pwd
DESCRIPTION
Pwd prints the pathname of the working (current) directory.
SEE ALSO
chdir (I)
BUGS

-1-
-

RC ( I ) 5/15/74 RC ( I )

NAME
rc − Ratfor compiler
SYNOPSIS
rc [ −c ] [ −r ] [ −f ] [ −v ] file ...
DESCRIPTION
Rc invokes the Ratfor preprocessor on a set of Ratfor source files. It accepts three types of arguments:
Arguments whose names end with ‘.r’ are taken to be Ratfor source programs; they are preprocessed into
Fortran and compiled. Each subroutine or function ‘name’ is placed on a separate file name.f, and its object
code is left on name.o. The main routine is on MAIN.f and MAIN.o; block data subprograms go on block-
data?.f and blockdata?.o. The files resulting from a ‘.r’ file are loaded into a single object file file.o, and the
intermediate object and Fortran files are removed.
The following flags are interpreted by rc. See ld (I) for load-time flags.
−c Suppresses the loading phase of the compilation, as does any error in anything.
−f Save Fortran intermediate files. This is primarily for debugging.
−r Ratfor only; don’t try to compile the Fortran. This implies −f and −c.
−v Don’t list intermediate file names while compiling.
Arguments whose names end with ‘.f’ are taken to be Fortran source programs; they are compiled in the
normal manner. (Only one Fortran routine is allowed in a ‘.f’ file.) Other arguments are taken to be either
loader flag arguments, or Fortran-compatible object programs, typically produced by an earlier rc run, or
perhaps libraries of Fortran-compatible routines. These programs, together with the results of any compila-
tions specified, are loaded to produce an executable program with name a.out.
FILES
ratjunk temporary
/usr/bin/ratfor preprocessor
/usr/fort/fc1 Fortran compiler
SEE ALSO
‘‘RATFOR − A Rational Fortran’’.
fc(I) for Fortran error messages.
DIAGNOSTICS
Yes, both from rc itself and from Fortran.
BUGS
Limit of about 50 arguments, 10 block data files.
#define and #include lines in ‘‘.f ’’ files are not processed.

-1-
-

REV ( I ) 4/24/75 REV ( I )

NAME
rev − reverse lines of a file
SYNOPSIS
rev
DESCRIPTION
Rev copies the standard input to the standard output, reversing the order of characters in every line.
BUGS

-1-
-

RM ( I ) 1/20/73 RM ( I )

NAME
rm − remove (unlink) files
SYNOPSIS
rm [ −f ] [ −r ] name ...
DESCRIPTION
Rm removes the entries for one or more files from a directory. If an entry was the last link to the file, the
file is destroyed. Removal of a file requires write permission in its directory, but neither read nor write per-
mission on the file itself.
If a file has no write permission, rm prints the file name and its mode, then reads a line from the standard
input. If the line begins with y, the file is removed, otherwise it is not. The question is not asked if option
−f was given or if the standard input is not a typewriter.
If a designated file is a directory, an error comment is printed unless the optional argument −r has been
used. In that case, rm recursively deletes the entire contents of the specified directory. To remove directo-
ries per se see rmdir(I).
FILES
/etc/glob to implement the −r flag
SEE ALSO
rmdir (I)
BUGS
When rm removes the contents of a directory under the −r flag, full pathnames are not printed in diagnos-
tics.

-1-
-

RMDIR ( I ) 3/15/72 RMDIR ( I )

NAME
rmdir − remove directory
SYNOPSIS
rmdir dir ...
DESCRIPTION
Rmdir removes (deletes) directories. The directory must be empty (except for the standard entries ‘.’ and
‘..’, which rmdir itself removes). Write permission is required in the directory in which the directory to be
removed appears.
BUGS
Needs a −r flag.
Actually, write permission in the directory’s parent is not required.
Mildly unpleasant consequences can follow removal of your own or someone else’s current directory.

-1-
-

ROFF ( I ) 11/4/74 ROFF ( I )

NAME
roff − format text
SYNOPSIS
roff [ +n ] [ −n ] [ −s ] [ −h ] file ...
DESCRIPTION
Roff formats text according to control lines embedded in the text in the given files. Encountering a nonexis-
tent file terminates printing. Incoming interconsole messages are turned off during printing. The optional
flag arguments mean:
+n Start printing at the first page with number n.
−n Stop printing at the first page numbered higher than n.
−s Stop before each page (including the first) to allow paper manipulation; resume on receipt of an inter-
rupt signal.
−h Insert tabs in the output stream to replace spaces whenever appropriate.
Input consists of intermixed text lines, which contain information to be formatted, and request lines, which
contain instructions about how to format it. Request lines begin with a distinguished control character, nor-
mally a period.
Output lines may be filled as nearly as possible with words without regard to input lineation. Line breaks
may be caused at specified places by certain commands, or by the appearance of an empty input line or an
input line beginning with a space.
The capabilities of roff are specified in the attached Request Summary. Numerical values are denoted there
by n or +n, titles by t, and single characters by c. Numbers denoted +n may be signed + or −, in which case
they signify relative changes to a quantity, otherwise they signify an absolute resetting. Missing n fields are
ordinarily taken to be 1, missing t fields to be empty, and c fields to shut off the appropriate special interpre-
tation.
Running titles usually appear at top and bottom of every page. They are set by requests like
.he ′part1′part2′part3′
Part1 is left justified, part2 is centered, and part3 is right justified on the page. Any % sign in a title is re-
placed by the current page number. Any nonblank may serve as a quote.
ASCII tab characters are replaced in the input by a replacement character, normally a space, according to
the column settings given by a .ta command. (See .tr for how to convert this character on output.)
Automatic hyphenation of filled output is done under control of .hy. When a word contains a designated
hyphenation character, that character disappears from the output and hyphens can be introduced into the
word at the marked places only.
FILES
/usr/lib/suftab suffix hyphenation tables
/tmp/rtm? temporary
SEE ALSO
nroff (I), troff (I)
BUGS
Roff is the simplest of the runoff programs, but is utterly frozen.

-1-
-

ROFF ( I ) 11/4/74 ROFF ( I )

REQUEST SUMMARY
Request Break Initial Meaning
.ad yes yes Begin adjusting right margins.
.ar no arabic Arabic page numbers.
.br yes - Causes a line break − the filling of the current line is stopped.
.bl n yes - Insert of n blank lines, on new page if necessary.
.bp +n yes n=1 Begin new page and number it n; no n means ‘+1’.
.cc c no c=. Control character becomes ‘c’.
.ce n yes - Center the next n input lines, without filling.
.de xx no - Define parameterless macro to be invoked by request ‘.xx’ (definition ends
on line beginning ‘..’).
.ds yes no Double space; same as ‘.ls 2’.
.ef t no t=´´´´ Even foot title becomes t.
.eh t no t=´´´´ Even head title becomes t.
.fi yes yes Begin filling output lines.
.fo no t=´´´´ All foot titles are t.
.hc c no none Hyphenation character becomes ‘c’.
.he t no t=´´´´ All head titles are t.
.hx no - Title lines are suppressed.
.hy n no n=1 Hyphenation is done, if n=1; and is not done, if n=0.
.ig no - Ignore input lines through a line beginning with ‘..’.
.in +n yes - Indent n spaces from left margin.
.ix +n no - Same as ‘.in’ but without break.
.li n no - Literal, treat next n lines as text.
.ll +n no n=65 Line length including indent is n characters.
.ls +n yes n=1 Line spacing set to n lines per output line.
.m1 n no n=2 Put n blank lines between the top of page and head title.
.m2 n no n=2 n blank lines put between head title and beginning of text on page.
.m3 n no n=1 n blank lines put between end of text and foot title.
.m4 n no n=3 n blank lines put between the foot title and the bottom of page.
.na yes no Stop adjusting the right margin.
.ne n no - Begin new page, if n output lines cannot fit on present page.
.nn +n no - The next n output lines are not numbered.
.n1 no no Add 5 to page offset; number lines in margin from 1 on each page.
.n2 n no no Add 5 to page offset; number lines from n; stop if n=0.
.ni +n no n=0 Line numbers are indented n.
.nf yes no Stop filling output lines.
.nx filename - Change to input file ‘filename’.
.of t no t=´´´´ Odd foot title becomes t.
.oh t no t=´´´´ Odd head title becomes t.
.pa +n yes n=1 Same as ‘.bp’.
.pl +n no n=66 Total paper length taken to be n lines.
.po +n no n=0 Page offset. All lines are preceded by n spaces.
.ro no arabic Roman page numbers.
.sk n no - Produce n blank pages starting next page.
.sp n yes - Insert block of n blank lines, except at top of page.
.ss yes yes Single space output lines, equivalent to ‘.ls 1’.
.ta n n.. - Pseudotab settings. Initial tab settings are columns 9 17 25 ...
.tc c no space Tab replacement character becomes ‘c’.
.ti +n yes - Temporarily indent next output line n spaces.
.tr cdef.. no - Translate c into d, e into f, etc.
.ul n no - Underline the letters and numbers in the next n input lines.

-2-
-

SH ( I ) 5/15/74 SH ( I )

NAME
sh − shell (command interpreter)
SYNOPSIS
sh [ −t ] [ −c ] [ name [ arg1 ... [ arg9 ] ] ]
DESCRIPTION
Sh is the standard command interpreter. It is the program which reads and arranges the execution of the
command lines typed by most users. It may itself be called as a command to interpret files of commands.
Before discussing the arguments to the Shell used as a command, the structure of command lines them-
selves will be given.
Commands. Each command is a sequence of non-blank command arguments separated by blanks. The
first argument specifies the name of a command to be executed. Except for certain types of special argu-
ments discussed below, the arguments other than the command name are passed without interpretation to
the invoked command.
If the first argument is the name of an executable file, it is invoked; otherwise the string ‘/bin/’ is prepended
to the argument. (In this way most standard commands, which reside in ‘/bin’, are found.) If no such com-
mand is found, the string ‘/usr’ is further prepended (to give ‘/usr/bin/command’) and another attempt is
made to execute the resulting file. (Certain lesser-used commands live in ‘/usr/bin’.)
If a non-directory file has executable mode, but not the form of an executable program (does not begin with
the proper magic number) then it is assumed to be an ASCII file of commands and a new Shell is created to
execute it. See ‘‘Argument passing’’ below.
If the file cannot be found, a diagnostic is printed.
Command lines. One or more commands separated by ‘|’ or ‘ˆ’ constitute a chain of filters. The standard
output of each command but the last is taken as the standard input of the next command. Each command is
run as a separate process, connected by pipes (see pipe(II)) to its neighbors. A command line contained in
parentheses ‘( )’ may appear in place of a simple command as a filter.
A command line consists of one or more pipelines separated, and perhaps terminated by ‘;’ or ‘&’. The
semicolon designates sequential execution. The ampersand causes the preceding pipeline to be executed
without waiting for it to finish. The process id of such a pipeline is reported, so that it may be used if nec-
essary for a subsequent wait or kill.
Termination Reporting. If a command (not followed by ‘&’) terminates abnormally, a message is printed.
(All terminations other than exit and interrupt are considered abnormal.) Termination reports for com-
mands followed by ‘&’ are given upon receipt of the first command subsequent to the termination of the
command, or when a wait is executed. The following is a list of the abnormal termination messages:
Bus error
Trace/BPT trap
Illegal instruction
IOT trap
EMT trap
Bad system call
Quit
Floating exception
Memory violation
Killed
Broken Pipe
If a core image is produced, ‘− Core dumped’ is appended to the appropriate message.
Redirection of I/O. There are three character sequences that cause the immediately following string to be
interpreted as a special argument to the Shell itself. Such an argument may appear anywhere among the ar-
guments of a simple command, or before or after a parenthesized command list, and is associated with that
command or command list.

-1-
-

SH ( I ) 5/15/74 SH ( I )

An argument of the form ‘<arg’ causes the file ‘arg’ to be used as the standard input (file descriptor 0) of
the associated command.
An argument of the form ‘>arg’ causes file ‘arg’ to be used as the standard output (file descriptor 1) for the
associated command. ‘Arg’ is created if it did not exist, and in any case is truncated at the outset.
An argument of the form ‘>>arg’ causes file ‘arg’ to be used as the standard output for the associated com-
mand. If ‘arg’ did not exist, it is created; if it did exist, the command output is appended to the file.
For example, either of the command lines
ls >junk; cat tail >>junk
( ls; cat tail ) >junk
creates, on file ‘junk’, a listing of the working directory, followed immediately by the contents of file ‘tail’.
Either of the constructs ‘>arg’ or ‘>>arg’ associated with any but the last command of a pipeline is ineffec-
tual, as is ‘<arg’ in any but the first.
In commands called by the Shell, file descriptor 2 refers to the standard output of the Shell before any redi-
rection. Thus filters may write diagnostics to a location where they have a chance to be seen.
Generation of argument lists. If any argument contains any of the characters ‘?’, ‘*’ or ‘[’, it is treated
specially as follows. The current directory is searched for files which match the given argument.
The character ‘*’ in an argument matches any string of characters in a file name (including the null string).
The character ‘?’ matches any single character in a file name.
Square brackets ‘[...]’ specify a class of characters which matches any single file-name character in the
class. Within the brackets, each ordinary character is taken to be a member of the class. A pair of charac-
ters separated by ‘−’ places in the class each character lexically greater than or equal to the first and less
than or equal to the second member of the pair.
Other characters match only the same character in the file name.
For example, ‘*’ matches all file names; ‘?’ matches all one-character file names; ‘[ab]*.s’ matches all file
names beginning with ‘a’ or ‘b’ and ending with ‘.s’; ‘?[zi−m]’ matches all two-character file names ending
with ‘z’ or the letters ‘i’ through ‘m’.
If the argument with ‘*’ or ‘?’ also contains a ‘/’, a slightly different procedure is used: instead of the cur-
rent directory, the directory used is the one obtained by taking the argument up to the last ‘/’ before a ‘*’ or
‘?’. The matching process matches the remainder of the argument after this ‘/’ against the files in the de-
rived directory. For example: ‘/usr/dmr/a*.s’ matches all files in directory ‘/usr/dmr’ which begin with ‘a’
and end with ‘.s’.
In any event, a list of names is obtained which match the argument. This list is sorted into alphabetical or-
der, and the resulting sequence of arguments replaces the single argument containing the ‘*’, ‘[’, or ‘?’.
The same process is carried out for each argument (the resulting lists are not merged) and finally the com-
mand is called with the resulting list of arguments.
Quoting. The character ‘\’ causes the immediately following character to lose any special meaning it may
have to the Shell; in this way ‘<’, ‘>’, and other characters meaningful to the Shell may be passed as part
of arguments. A special case of this feature allows the continuation of commands onto more than one line:
a new-line preceded by ‘\’ is translated into a blank.
Sequences of characters enclosed in double (") or single (´) quotes are also taken literally. For example:
ls | pr −h "My directory"
causes a directory listing to be produced by ls, and passed on to pr to be printed with the heading ‘My di-
rectory’. Quotes permit the inclusion of blanks in the heading, which is a single argument to pr.
Argument passing. When the Shell is invoked as a command, it has additional string processing capabili-
ties. Recall that the form in which the Shell is invoked is
sh [ name [ arg1 ... [ arg9 ] ] ]

-2-
-

SH ( I ) 5/15/74 SH ( I )

The name is the name of a file which is read and interpreted. If not given, this subinstance of the Shell con-
tinues to read the standard input file.
In command lines in the file (not in command input), character sequences of the form ‘$n’, where n is a
digit, are replaced by the nth argument to the invocation of the Shell (argn). ‘$0’ is replaced by name.
The argument ‘−t,’ used alone, causes sh to read the standard input for a single line, execute it as a com-
mand, and then exit. This facility replaces the older ‘mini-shell.’ It is useful for interactive programs
which allow users to execute system commands.
The argument ‘−c’ (used with one following argument) causes the next argument to be taken as a command
line and executed. No new-line need be present, but new-line characters are treated appropriately. This fa-
cility is useful as an alternative to ‘−t’ where the caller has already read some of the characters of the com-
mand to be executed.
End of file. An end-of-file in the Shell’s input causes it to exit. A side effect of this fact means that the
way to log out from UNIX is to type an EOT.
Special commands. The following commands are treated specially by the Shell.
chdir is done without spawning a new process by executing sys chdir (II).
login is done by executing /bin/login without creating a new process.
wait is done without spawning a new process by executing sys wait (II).
shift is done by manipulating the arguments to the Shell.
‘:’ is simply ignored.
Command file errors; interrupts. Any Shell-detected error, or an interrupt signal, during the execution of
a command file causes the Shell to cease execution of that file.
Processes that are created with ‘&’ ignore interrupts. Also if such a process has not redirected its input
with a ‘<’, its input is automatically redirected to the zero length file /dev/null.
FILES
/etc/glob, which interprets ‘*’, ‘?’, and ‘[’.
/dev/null as a source of end-of-file.
SEE ALSO
‘The UNIX Time-Sharing System’, CACM, July, 1974, which gives the theory of operation of the Shell.
chdir (I), login (I), wait (I), shift (I)
BUGS
There is no way to redirect the diagnostic output.

-3-
-

SHIFT ( I ) 8/21/73 SHIFT ( I )

NAME
shift − adjust Shell arguments
SYNOPSIS
shift
DESCRIPTION
Shift is used in Shell command files to shift the argument list left by 1, so that old $2 can now be referred to
by $1 and so forth. Shift is useful to iterate over several arguments to a command file. For example, the
command file
: loop
if $1x = x exit
pr −3 $1
shift
goto loop
prints each of its arguments in 3-column format.
Shift is executed within the Shell.
SEE ALSO
sh (I)
BUGS

-1-
-

SIZE ( I ) 9/2/72 SIZE ( I )

NAME
size − size of an object file
SYNOPSIS
size [ object ... ]
DESCRIPTION
Size prints the (decimal) number of bytes required by the text, data, and bss portions, and their sum in octal
and decimal, of each object-file argument. If no file is specified, a.out is used.
BUGS

-1-
-

SLEEP ( I ) 11/1/73 SLEEP ( I )

NAME
sleep − suspend execution for an interval
SYNOPSIS
sleep time
DESCRIPTION
Sleep suspends execution for time seconds. It is used to execute a command in a certain amount of time as
in:
(sleep 105; command)&
Or to execute a command every so often as in this shell command file:
: loop
command
sleep 37
goto loop
SEE ALSO
sleep (II)
BUGS
Time must be less than 65536 seconds.

-1-
-

SORT ( I ) 5/13/75 SORT ( I )

NAME
sort, usort − sort or merge files
SYNOPSIS
sort [ −abdnrtx ] [ +pos [ −pos ] ] . . . [ −mo ] [ name ] . . .
usort [ −umo ] [ name ] . . .
DESCRIPTION
Sort sorts all the named files together and writes the result on the standard output. The name ‘−’ means the
standard input. The standard input is also used if no input file names are given. Thus sort may be used as a
filter.
The default sort key is an entire line. Default ordering is lexicographic in ASCII collating sequence, except
that lower-case letters are considered the same as the corresponding upper-case letters. Non-ASCII bytes
are ignored. The ordering is affected by the flags −abdnrt, one or more of which may appear:
a Do not map lower case letters.
b Leading blanks (spaces and tabs) are not included in fields.
d ‘Dictionary’ order: only letters, digits and blanks are significant in ASCII comparisons.
n An initial numeric string, consisting of optional minus sign, digits and optionally included decimal
point, is sorted by arithmetic value.
r Reverse the sense of comparisons.
tx Tab character between fields is x.
Selected parts of the line, specified by +pos and −pos, may be used as sort keys. Pos has the form m.n,
where m specifies a number of fields to skip, and n a number of characters to skip further into the next field.
A missing is taken to be 0. +pos denotes the beginning of the key; −pos denotes the first position after the
key (end of line by default). The ordering rule may be overridden for a particular key by appending one or
more of the flags abdnr to +pos.
When no tab character has been specified, a field consists of nonblanks and any preceding blanks. Under
the −b flag, leading blanks are excluded from a field. When a tab character has been specified, a field is a
string ending with a tab character.
When keys are specified, later keys are compared only when all earlier ones compare equal. Lines that
compare equal are ordered with all bytes significant.
These flag arguments are also understood:
−m Merge only, the input files are already sorted.
−o The next argument is the name of an output file to use instead of the standard output. This file may be
the same as one of the inputs, except under the merge flag −m.
Usort is a somewhat specialized version of sort which accepts no collating sequence options: order is al-
ways plain ASCII. It also strips out the second and following copies of duplicated lines. A u flag prevents
this stripping. Usort also understands the m and o options in the same way as sort.
FILES
/usr/tmp/stm???
BUGS

-1-
-

SPELL ( I ) 4/15/75 SPELL ( I )

NAME
spell − find spelling errors
SYNOPSIS
spell [ −v ] file ...
DESCRIPTION
Spell collects the words from the named documents, and looks them up in a dictionary. The words not
found are printed on the standard output. Words which are reasonable transformations of dictionary entries
(e.g. a dictionary entry plus s ) are not printed. If no files are given, the input is from the standard input.
If the −v flag is given, all words which are not literally in the dictionary are printed; those which can be
transformed to lie in the dictionary are so marked, and the others are marked with asterisks.
The process takes several minutes.
FILES
/usr/lib/w2006, /usr/dict/words, /usr/lib/spell[123]
SEE ALSO
typo (I)
BUGS

Because of the mapping into lower case and the stripping of special characters, words may be hard to locate
in the original text.
The escape sequences of troff (I) are not correctly recognized.
More suffixes, and perhaps some prefixes, should be added.
The dictionary cannot be distributed because of copyright limitations.

-1-
-

SPLIT ( I ) 1/15/73 SPLIT ( I )

NAME
split − split a file into pieces
SYNOPSIS
split −n [ file [ name ] ]
DESCRIPTION
Split reads file and writes it in n-line pieces (default 1000), as many as necessary, onto a set of output files.
The name of the first output file is name with aa appended, and so on lexicographically. If no output name
is given, x is default.
If no input file is given, or if − is given in its stead, then the standard input file is used.
BUGS

-1-
-

STRIP ( I ) 3/15/72 STRIP ( I )

NAME
strip − remove symbols and relocation bits
SYNOPSIS
strip name ...
DESCRIPTION
Strip removes the symbol table and relocation bits ordinarily attached to the output of the assembler and
loader. This is useful to save space after a program has been debugged.
The effect of strip is the the same as use of the −s option of ld.
FILES
/tmp/stm? temporary file
SEE ALSO
ld (I), as (I)
BUGS

-1-
-

STTY ( I ) 6/12/72 STTY ( I )

NAME
stty − set typewriter options
SYNOPSIS
stty [ option ... ]
DESCRIPTION
Stty sets certain I/O options on the current output typewriter. With no argument, it reports the current set-
tings of the options. The option strings are selected from the following set:
even allow even parity
−even disallow even parity
odd allow odd parity
−odd disallow odd parity
raw raw mode input (no erase, kill, interrupt, quit, EOT; parity bit passed back)
−raw negate raw mode
cooked same as ‘−raw’
−nl allow carriage return for new-line, and output CR-LF for carriage return or new-line
nl accept only new-line to end lines
echo echo back every character typed
−echo do not echo characters
lcase map upper case to lower case
−lcase do not map case
−tabs replace tabs by spaces when printing
tabs preserve tabs
ek reset erase and kill characters back to normal # and @.
erase c set erase character to c.
kill c set kill character to c.
cr0 cr1 cr2 cr3
select style of delay for carriage return (see below)
nl0 nl1 nl2 nl3
select style of delay for linefeed (see below)
tab0 tab1 tab2 tab3
select style of delay for tab (see below)
ff0 ff1
select style of delay for form feed (see below)
tty33 set all modes suitable for Teletype model 33
tty37 set all modes suitable for Teletype model 37
vt05 set all modes suitable for DEC VT05 terminal
tn300 set all modes suitable for GE Terminet 300
ti700 set all modes suitable for Texas Instruments 700 terminal
tek set all modes suitable for Tektronix 4014 terminal
hup hang up dataphone on last close.
−hup do not hang up dataphone on last close.
0 hang up phone line immediately
50 75 110 134 150 200 300 600 1200 1800 2400 4800 9600 exta extb
Set typewriter baud rate to the number given, if possible. (These are the speeds supported by
the DH-11 interface).
The various delay algorithms are tuned to various kinds of terminals. In general the specifications ending
in ‘0’ mean no delay for the corresponding character.
SEE ALSO
stty (II)
BUGS

-1-
-

TEE ( I ) 3/6/74 TEE ( I )

NAME
tee − pipe fitting
SYNOPSIS
tee [ name ... ]
DESCRIPTION
Tee transcribes the standard input to the standard output and makes copies in the named files.
BUGS

-1-
-

TIME ( I ) 8/16/73 TIME ( I )

NAME
time − time a command
SYNOPSIS
time command
DESCRIPTION
The given command is executed; after it is complete, time prints the elapsed time during the command, the
time spent in the system, and the time spent in execution of the command.
The execution time can depend on what kind of memory the program happens to land in; the user time in
MOS is often half what it is in core.
The times are printed on the diagnostic output stream.
BUGS
Elapsed time is accurate to the second, while the CPU times are measured to the 60th second. Thus the
sum of the CPU times can be up to a second larger than the elapsed time.

-1-
-

TP ( I ) 10/15/73 TP ( I )

NAME
tp − manipulate DECtape and magtape
SYNOPSIS
tp [ key ] [ name ... ]
DESCRIPTION
Tp saves and restores files on DECtape or magtape. Its actions are controlled by the key argument. The key
is a string of characters containing at most one function letter and possibly one or more function modifiers.
Other arguments to the command are file or directory names specifying which files are to be dumped, re-
stored, or listed. In all cases, appearance of a directory name refers to the files and (recursively) subdirecto-
ries of that directory.
The function portion of the key is specified by one of the following letters:
r The named files are written on the tape. If files with the same names already exist, they are re-
placed. ‘Same’ is determined by string comparison, so ‘./abc’ can never be the same as
‘/usr/dmr/abc’ even if ‘/usr/dmr’ is the current directory. If no file argument is given, ‘.’ is the de-
fault.
u updates the tape. u is like r, but a file is replaced only if its modification date is later than the date
stored on the tape; that is to say, if it has changed since it was dumped. u is the default command
if none is given.
d deletes the named files from the tape. At least one name argument must be given. This function
is not permitted on magtapes.
x extracts the named files from the tape to the file system. The owner and mode are restored. If no
file argument is given, the entire contents of the tape are extracted.
t lists the names of the specified files. If no file argument is given, the entire contents of the tape is
listed.
The following characters may be used in addition to the letter which selects the function desired.
m Specifies magtape as opposed to DECtape.
0,...,7 This modifier selects the drive on which the tape is mounted. For DECtape, ‘x’ is default; for
magtape ‘0’ is the default.
v Normally tp does its work silently. The v (verbose) option causes it to type the name of each
file it treats preceded by the function letter. With the t function, v gives more information
about the tape entries than just the name.
c means a fresh dump is being created; the tape directory is zeroed before beginning. Usable on-
ly with r and u. This option is assumed with magtape since it is impossible to selectively over-
write magtape.
f causes new entries on tape to be ‘fake’ in that no data is present for these entries. Such fake
entries cannot be extracted. Usable only with r and u.
i Errors reading and writing the tape are noted, but no action is taken. Normally, errors cause a
return to the command level.
w causes tp to pause before treating each file, type the indicative letter and the file name (as with
v) and await the user’s response. Response y means ‘yes’, so the file is treated. Null response
means ‘no’, and the file does not take part in whatever is being done. Response x means ‘exit’;
the tp command terminates immediately. In the x function, files previously asked about have
been extracted already. With r, u, and d no change has been made to the tape.

FILES
/dev/tap?
/dev/mt?

-1-
-

TP ( I ) 10/15/73 TP ( I )

DIAGNOSTICS
Several; the non-obvious one is ‘Phase error’, which means the file changed after it was selected for dump-
ing but before it was dumped.
BUGS
A single file with several links to it is treated like several files.

-2-
-

TR ( I ) 5/20/74 TR ( I )

NAME
tr − transliterate
SYNOPSIS
tr [ −cds ] [ string1 [ string2 ] ]
DESCRIPTION
Tr copies the standard input to the standard output with substitution or deletion of selected characters. In-
put characters found in string1 are mapped into the corresponding characters of string2. Any combination
of the options −cds may be used. −c complements the set of characters in string1 with respect to the uni-
verse of characters whose ascii codes are 001 through 377 octal. −d deletes all input characters in string1.
−s squeezes all strings of repeated output characters that are in string2 to single characters.
The following abbreviation conventions may be used to introduce ranges of characters or repeated charac-
ters into the strings:
[a −b ] stands for the string of characters whose ascii codes run from character a to character b.
[a *n ], where n is an integer or empty, stands for n-fold repetition of character a. n is taken to be octal or
decimal according as its first digit is or is not zero. A zero or missing n is taken to be huge; this facility is
useful for padding string2.
The escape character ‘\’ may be used as in sh to remove special meaning from any character in a string. In
addition, ‘\’ followed by 1, 2 or 3 octal digits stands for the character whose ascii code is given by those
digits.
The following example creates a list of all the words in ‘file1’ one per line in ‘file2’, where a word is taken
to be a maximal string of alphabetics. The strings are quoted to protect the special characters from interpre-
tation by the Shell; 012 is the ascii code for newline.
tr −cs "[A−Z][a−z]" "[\012*]" <file1 >file2
SEE ALSO
sh (I), ed (I), ascii (V)
BUGS
Won’t handle ascii NUL in string1 or string2; always deletes NUL from input.

-1-
-

TROFF ( I ) 4/15/75 TROFF ( I )

NAME
troff − format text
SYNOPSIS
troff [ +n ] [ −n ] [ −sn ] [ −nn ] [ −ran ] [ −mname ] [ −t ] [ −f ] [ −w ] [ −a ] [ −pn ] files
DESCRIPTION
Troff formats text for a Graphic Systems phototypesetter according to control lines embedded in the text
files. It reads the standard input if no file arguments are given. An argument of just ‘‘−’’ refers to the stan-
dard input. The non-file option arguments are interpreted as follows:
+n Commence typesetting at the first page numbered n or larger.
−n Stop after page n.
−sn Print output in groups of n pages, stopping the typesetter after each group.
−nn First generated (not necessarily printed) page is given the number n; simulates ‘‘.pn n’’.
−ran Set number register a to the value n.
−mname Prepends a standard macro file; simulates ‘‘.so /usr/lib/tmac.name’’.
−t Place output on standard output instead of the phototypesetter.
−f Refrain from feeding out paper and stopping the phototypesetter at the end.
−w Wait until phototypesetter is available, if currently busy.
−a Send a printable approximation of the results to the standard output.
−pn Print all characters with point-size n while retaining all prescribed spacings and motions.

FILES
/usr/lib/suftab suffix hyphenation tables
/tmp/rtm? temporary
/usr/lib/tmac.* standard macro files
SEE ALSO
TROFF User’s Manual (internal memorandum).
TROFF Made Trivial (internal memorandum).
nroff (I), eqn (I), catsim (VI)
BUGS

-1-
-

TTY ( I ) 3/15/72 TTY ( I )

NAME
tty − get typewriter name
SYNOPSIS
tty
DESCRIPTION
Tty gives the name of the user’s typewriter in the form ‘ttyn’ for n a digit or letter. The actual path name is
then ‘/dev/ttyn’.
DIAGNOSTICS
‘not a tty’ if the standard input file is not a typewriter.
BUGS

-1-
-

TYPO ( I ) 5/15/74 TYPO ( I )

NAME
typo − find possible typos
SYNOPSIS
typo [ −1 ] [ −n ] file ...
DESCRIPTION
Typo hunts through a document for unusual words, typographic errors, and hapax legomena and prints them
on the standard output.
The words used in the document are printed out in decreasing order of peculiarity along with an index of
peculiarity. An index of 10 or more is considered peculiar. Printing of certain very common English words
is suppressed.
The statistics for judging words are taken from the document itself, with some help from known statistics of
English. The −n option suppresses the help from English and should be used if the document is written in,
for example, Urdu.
The −1 option causes the final output to appear in a single column instead of three columns. The normal
header and pagination is also suppressed.
Roff (I) and nroff (I) control lines are ignored. Upper case is mapped into lower case. Quote marks, verti-
cal bars, hyphens, and ampersands within words are equivalent to spaces. Words hyphenated across lines
are put back together.
FILES
/tmp/ttmp??
/usr/lib/salt
/usr/lib/w2006
BUGS
Because of the mapping into lower case and the stripping of special characters, words may be hard to locate
in the original text.
The escape sequences of troff (I) are not correctly recognized.

-1-
-

UNIQ ( I ) 12/1/72 UNIQ ( I )

NAME
uniq − report repeated lines in a file
SYNOPSIS
uniq [ −udc [ +n ] [ −n ] ] [ input [ output ] ]
DESCRIPTION
Uniq reads the input file comparing adjacent lines. In the normal case, the second and succeeding copies of
repeated lines are removed; the remainder is written on the output file. Note that repeated lines must be ad-
jacent in order to be found; see sort(I). If the −u flag is used, just the lines that are not repeated in the origi-
nal file are output. The −d option specifies that one copy of just the repeated lines is to be written. The
normal mode output is the union of the −u and −d mode outputs.
The −c option supersedes −u and −d and generates an output report in default style but with each line pre-
ceded by a count of the number of times it occurred.
The n arguments specify skipping an initial portion of each line in the comparison:
−n The first n fields together with any blanks before each are ignored. A field is defined as a string
of non-space, non-tab characters separated by tabs and spaces from its neighbors.
+n The first n characters are ignored. Fields are skipped before characters.

SEE ALSO
sort (I), comm (I)
BUGS

-1-
-

WAIT ( I ) 4/9/73 WAIT ( I )

NAME
wait − await completion of process
SYNOPSIS
wait
DESCRIPTION
Wait until all processes started with & have completed, and report on abnormal terminations.
Because sys wait must be executed in the parent process, the Shell itself executes wait, without creating a
new process.
SEE ALSO
sh (I)
BUGS
After executing wait you are committed to waiting until termination, because interrupts and quits are ig-
nored by all processes concerned. The only out, if the process does not terminate, is to kill it from another
terminal or to hang up.

-1-
-

WC ( I ) 7/26/74 WC ( I )

NAME
wc − word count
SYNOPSIS
wc [ name ... ]
DESCRIPTION
Wc counts lines and words in the named files, or in the standard input if no name appears. A word is a
maximal string of printing characters delimited by spaces, tabs or newlines. All other characters are simply
ignored.
BUGS

-1-
-

WHO ( I ) 3/15/72 WHO ( I )

NAME
who − who is on the system
SYNOPSIS
who [ who-file ] [ am I ]
DESCRIPTION
Who, without an argument, lists the name, typewriter channel, and login time for each current UNIX user.
Without an argument, who examines the /etc/utmp file to obtain its information. If a file is given, that file is
examined. Typically the given file will be /usr/adm/wtmp, which contains a record of all the logins since it
was created. Then who lists logins, logouts, and crashes since the creation of the wtmp file. Each login is
listed with user name, typewriter name (with ‘/dev/’ suppressed), and date and time. When an argument is
given, logouts produce a similar line without a user name. Reboots produce a line with ‘x’ in the place of
the device name, and a fossil time indicative of when the system went down.
With two arguments, who behaves as if it had no arguments except for restricting the printout to the line for
the current typewriter. Thus ‘who am I’ (and also ‘who are you’) tells you who you are logged in as.
FILES
/etc/utmp
SEE ALSO
login (I), init (VIII)
BUGS

-1-
-

WRITE ( I ) 8/5/73 WRITE ( I )

NAME
write − write to another user
SYNOPSIS
write user [ ttyno ]
DESCRIPTION
Write copies lines from your typewriter to that of another user. When first called, it sends the message
message from yourname...
The recipient of the message should write back at this point. Communication continues until an end of file
is read from the typewriter or an interrupt is sent. At that point write writes ‘EOT’ on the other terminal
and exits.
If you want to write to a user who is logged in more than once, the ttyno argument may be used to indicate
the last character of the appropriate typewriter name.
Permission to write may be denied or granted by use of the mesg command. At the outset writing is al-
lowed. Certain commands, in particular roff and pr, disallow messages in order to prevent messy output.
If the character ‘!’ is found at the beginning of a line, write calls the shell to execute the rest of the line as a
command.
The following protocol is suggested for using write: when you first write to another user, wait for him to
write back before starting to send. Each party should end each message with a distinctive signal ( (o) for
‘over’ is conventional) that the other may reply. (oo) (for ‘over and out’) is suggested when conversation is
about to be terminated.
FILES
/etc/utmp to find user
/bin/sh to execute ‘!’
SEE ALSO
mesg (I), who (I), mail (I)
BUGS

-1-
-

YACC ( I ) 11/25/74 YACC ( I )

NAME
yacc − yet another compiler-compiler
SYNOPSIS
yacc [ −vor ] [ grammar ]
DESCRIPTION
Yacc converts a context-free grammar into a set of tables for a simple automaton which executes an LR(1)
parsing algorithm. The grammar may be ambiguous; specified precedence rules are used to break ambigui-
ties.
The output is y.tab.c, which must be compiled by the C compiler and loaded with any other routines re-
quired (perhaps a lexical analyzer) and the Yacc library:
cc y.tab.c other.o −ly
If the −v flag is given, the file y.output is prepared, which contains a description of the parsing tables and a
report on conflicts generated by ambiguities in the grammar.
The −o flag calls an optimizer for the tables; the optimized tables, with parser included, appear on file
y.tab.c
The −r flag causes Yacc to accept grammars with Ratfor actions, and produce Ratfor output on y.tab.r; −r
implies the −o flag. Typical usage is then
rc y.tab.r other.o

SEE ALSO
‘‘LR Parsing’’, by A. V. Aho and S. C. Johnson, Computing Surveys, June, 1974. ‘‘The YACC Compiler-
compiler’’, internal memorandum.
AUTHOR
S. C. Johnson
FILES
y.output
y.tab.c
y.tab.r when ratfor output is obtained
yacc.tmp when optimizer is called
/lib/liby.a runtime library for compiler
/usr/yacc/fpar.r ratfor parser
/usr/yacc/opar.c parser for optimized tables
/usr/yacc/yopti optimizer postpass
DIAGNOSTICS
The number of reduce-reduce and shift-reduce conflicts is reported on the standard output; a more detailed
report is found in the y.output file.
BUGS
Because file names are fixed, at most one Yacc process can be active in a given directory at a time.

-1-
-

BREAK ( II ) 8/5/73 BREAK ( II )

NAME
break, brk, sbrk − change core allocation
SYNOPSIS
(break = 17.)
sys break; addr
char *brk(addr)
char *sbrk(incr)
DESCRIPTION
Break sets the system’s idea of the lowest location not used by the program (called the break) to addr
(rounded up to the next multiple of 64 bytes). Locations not less than addr and below the stack pointer are
not in the address space and will thus cause a memory violation if accessed.
From C, brk will set the break to addr. The old break is returned.
In the alternate entry sbrk, incr more bytes are added to the program’s data space and a pointer to the start
of the new area is returned.
When a program begins execution via exec the break is set at the highest location defined by the program
and data storage areas. Ordinarily, therefore, only programs with growing data areas need to use break.
SEE ALSO
exec (II), alloc (III), end (III)
DIAGNOSTICS
The c-bit is set if the program requests more memory than the system limit or if more than 8 segmentation
registers would be required to implement the break. From C, −1 is returned for these errors.
BUGS
Setting the break in the range 0177700 to 0177777 is the same as setting it to zero.

-1-
-

CHDIR ( II ) 8/5/73 CHDIR ( II )

NAME
chdir − change working directory
SYNOPSIS
(chdir = 12.)
sys chdir; dirname
chdir(dirname)
char *dirname;
DESCRIPTION
Dirname is the address of the pathname of a directory, terminated by a null byte. Chdir causes this directo-
ry to become the current working directory.
SEE ALSO
chdir (I)
DIAGNOSTICS
The error bit (c-bit) is set if the given name is not that of a directory or is not readable. From C, a −1 re-
turned value indicates an error, 0 indicates success.

-1-
-

CHMOD ( II ) 12/15/74 CHMOD ( II )

NAME
chmod − change mode of file
SYNOPSIS
(chmod = 15.)
sys chmod; name; mode
chmod(name, mode)
char *name;
DESCRIPTION
The file whose name is given as the null-terminated string pointed to by name has its mode changed to
mode. Modes are constructed by ORing together some combination of the following:
4000 set user ID on execution
2000 set group ID on execution
1000 save text image after execution
0400 read by owner
0200 write by owner
0100 execute (search on directory) by owner
0070 read, write, execute (search) by group
0007 read, write, execute (search) by others
Only the owner of a file (or the super-user) may change the mode. Only the super-user can set the 1000
mode.
SEE ALSO
chmod (I)
DIAGNOSTIC
Error bit (c-bit) set if name cannot be found or if current user is neither the owner of the file nor the super-
user. From C, a −1 returned value indicates an error, 0 indicates success.

-1-
-

CHOWN ( II ) 12/15/74 CHOWN ( II )

NAME
chown − change owner and group of a file
SYNOPSIS
(chmod = 16.)
sys chown; name; owner
chown(name, owner)
char *name;
DESCRIPTION
The file whose name is given by the null-terminated string pointed to by name has its owner and group
changed to the low and high bytes of owner respectively. Only the super-user may execute this call, be-
cause if users were able to give files away, they could defeat the (nonexistent) file-space accounting proce-
dures.
SEE ALSO
chown (VIII), chgrp (VIII), passwd (V)
DIAGNOSTICS
The error bit (c-bit) is set on illegal owner changes. From C a −1 returned value indicates error, 0 indicates
success.

-1-
-

CLOSE ( II ) 8/5/73 CLOSE ( II )

NAME
close − close a file
SYNOPSIS
(close = 6.)
(file descriptor in r0)
sys close
close(fildes)

DESCRIPTION
Given a file descriptor such as returned from an open, creat, or pipe call, close closes the associated file. A
close of all files is automatic on exit, but since processes are limited to 15 simultaneously open files, close
is necessary for programs which deal with many files.
SEE ALSO
creat (II), open (II), pipe (II)
DIAGNOSTICS
The error bit (c-bit) is set for an unknown file descriptor. From C a −1 indicates an error, 0 indicates suc-
cess.

-1-
-

CREAT ( II ) 8/5/73 CREAT ( II )

NAME
creat − create a new file
SYNOPSIS
(creat = 8.)
sys creat; name; mode
(file descriptor in r0)
creat(name, mode)
char *name;
DESCRIPTION
Creat creates a new file or prepares to rewrite an existing file called name, given as the address of a null-ter-
minated string. If the file did not exist, it is given mode mode. See chmod (II) for the construction of the
mode argument.
If the file did exist, its mode and owner remain unchanged but it is truncated to 0 length.
The file is also opened for writing, and its file descriptor is returned (in r0).
The mode given is arbitrary; it need not allow writing. This feature is used by programs which deal with
temporary files of fixed names. The creation is done with a mode that forbids writing. Then if a second in-
stance of the program attempts a creat, an error is returned and the program knows that the name is unus-
able for the moment.
SEE ALSO
write (II), close (II), stat (II)
DIAGNOSTICS
The error bit (c-bit) may be set if: a needed directory is not searchable; the file does not exist and the direc-
tory in which it is to be created is not writable; the file does exist and is unwritable; the file is a directory;
there are already too many files open.
From C, a −1 return indicates an error.

-1-
-

CSW ( II ) 8/5/73 CSW ( II )

NAME
csw − read console switches
SYNOPSIS
(csw = 38.; not in assembler)
sys csw
getcsw( )
DESCRIPTION
The setting of the console switches is returned (in r0).

-1-
-

DUP ( II ) 8/5/73 DUP ( II )

NAME
dup − duplicate an open file descriptor
SYNOPSIS
(dup = 41.; not in assembler)
(file descriptor in r0)
sys dup
dup(fildes)
int fildes;
DESCRIPTION
Given a file descriptor returned from an open, pipe, or creat call, dup will allocate another file descriptor
synonymous with the original. The new file descriptor is returned in r0.
Dup is used more to reassign the value of file descriptors than to genuinely duplicate a file descriptor. Since
the algorithm to allocate file descriptors returns the lowest available value, combinations of dup and close
can be used to manipulate file descriptors in a general way. This is handy for manipulating standard input
and/or standard output.
SEE ALSO
creat (II), open (II), close (II), pipe (II)
DIAGNOSTICS
The error bit (c-bit) is set if: the given file descriptor is invalid; there are already too many open files. From
C, a −1 returned value indicates an error.

-1-
-

EXEC ( II ) 8/5/73 EXEC ( II )

NAME
exec, execl, execv − execute a file
SYNOPSIS
(exec = 11.)
sys exec; name; args
...
name: <...\0>
...
args: arg0; arg1; ...; 0
arg0: <...\0>
arg1: <...\0>
...
execl(name, arg0, arg1, ..., argn, 0)
char *name, *arg0, *arg1, ..., *argn;
execv(name, argv)
char *name;
char *argv[ ];
DESCRIPTION
Exec overlays the calling process with the named file, then transfers to the beginning of the core image of
the file. There can be no return from the file; the calling core image is lost.
Files remain open across exec calls. Ignored signals remain ignored across exec, but signals that are caught
are reset to their default values.
Each user has a real user ID and group ID and an effective user ID and group ID. The real ID identifies the
person using the system; the effective ID determines his access privileges. Exec changes the effective user
and group ID to the owner of the executed file if the file has the ‘‘set-user-ID’’ or ‘‘set-group-ID’’ modes.
The real user ID is not affected.
The form of this call differs somewhat depending on whether it is called from assembly language or C; see
below for the C version.
The first argument to exec is a pointer to the name of the file to be executed. The second is the address of a
null-terminated list of pointers to arguments to be passed to the file. Conventionally, the first argument is
the name of the file. Each pointer addresses a string terminated by a null byte.
Once the called file starts execution, the arguments are available as follows. The stack pointer points to a
word containing the number of arguments. Just above this number is a list of pointers to the argument
strings. The arguments are placed as high as possible in core.
sp→ nargs
arg0
...
argn
−1
arg0: <arg0\0>
...
argn: <argn\0>
From C, two interfaces are available. execl is useful when a known file with known arguments is being
called; the arguments to execl are the character strings constituting the file and the arguments; as in the ba-
sic call, the first argument is conventionally the same as the file name (or its last component). A 0 argument
must end the argument list.
The execv version is useful when the number of arguments is unknown in advance; the arguments to execv
are the name of the file to be executed and a vector of strings containing the arguments. The last argument
string must be followed by a 0 pointer.

-1-
-

EXEC ( II ) 8/5/73 EXEC ( II )

When a C program is executed, it is called as follows:

main(argc, argv)
int argc;
char **argv;
where argc is the argument count and argv is an array of character pointers to the arguments themselves.
As indicated, argc is conventionally at least one and the first member of the array points to a string contain-
ing the name of the file.
Argv is not directly usable in another execv, since argv[argc] is −1 and not 0.
SEE ALSO
fork (II)
DIAGNOSTICS
If the file cannot be found, if it is not executable, if it does not have a valid header (407, 410, or 411 octal as
first word), if maximum memory is exceeded, or if the arguments require more than 512 bytes a return from
exec constitutes the diagnostic; the error bit (c-bit) is set. Even for the super-user, at least one of the ex-
ecute-permission bits must be set for a file to be executed. From C the returned value is −1.
BUGS
Only 512 characters of arguments are allowed.

-2-
-

EXIT ( II ) 8/5/73 EXIT ( II )

NAME
exit − terminate process

SYNOPSIS
(exit = 1.)
(status in r0)
sys exit
exit(status)
int status;
DESCRIPTION
Exit is the normal means of terminating a process. Exit closes all the process’s files and notifies the parent
process if it is executing a wait. The low byte of r0 (resp. the argument to exit) is available as status to the
parent process.
This call can never return.
SEE ALSO
wait (II)
DIAGNOSTICS
None.

-1-
-

FORK ( II ) 8/5/73 FORK ( II )

NAME
fork − spawn new process
SYNOPSIS
(fork = 2.)
sys fork
(new process return)
(old process return)
fork( )
DESCRIPTION
Fork is the only way new processes are created. The new process’s core image is a copy of that of the caller
of fork. The only distinction is the return location and the fact that r0 in the old (parent) process contains
the process ID of the new (child) process. This process ID is used by wait.
The two returning processes share all open files that existed before the call. In particular, this is the way
that standard input and output files are passed and also how pipes are set up.
From C, the child process receives a 0 return, and the parent receives a non-zero number which is the pro-
cess ID of the child; a return of −1 indicates inability to create a new process.
SEE ALSO
wait (II), exec (II)
DIAGNOSTICS
The error bit (c-bit) is set in the old process if a new process could not be created because of lack of process
space. From C, a return of −1 (not just negative) indicates an error.

-1-
-

FSTAT ( II ) 8/5/73 FSTAT ( II )

NAME
fstat − get status of open file

SYNOPSIS
(fstat = 28.)
(file descriptor in r0)
sys fstat; buf
fstat(fildes, buf)
struct inode *buf;
DESCRIPTION
This call is identical to stat, except that it operates on open files instead of files given by name. It is most
often used to get the status of the standard input and output files, whose names are unknown.
SEE ALSO
stat (II)
DIAGNOSTICS
The error bit (c-bit) is set if the file descriptor is unknown; from C, a −1 return indicates an error, 0 indi-
cates success.

-1-
-

GETGID ( II ) 5/15/74 GETGID ( II )

NAME
getgid − get group identifications
SYNOPSIS
(getgid = 47.; not in assembler)
sys getgid
getgid( )
DESCRIPTION
Getgid returns a word (in r0), the low byte of which contains the real group ID of the current process. The
high byte contains the effective group ID of the current process. The real group ID identifies the group of
the person who is logged in, in contradistinction to the effective group ID, which determines his access per-
mission at the moment. It is thus useful to programs which operate using the ‘‘set group ID’’ mode, to find
out who invoked them.
SEE ALSO
setgid (II)
DIAGNOSTICS
−

-1-
-

GETPID ( II ) 2/8/75 GETPID ( II )

NAME
getpid − get process identification
SYNOPSIS
(getpid = 20.; not in assembler)
sys getpid
(pid in r0)
getpid( )
DESCRIPTION
Getpid returns the process ID of the current process. Most often it is used to generate uniquely-named tem-
porary files.
SEE ALSO
−
DIAGNOSTICS
−

-1-
-

GETUID ( II ) 5/15/74 GETUID ( II )

NAME
getuid − get user identifications
SYNOPSIS
(getuid = 24.)
sys getuid
getuid( )
DESCRIPTION
Getuid returns a word (in r0), the low byte of which contains the real user ID of the current process. The
high byte contains the effective user ID of the current process. The real user ID identifies the person who is
logged in, in contradistinction to the effective user ID, which determines his access permission at the mo-
ment. It is thus useful to programs which operate using the ‘‘set user ID’’ mode, to find out who invoked
them.
SEE ALSO
setuid (II)
DIAGNOSTICS
−

-1-
-

GTTY ( II ) 8/5/73 GTTY ( II )

NAME
gtty − get typewriter status
SYNOPSIS
(gtty = 32.)
(file descriptor in r0)
sys gtty; arg
...
arg: .=.+6
gtty(fildes, arg)
int arg[3];
DESCRIPTION
Gtty stores in the three words addressed by arg the status of the typewriter whose file descriptor is given in
r0 (resp. given as the first argument). The format is the same as that passed by stty.
SEE ALSO
stty (II)
DIAGNOSTICS
Error bit (c-bit) is set if the file descriptor does not refer to a typewriter. From C, a −1 value is returned for
an error, 0, for a successful call.

-1-
-

INDIR ( II ) 8/5/73 INDIR ( II )

NAME
indir − indirect system call
SYNOPSIS
(indir = 0.; not in assembler)
sys indir; syscall
DESCRIPTION
The system call at the location syscall is executed. Execution resumes after the indir call.
The main purpose of indir is to allow a program to store arguments in system calls and execute them out of
line in the data segment. This preserves the purity of the text segment.
If indir is executed indirectly, it is a no-op. If the instruction at the indirect location is not a system call, the
executing process will get a fault.
SEE ALSO
−
DIAGNOSTICS
−

-1-
-

INTRO ( II ) 11/5/73 INTRO ( II )

INTRODUCTION TO SYSTEM CALLS

Section II of this manual lists all the entries into the system. In most cases two calling sequences are specified, one
of which is usable from assembly language, and the other from C. Most of these calls have an error return. From
assembly language an erroneous call is always indicated by turning on the c-bit of the condition codes. The pres-
ence of an error is most easily tested by the instructions bes and bec (‘‘branch on error set (or clear)’’). These are
synonyms for the bcs and bcc instructions.

From C, an error condition is indicated by an otherwise impossible returned value. Almost always this is −1; the in-
dividual sections specify the details.

In both cases an error number is also available. In assembly language, this number is returned in r0 on erroneous
calls. From C, the external variable errno is set to the error number. Errno is not cleared on successful calls, so it
should be tested only after an error has occurred. There is a table of messages associated with each error, and a rou-
tine for printing the message. See perror (III).

The possible error numbers are not recited with each writeup in section II, since many errors are possible for most of
the calls. Here is a list of the error numbers, their names inside the system (for the benefit of system-readers), and
the messages available using perror. A short explanation is also provided.

0 − (unused)

1 EPERM Not owner and not super-user

Typically this error indicates an attempt to modify a file in some way forbidden except to its owner. It is also
returned for attempts by ordinary users to do things allowed only to the super-user.

2 ENOENT No such file or directory

This error occurs when a file name is specified and the file should exist but doesn’t, or when one of the direc-
tories in a path name does not exist.

3 ESRCH No such process

The process whose number was given to signal does not exist, or is already dead.

4 EINTR Interrupted system call

An asynchronous signal (such as interrupt or quit), which the user has elected to catch, occurred during a
system call. If execution is resumed after processing the signal, it will appear as if the interrupted system
call returned this error condition.

5 EIO I/O error

Some physical I/O error occurred during a read or write. This error may in some cases occur on a call fol-
lowing the one to which it actually applies.

6 ENXIO No such device or address

I/O on a special file refers to a subdevice which does not exist, or beyond the limits of the device. It may al-
so occur when, for example, a tape drive is not dialled in or no disk pack is loaded on a drive.

7 E2BIG Arg list too long

An argument list longer than 512 bytes (counting the null at the end of each argument) is presented to exec.

8 ENOEXEC Exec format error

A request is made to execute a file which, although it has the appropriate permissions, does not start with one
of the magic numbers 407 or 410.

9 EBADF Bad file number

Either a file descriptor refers to no open file, or a read (resp. write) request is made to a file which is open on-

-1-
-

INTRO ( II ) 11/5/73 INTRO ( II )

ly for writing (resp. reading).

10 ECHILD No children
Wait and the process has no living or unwaited-for children.

11 EAGAIN No more processes

In a fork, the system’s process table is full and no more processes can for the moment be created.

12 ENOMEM Not enough core

During an exec or break, a program asks for more core than the system is able to supply. This is not a tempo-
rary condition; the maximum core size is a system parameter. The error may also occur if the arrangement of
text, data, and stack segments is such as to require more than the existing 8 segmentation registers.

13 EACCES Permission denied

An attempt was made to access a file in a way forbidden by the protection system.

14 − (unused)

15 ENOTBLK Block device required

A plain file was mentioned where a block device was required, e.g. in mount.

16 EBUSY Mount device busy

An attempt to mount a device that was already mounted or an attempt was made to dismount a device on
which there is an open file or some process’s current directory.

17 EEXIST File exists

An existing file was mentioned in an inappropriate context, e.g. link.

18 EXDEV Cross-device link

A link to a file on another device was attempted.

19 ENODEV No such device

An attempt was made to apply an inappropriate system call to a device; e.g. read a write-only device.

20 ENOTDIR Not a directory

A non-directory was specified where a directory is required, for example in a path name or as an argument to
chdir.

21 EISDIR Is a directory
An attempt to write on a directory.

22 EINVAL Invalid argument

Some invalid argument: currently, dismounting a non-mounted device, mentioning an unknown signal in sig-
nal, and giving an unknown request in stty to the TIU special file.

23 ENFILE File table overflow

The system’s table of open files is full, and temporarily no more opens can be accepted.

24 EMFILE Too many open files

Only 15 files can be open per process.

25 ENOTTY Not a typewriter

The file mentioned in stty or gtty is not a typewriter or one of the other devices to which these calls apply.

26 ETXTBSY Text file busy

An attempt to execute a pure-procedure program which is currently open for writing (or reading!). Also an
attempt to open for writing a pure-procedure program that is being executed.

-2-
-

INTRO ( II ) 11/5/73 INTRO ( II )

27 EFBIG File too large

An attempt to make a file larger than the maximum of 32768 blocks.

28 ENOSPC No space left on device

During a write to an ordinary file, there is no free space left on the device.

29 ESPIPE Seek on pipe

A seek was issued to a pipe. This error should also be issued for other non-seekable devices.

30 EROFS Read-only file system

An attempt to modify a file or directory was made on a device mounted read-only.

31 EMLINK Too many links

An attempt to make more than 127 links to a file.

32 EPIPE Write on broken pipe

A write on a pipe for which there is no process to read the data. This condition normally generates a signal;
the error is returned if the signal is ignored.

-3-
-

KILL ( II ) 12/15/74 KILL ( II )

NAME
kill − send signal to a process
SYNOPSIS
(kill = 37.; not in assembler)
(process number in r0)
sys kill; sig
kill(pid, sig);
DESCRIPTION
Kill sends the signal sig to the process specified by the process number in r0. See signal (II) for a list of
signals.
The sending and receiving processes must have the same effective user ID, otherwise this call is restricted
to the super-user.
If the process number is 0, the signal is sent to all other processes which have the same controlling type-
writer and user ID.
In no case is it possible for a process to kill itself.
SEE ALSO
signal (II), kill (I)
DIAGNOSTICS
The error bit (c-bit) is set if the process does not have the same effective user ID and the user is not super-
user, or if the process does not exist. From C, −1 is returned.

-1-
-

LINK ( II ) 8/5/73 LINK ( II )

NAME
link − link to a file
SYNOPSIS
(link = 9.)
sys link; name1; name2
link(name1, name2)
char *name1, *name2;
DESCRIPTION
A link to name1 is created; the link has the name name2. Either name may be an arbitrary path name.
SEE ALSO
link (I), unlink (II)
DIAGNOSTICS
The error bit (c-bit) is set when name1 cannot be found; when name2 already exists; when the directory of
name2 cannot be written; when an attempt is made to link to a directory by a user other than the super-user;
when an attempt is made to link to a file on another file system; when more than 127 links are made. From
C, a −1 return indicates an error, a 0 return indicates success.

-1-
-

MKNOD ( II ) 8/5/73 MKNOD ( II )

NAME
mknod − make a directory or a special file
SYNOPSIS
(mknod = 14.; not in assembler)
sys mknod; name; mode; addr
mknod(name, mode, addr)
char *name;
DESCRIPTION
Mknod creates a new file whose name is the null-terminated string pointed to by name. The mode of the
new file (including directory and special file bits) is initialized from mode. The first physical address of the
file is initialized from addr. Note that in the case of a directory, addr should be zero. In the case of a spe-
cial file, addr specifies which special file.
Mknod may be invoked only by the super-user.
SEE ALSO
mkdir (I), mknod (VIII), fs (V)
DIAGNOSTICS
Error bit (c-bit) is set if the file already exists or if the user is not the super-user. From C, a −1 value indi-
cates an error.

-1-
-

MOUNT ( II ) 5/15/74 MOUNT ( II )

NAME
mount − mount file system
SYNOPSIS
(mount = 21.)
sys mount; special; name; rwflag
mount(special, name, rwflag)
char *special, *name;
DESCRIPTION
Mount announces to the system that a removable file system has been mounted on the block-structured spe-
cial file special; from now on, references to file name will refer to the root file on the newly mounted file
system. Special and name are pointers to null-terminated strings containing the appropriate path names.
Name must exist already. Its old contents are inaccessible while the file system is mounted.
The rwflag argument determines whether the file system can be written on; if it is 0 writing is allowed, if
non-zero no writing is done. Physically write-protected and magnetic tape file systems must be mounted
read-only or errors will occur when access times are updated, whether or not any explicit write is attempt-
ed.
SEE ALSO
mount (VIII), umount (II)
DIAGNOSTICS
Error bit (c-bit) set if: special is inaccessible or not an appropriate file; name does not exist; special is al-
ready mounted; name is in use; there are already too many file systems mounted.
BUGS
−

-1-
-

NICE ( II ) 8/5/73 NICE ( II )

NAME
nice − set program priority
SYNOPSIS
(nice = 34.)
(priority in r0)
sys nice
nice(priority)
DESCRIPTION
The scheduling priority of the process is changed to the argument. Positive priorities get less service than
normal; 0 is default. Only the super-user may specify a negative priority. The valid range of priority is 20
to −220. The value of 4 is recommended to users who wish to execute long-running programs without flak
from the administration.
The effect of this call is passed to a child process by the fork system call. The effect can be cancelled by
another call to nice with a priority of 0.
The actual running priority of a process is the priority argument plus a number that ranges from 100 to 119
depending on the cpu usage of the process.
SEE ALSO
nice (I)
DIAGNOSTICS
The error bit (c-bit) is set if the user requests a priority outside the range of 0 to 20 and is not the super-us-
er.

-1-
-

OPEN ( II ) 8/5/73 OPEN ( II )

NAME
open − open for reading or writing
SYNOPSIS
(open = 5.)
sys open; name; mode
(file descriptor in r0)
open(name, mode)
char *name;
DESCRIPTION
Open opens the file name for reading (if mode is 0), writing (if mode is 1) or for both reading and writing
(if mode is 2). Name is the address of a string of ASCII characters representing a path name, terminated by
a null character.
The returned file descriptor should be saved for subsequent calls to read, write, and close.
SEE ALSO
creat (II), read (II), write (II), close (II)
DIAGNOSTICS
The error bit (c-bit) is set if the file does not exist, if one of the necessary directories does not exist or is un-
readable, if the file is not readable (resp. writable), or if too many files are open. From C, a −1 value is re-
turned on an error.

-1-
-

PIPE ( II ) 8/5/73 PIPE ( II )

NAME
pipe − create an interprocess channel
SYNOPSIS
(pipe = 42.)
sys pipe
(read file descriptor in r0)
(write file descriptor in r1)
pipe(fildes)
int fildes[2];
DESCRIPTION
The pipe system call creates an I/O mechanism called a pipe. The file descriptors returned can be used in
read and write operations. When the pipe is written using the descriptor returned in r1 (resp. fildes[1]), up
to 4096 bytes of data are buffered before the writing process is suspended. A read using the descriptor re-
turned in r0 (resp. fildes[0]) will pick up the data.
It is assumed that after the pipe has been set up, two (or more) cooperating processes (created by subse-
quent fork calls) will pass data through the pipe with read and write calls.
The Shell has a syntax to set up a linear array of processes connected by pipes.
Read calls on an empty pipe (no buffered data) with only one end (all write file descriptors closed) return an
end-of-file. Write calls under similar conditions generate a fatal signal (signal (II)); if the signal is ignored,
an error is returned on the write.
SEE ALSO
sh (I), read (II), write (II), fork (II)
DIAGNOSTICS
The error bit (c-bit) is set if too many files are already open. From C, a −1 returned value indicates an error.
A signal is generated if a write on a pipe with only one end is attempted.
BUGS

-1-
-

PROFIL ( II ) 5/15/74 PROFIL ( II )

NAME
profil − execution time profile
SYNOPSIS
(profil = 44.; not in assembler)
sys profil; buff; bufsiz; offset; scale
profil(buff, bufsiz, offset, scale)
char buff[ ];
int bufsiz, offset, scale;
DESCRIPTION
Buff points to an area of core whose length (in bytes) is given by bufsiz. After this call, the user’s program
counter (pc) is examined each clock tick (60th second); offset is subtracted from it, and the result multiplied
by scale. If the resulting number corresponds to a word inside buff, that word is incremented.
The scale is interpreted as an unsigned, fixed-point fraction with binary point at the left: 177777(8) gives a
1-1 mapping of pc’s to words in buff; 77777(8) maps each pair of instruction words together. 2(8) maps all
instructions onto the beginning of buff (producing a non-interrupting core clock).
Profiling is turned off by giving a scale of 0 or 1. It is rendered ineffective by giving a bufsiz of 0. Profiling
is also turned off when an exec is executed but remains on in child and parent both after a fork.
SEE ALSO
monitor (III), prof (I)
DIAGNOSTICS
−

-1-
-

PTRACE ( II ) 1/25/75 PTRACE ( II )

NAME
ptrace − process trace
SYNOPSIS
(ptrace = 26.; not in assembler)
(data in r0)
sys ptrace; pid; addr; request
(value in r0)
ptrace(request, pid, addr, data);
DESCRIPTION
Ptrace provides a means by which a parent process may control the execution of a child process, and exam-
ine and change its core image. Its primary use is for the implementation of breakpoint debugging, but it
should be adaptable for simulation of non-UNIX environments. There are four arguments whose interpre-
tation depends on a request argument. Generally, pid is the process ID of the traced process, which must be
a child (no more distant descendant) of the tracing process. A process being traced behaves normally until
it encounters some signal whether internally generated like ‘‘illegal instruction’’ or externally generated
like ‘‘interrupt.’’ See signal (II) for the list. Then the traced process enters a stopped state and its parent is
notified via wait (II). When the child is in the stopped state, its core image can be examined and modified
using ptrace. If desired, another ptrace request can then cause the child either to terminate or to continue,
possibly ignoring the signal.
The value of the request argument determines the precise action of the call:
0 This request is the only one used by the child process; it declares that the process is to be traced by its
parent. All the other arguments are ignored. Peculiar results will ensue if the parent does not expect
to trace the child.
1,2 The word in the child process’s address space at addr is returned (in r0). Request 1 indicates the data
space (normally used); 2 indicates the instruction space (when I and D space are separated). addr
must be even. The child must be stopped. The input data is ignored.
3 The word of the system’s per-process data area corresponding to addr is returned. Addr must be even
and less than 512. This space contains the registers and other information about the process; its layout
corresponds to the user structure in the system.
4,5 The given data is written at the word in the process’s address space corresponding to addr, which must
be even. No useful value is returned. Request 4 specifies data space (normally used), 5 specifies in-
struction space. Attempts to write in pure procedure result in termination of the child, instead of going
through or causing an error for the parent.
6 The process’s system data is written, as it is read with request 3. Only a few locations can be written
in this way: the general registers, the floating point status and registers, and certain bits of the proces-
sor status word.
7 The data argument is taken as a signal number and the child’s execution continues as if it had incurred
that signal. Normally the signal number will be either 0 to indicate that the signal which caused the
stop should be ignored, or that value fetched out of the process’s image indicating which signal caused
the stop.
8 The traced process terminates.
As indicated, these calls (except for request 0) can be used only when the subject process has stopped. The
wait call is used to determine when a process stops; in such a case the ‘‘termination’’ status returned by
wait has the value 0177 to indicate stoppage rather than genuine termination.
To forestall possible fraud, ptrace inhibits the set-user-id facility on subsequent exec (II)
calls.
SEE ALSO
wait (II), signal (II), cdb (I)

-1-
-

PTRACE ( II ) 1/25/75 PTRACE ( II )

DIAGNOSTICS
From assembler, the c-bit (error bit) is set on errors; from C, −1 is returned and errno has the error code.
BUGS
The request 0 call should be able to specify signals which are to be treated normally and not cause a stop.
In this way, for example, programs with simulated floating point (which use ‘‘illegal instruction’’ signals at
a very high rate) could be efficiently debugged.
Also, it should be possible to stop a process on occurrence of a system call; in this way a completely con-
trolled environment could be provided.

-2-
-

READ ( II ) 8/5/73 READ ( II )

NAME
read − read from file
SYNOPSIS
(read = 3.)
(file descriptor in r0)
sys read; buffer; nbytes
read(fildes, buffer, nbytes)
char *buffer;
DESCRIPTION
A file descriptor is a word returned from a successful open, creat, dup, or pipe call. Buffer is the location of
nbytes contiguous bytes into which the input will be placed. It is not guaranteed that all nbytes bytes will
be read; for example if the file refers to a typewriter at most one line will be returned. In any event the
number of characters read is returned (in r0).
If the returned value is 0, then end-of-file has been reached.
SEE ALSO
open (II), creat (II), dup (II), pipe (II)
DIAGNOSTICS
As mentioned, 0 is returned when the end of the file has been reached. If the read was otherwise unsuc-
cessful the error bit (c-bit) is set. Many conditions can generate an error: physical I/O errors, bad buffer ad-
dress, preposterous nbytes, file descriptor not that of an input file. From C, a −1 return indicates the error.

-1-
-

SEEK ( II ) 8/5/73 SEEK ( II )

NAME
seek − move read/write pointer
SYNOPSIS
(seek = 19.)
(file descriptor in r0)
sys seek; offset; ptrname
seek(fildes, offset, ptrname)
DESCRIPTION
The file descriptor refers to a file open for reading or writing. The read (resp. write) pointer for the file is
set as follows:
if ptrname is 0, the pointer is set to offset.
if ptrname is 1, the pointer is set to its current location plus offset.
if ptrname is 2, the pointer is set to the size of the file plus offset.
if ptrname is 3, 4 or 5, the meaning is as above for 0, 1 and 2 except that the offset is multiplied by 512.
If ptrname is 0 or 3, offset is unsigned, otherwise it is signed.
SEE ALSO
open (II), creat (II)
DIAGNOSTICS
The error bit (c-bit) is set for an undefined file descriptor. From C, a −1 return indicates an error.

-1-
-

SETGID ( II ) 8/5/73 SETGID ( II )

NAME
setgid − set process group ID
SYNOPSIS
(setgid = 46.; not in assembler)
(group ID in r0)
sys setgid
setgid(gid)
DESCRIPTION
The group ID of the current process is set to the argument. Both the effective and the real group ID are set.
This call is only permitted to the super-user or if the argument is the real group ID.
SEE ALSO
getgid (II)
DIAGNOSTICS
Error bit (c-bit) is set as indicated; from C, a −1 value is returned.

-1-
-

SETUID ( II ) 8/5/73 SETUID ( II )

NAME
setuid − set process user ID
SYNOPSIS
(setuid = 23.)
(user ID in r0)
sys setuid
setuid(uid)
DESCRIPTION
The user ID of the current process is set to the argument. Both the effective and the real user ID are set.
This call is only permitted to the super-user or if the argument is the real user ID.
SEE ALSO
getuid (II)
DIAGNOSTICS
Error bit (c-bit) is set as indicated; from C, a −1 value is returned.

-1-
-

SIGNAL ( II ) 8/5/73 SIGNAL ( II )

NAME
signal − catch or ignore signals
SYNOPSIS
(signal = 48.)
sys signal; sig; label
(old value in r0)
signal(sig, func)
int (*func)( );
DESCRIPTION
A signal is generated by some abnormal event, initiated either by user at a typewriter (quit, interrupt), by a
program error (bus error, etc.), or by request of another program (kill). Normally all signals cause termina-
tion of the receiving process, but this call allows them either to be ignored or to cause an interrupt to a spec-
ified location. Here is the list of signals:
1 hangup
2 interrupt
3* quit
4* illegal instruction (not reset when caught)
5* trace trap (not reset when caught)
6* IOT instruction
7* EMT instruction
8* floating point exception
9 kill (cannot be caught or ignored)
10* bus error
11* segmentation violation
12* bad argument to system call
13 write on a pipe with no one to read it
In the assembler call, if label is 0, the process is terminated when the signal occurs; this is the default ac-
tion. If label is odd, the signal is ignored. Any other even label specifies an address in the process where
an interrupt is simulated. An RTI or RTT instruction will return from the interrupt. Except as indicated, a
signal is reset to 0 after being caught. Thus if it is desired to catch every such signal, the catching routine
must issue another signal call.
In C, if func is 0, the default action for signal sig (termination) is reinstated. If func is 1, the signal is ig-
nored. If func is non-zero and even, it is assumed to be the address of a function entry point. When the sig-
nal occurs, the function will be called. A return from the function will continue the process at the point it
was interrupted. As in the assembler call, signal must in general be called again to catch subsequent sig-
nals.
When a caught signal occurs during certain system calls, the call terminates prematurely. In particular this
can occur during a read or write on a slow device (like a typewriter; but not a file); and during or wait.
When such a signal occurs, the saved user status is arranged in such a way that when return from the signal-
catching takes place, it will appear that the system call returned a characteristic error status. The user’s pro-
gram may then, if it wishes, re-execute the call.
The starred signals in the list above cause a core image if not caught or ignored.
The value of the call is the old action defined for the signal.
After a fork (II) the child inherits all signals. Exec (II) resets all caught signals to default action.
SEE ALSO
kill (I), kill (II), ptrace (II), reset (III)
DIAGNOSTICS
The error bit (c-bit) is set if the given signal is out of range. In C, a −1 indicates an error; 0 indicates suc-
cess.

-1-
-

SIGNAL ( II ) 8/5/73 SIGNAL ( II )

BUGS

-2-
-

SLEEP ( II ) 8/5/73 SLEEP ( II )

NAME
sleep − stop execution for interval
SYNOPSIS
(sleep = 35.; not in assembler)
(seconds in r0)
sys sleep
sleep(seconds)
DESCRIPTION
The current process is suspended from execution for the number of seconds specified by the argument.
SEE ALSO
sleep (I)
DIAGNOSTICS
−

-1-
-

STAT ( II ) 8/5/73 STAT ( II )

NAME
stat − get file status
SYNOPSIS
(stat = 18.)
sys stat; name; buf
stat(name, buf)
char *name;
struct inode *buf;
DESCRIPTION
Name points to a null-terminated string naming a file; buf is the address of a 36(10) byte buffer into which
information is placed concerning the file. It is unnecessary to have any permissions at all with respect to
the file, but all directories leading to the file must be readable. After stat, buf has the following structure
(starting offset given in bytes):
struct inode {
char minor; /* +0: minor device of i-node */
char major; /* +1: major device */
int inumber; /* +2 */
int flags; /* +4: see below */
char nlinks; /* +6: number of links to file */
char uid; /* +7: user ID of owner */
char gid; /* +8: group ID of owner */
char size0; /* +9: high byte of 24-bit size */
int size1; /* +10: low word of 24-bit size */
int addr[8]; /* +12: block numbers or device number */
int actime[2]; /* +28: time of last access */
int modtime[2]; /* +32: time of last modification */
};
The flags are as follows:
100000 i-node is allocated
060000 2-bit file type:
000000 plain file
040000 directory
020000 character-type special file
060000 block-type special file.
010000 large file
004000 set user-ID on execution
002000 set group-ID on execution
001000 save text image after execution
000400 read (owner)
000200 write (owner)
000100 execute (owner)
000070 read, write, execute (group)
000007 read, write, execute (others)
SEE ALSO
ls (I), fstat (II), fs (V)
DIAGNOSTICS
Error bit (c-bit) is set if the file cannot be found. From C, a −1 return indicates an error.

-1-
-

STIME ( II ) 8/5/73 STIME ( II )

NAME
stime − set time
SYNOPSIS
(stime = 25.)
(time in r0-r1)
sys stime
stime(tbuf)
int tbuf[2];
DESCRIPTION
Stime sets the system’s idea of the time and date. Time is measured in seconds from 0000 GMT Jan 1
1970. Only the super-user may use this call.
SEE ALSO
date (I), time (II), ctime (III)
DIAGNOSTICS
Error bit (c-bit) set if user is not the super-user.

-1-
-

STTY ( II ) 12/15/74 STTY ( II )

NAME
stty − set mode of typewriter
SYNOPSIS
(stty = 31.)
(file descriptor in r0)
sys stty; arg
...
arg: .byte ispeed, ospeed; .byte erase, kill; mode
stty(fildes, arg)
struct {
char ispeed, ospeed;
char erase, kill;
int mode;
} *arg;

DESCRIPTION
Stty sets mode bits and character speeds for the typewriter whose file descriptor is passed in r0 (resp. is the
first argument to the call). First, the system delays until the typewriter is quiescent. The input and output
speeds are set from the first two bytes of the argument structure as indicated by the following table, which
corresponds to the speeds supported by the DH-11 interface. If DC-11, DL-11 or KL-11 interfaces are
used, impossible speed changes are ignored.
0 (hang up dataphone)
1 50 baud
2 75 baud
3 110 baud
4 134.5 baud
5 150 baud
6 200 baud
7 300 baud
8 600 baud
9 1200 baud
10 1800 baud
11 2400 baud
12 4800 baud
13 9600 baud
14 External A
15 External B
In the current configuration, only 110, 150 and 300 baud are really supported on dial-up lines, in that the
code conversion and line control required for IBM 2741’s (134.5 baud) must be implemented by the user’s
program, and the half-duplex line discipline required for the 202 dataset (1200 baud) is not supplied.
The next two characters of the argument structure specify the erase and kill characters respectively. (De-
faults are # and @.)
The mode contains several bits which determine the system’s treatment of the typewriter:
100000 Select one of two algorithms for backspace delays
040000 Select one of two algorithms for form-feed and vertical-tab delays
030000 Select one of four algorithms for carriage-return delays
006000 Select one of four algorithms for tab delays
001400 Select one of four algorithms for new-line delays
000200 even parity allowed on input (e. g. for M37s)
000100 odd parity allowed on input
000040 raw mode: wake up on all characters
000020 map CR into LF; echo LF or CR as CR-LF
000010 echo (full duplex)

-1-
-

STTY ( II ) 12/15/74 STTY ( II )

000004 map upper case to lower on input (e. g. M33)

000002 echo and print tabs as spaces
000001 hang up (remove ‘data terminal ready,’ lead CD) after last close
The delay bits specify how long transmission stops to allow for mechanical or other movement when cer-
tain characters are sent to the terminal. In all cases a value of 0 indicates no delay.
Backspace delays are currently ignored but will be used for Terminet 300’s.
If a form-feed/vertical tab delay is specified, it lasts for about 2 seconds.
Carriage-return delay type 1 lasts about .08 seconds and is suitable for the Terminet 300. Delay type 2 lasts
about .16 seconds and is suitable for the VT05 and the TI 700. Delay type 3 is unimplemented and is 0.
New-line delay type 1 is dependent on the current column and is tuned for Teletype model 37’s. Type 2 is
useful for the VT05 and is about .10 seconds. Type 3 is unimplemented and is 0.
Tab delay type 1 is dependent on the amount of movement and is tuned to the Teletype model 37. Other
types are unimplemented and are 0.
Characters with the wrong parity, as determined by bits 200 and 100, are ignored.
In raw mode, every character is passed immediately to the program without waiting until a full line has
been typed. No erase or kill processing is done; the end-of-file character (EOT), the interrupt character
(DEL) and the quit character (FS) are not treated specially.
Mode 020 causes input carriage returns to be turned into new-lines; input of either CR or LF causes LF-CR
both to be echoed (used for GE TermiNet 300’s and other terminals without the newline function).
The hangup mode 01 causes the line to be disconnected when the last process with the line open closes it or
terminates. It is useful when a port is to be used for some special purpose; for example, if it is associated
with an ACU used to place outgoing calls.
This system call is also used with certain special files other than typewriters, but since none of them are part
of the standard system the specifications will not be given.
SEE ALSO
stty (I), gtty (II)
DIAGNOSTICS
The error bit (c-bit) is set if the file descriptor does not refer to a typewriter. From C, a negative value indi-
cates an error.

-2-
-

SYNC ( II ) 8/5/73 SYNC ( II )

NAME
sync − update super-block
SYNOPSIS
(sync = 36.; not in assembler)
sys sync
DESCRIPTION
Sync causes all information in core memory that should be on disk to be written out. This includes modi-
fied super blocks, modified i-nodes, and delayed block I/O.
It should be used by programs which examine a file system, for example icheck, df, etc. It is mandatory be-
fore a boot.
SEE ALSO
sync (VIII), update (VIII)
DIAGNOSTICS
−

-1-
-

TIME ( II ) 8/5/73 TIME ( II )

NAME
time − get date and time
SYNOPSIS
(time = 13.)
sys time
time(tvec)
int tvec[2];
DESCRIPTION
Time returns the time since 00:00:00 GMT, Jan. 1, 1970, measured in seconds. From as, the high order
word is in the r0 register and the low order is in r1. From C, the user-supplied vector is filled in.
SEE ALSO
date (I), stime (II), ctime (III)
DIAGNOSTICS
−

-1-
-

TIMES ( II ) 8/5/73 TIMES ( II )

NAME
times − get process times
SYNOPSIS
(times = 43.; not in assembler)
sys times; buffer
times(buffer)
struct tbuffer *buffer;
DESCRIPTION
Times returns time-accounting information for the current process and for the terminated child processes of
the current process. All times are in 1/60 seconds.
After the call, the buffer will appear as follows:
struct tbuffer {
int proc user time;
int proc system time;
int child user time[2];
int child system time[2];
};
The children times are the sum of the children’s process times and their children’s times.
SEE ALSO
time (I)
DIAGNOSTICS
−
BUGS
The process times should be 32 bits as well.

-1-
-

UMOUNT ( II ) 8/5/73 UMOUNT ( II )

NAME
umount − dismount file system
SYNOPSIS
(umount = 22.)
sys umount; special
DESCRIPTION
Umount announces to the system that special file special is no longer to contain a removable file system.
The file associated with the special file reverts to its ordinary interpretation; see mount (II).

SEE ALSO
umount (VIII), mount (II)
DIAGNOSTICS
Error bit (c-bit) set if no file system was mounted on the special file or if there are still active files on the
mounted file system.

-1-
-

UNLINK ( II ) 8/5/73 UNLINK ( II )

NAME
unlink − remove directory entry
SYNOPSIS
(unlink = 10.)
sys unlink; name
unlink(name)
char *name;
DESCRIPTION
Name points to a null-terminated string. Unlink removes the entry for the file pointed to by name from its
directory. If this entry was the last link to the file, the contents of the file are freed and the file is destroyed.
If, however, the file was open in any process, the actual destruction is delayed until it is closed, even though
the directory entry has disappeared.
SEE ALSO
rm (I), rmdir (I), link (II)
DIAGNOSTICS
The error bit (c-bit) is set to indicate that the file does not exist or that its directory cannot be written. Write
permission is not required on the file itself. It is also illegal to unlink a directory (except for the super-us-
er). From C, a −1 return indicates an error.

-1-
-

WAIT ( II ) 2/9/75 WAIT ( II )

NAME
wait − wait for process to terminate
SYNOPSIS
(wait = 7.)
sys wait
(process ID in r0)
(status in r1)
wait(status)
int *status;
DESCRIPTION
Wait causes its caller to delay until one of its child processes terminates. If any child has died since the last
wait, return is immediate; if there are no children, return is immediate with the error bit set (resp. with a
value of −1 returned). The normal return yields the process ID of the terminated child (in r0). In the case
of several children several wait calls are needed to learn of all the deaths.
If no error is indicated on return, the r1 high byte (resp. the high byte stored into status ) contains the low
byte of the child process r0 (resp. the argument of exit ) when it terminated. The r1 (resp. status ) low byte
contains the termination status of the process. See signal (II) for a list of termination statuses (signals); 0
status indicates normal termination. A special status (0177) is returned for a stopped process which has not
terminated and can be restarted. See ptrace (II). If the 0200 bit of the termination status is set, a core im-
age of the process was produced by the system.
If the parent process terminates without waiting on its children, the initialization process (process ID = 1)
inherits the children.
SEE ALSO
exit (II), fork (II), signal (II)
DIAGNOSTICS
The error bit (c-bit) is set if there are no children not previously waited for. From C, a returned value of −1
indicates an error.

-1-
-

WRITE ( II ) 8/5/73 WRITE ( II )

NAME
write − write on a file
SYNOPSIS
(write = 4.)
(file descriptor in r0)
sys write; buffer; nbytes
write(fildes, buffer, nbytes)
char *buffer;
DESCRIPTION
A file descriptor is a word returned from a successful open, creat, dup, or pipe call.
Buffer is the address of nbytes contiguous bytes which are written on the output file. The number of charac-
ters actually written is returned (in r0). It should be regarded as an error if this is not the same as requested.
Writes which are multiples of 512 characters long and begin on a 512-byte boundary in the file are more ef-
ficient than any others.
SEE ALSO
creat (II), open (II), pipe (II)
DIAGNOSTICS
The error bit (c-bit) is set on an error: bad descriptor, buffer address, or count; physical I/O errors. From C,
a returned value of −1 indicates an error.

-1-
-

ABORT ( III ) 4/10/75 ABORT ( III )

NAME
abort − generate an IOT fault
SYNOPSIS
abort()
DESCRIPTION
Abort executes the IOT instruction. This is usually considered a program fault by the system and results in
termination with a core dump. It is used to generate a core image for debugging.
SEE ALSO
db (I), cdb (I), signal (II)
DIAGNOSTICS
usually ‘‘IOT trap -- core dumped’’ from the Shell.
BUGS

-1-
-

ABS ( III ) 2/9/75 ABS ( III )

NAME
abs, fabs − absolute value
SYNOPSIS
abs(i)
int i;
double fabs(x)
double x;

DESCRIPTION
Abs returns the absolute value of its integer operand; fabs is the double version.

-1-
-

ALLOC ( III ) 3/1/74 ALLOC ( III )

NAME
alloc, free − core allocator
SYNOPSIS
char *alloc(size)
free(ptr)
char *ptr;
DESCRIPTION
Alloc and free provide a simple general-purpose core management package. Alloc is given a size in bytes;
it returns a pointer to an area at least that size which is even and hence can hold an object of any type. The
argument to free is a pointer to an area previously allocated by alloc; this space is made available for fur-
ther allocation.
Needless to say, grave disorder will result if the space assigned by alloc is overrun or if some random num-
ber is handed to free.
The routine uses a first-fit algorithm which coalesces blocks being freed with other blocks already free. It
calls sbrk (see break (II)) to get more core from the system when there is no suitable space already free.
DIAGNOSTICS
Returns −1 if there is no available core.
BUGS
Allocated memory contains garbage instead of being cleared.

-1-
-

ATAN ( III ) 4/30/73 ATAN ( III )

NAME
atan, atan2 − arc tangent function
SYNOPSIS
jsr pc,atan[2]
double atan(x)
double x;
double atan2(x, y)
double x, y;
DESCRIPTION
The atan entry returns the arc tangent of fr0 in fr0; from C, the arc tangent of x is returned. The range is
−π/2 to π/2. The atan2 entry returns the arc tangent of fr0/fr1 in fr0; from C, the arc tangent of x/y is re-
turned. The range is −π to π.
DIAGNOSTIC
There is no error return.
BUGS

-1-
-

ATOF ( III ) 4/30/73 ATOF ( III )

NAME
atof − convert ASCII to floating
SYNOPSIS
double atof(nptr)
char *nptr;
DESCRIPTION
Atof converts a string to a floating number. Nptr should point to a string containing the number; the first un-
recognized character ends the number.
The only numbers recognized are: an optional minus sign followed by a string of digits optionally contain-
ing one decimal point, then followed optionally by the letter e followed by a signed integer.
DIAGNOSTICS
There are none; overflow results in a very large number and garbage characters terminate the scan.
BUGS
The routine should accept initial +, initial blanks, and E for e. Overflow should be signalled.

-1-
-

ATOI ( III ) 2/8/75 ATOI ( III )

NAME
atoi − convert ASCII to integer
SYNOPSIS
atoi(nptr)
char *nptr;
DESCRIPTION
Atoi converts the string pointed to by nptr to an integer. The string can contain leading blanks or tabs, an
optional ‘−’, and then an unbroken string of digits. Conversion stops at the first non-digit.
SEE ALSO
atof (III)
BUGS
There is no provision for overflow.

-1-
-

CRYPT ( III ) 4/30/73 CRYPT ( III )

NAME
crypt − password encoding
SYNOPSIS
mov $key,r0
jsr pc,crypt
char *crypt(key)
char *key;
DESCRIPTION
On entry, r0 points to a string of characters terminated by an ASCII NUL. The routine performs an opera-
tion on the key which is difficult to invert (i.e. encrypts it) and leaves the resulting eight bytes of ASCII al-
phanumerics in a global cell called ‘‘word’’.
From C, the key argument is a string and the value returned is a pointer to the eight-character result.
This routine is used to encrypt all passwords.
SEE ALSO
passwd(I), passwd(V), login(I)
BUGS
Short or otherwise simple passwords can be decrypted easily by exhaustive search. Six characters of gib-
berish is reasonably safe.

-1-
-

CTIME ( III ) 10/15/73 CTIME ( III )

NAME
ctime, localtime, gmtime − convert date and time to ASCII
SYNOPSIS
char *ctime(tvec)
int tvec[2];
[from Fortran]
double precision ctime
... = ctime(dummy)
int *localtime(tvec)
int tvec[2];
int *gmtime(tvec)
int tvec[2];
DESCRIPTION
Ctime converts a time in the vector tvec such as returned by time (II) into ASCII and returns a pointer to a
character string in the form
Sun Sep 16 01:03:52 1973\n\0
All the fields have constant width.
The localtime and gmtime entries return pointers to integer vectors containing the broken-down time. Lo-
caltime corrects for the time zone and possible daylight savings time; gmtime converts directly to GMT,
which is the time UNIX uses. The value is a pointer to an array whose components are
0 seconds
1 minutes
2 hours
3 day of the month (1-31)
4 month (0-11)
5 year − 1900
6 day of the week (Sunday = 0)
7 day of the year (0-365)
8 Daylight Saving Time flag if non-zero
The external variable timezone contains the difference, in seconds, between GMT and local standard time
(in EST, is 5*60*60); the external variable daylight is non-zero iff the standard U.S.A. Daylight Savings
Time conversion should be applied. The program knows about the peculiarities of this conversion in 1974
and 1975; if necessary, a table for these years can be extended.
A routine named ctime is also available from Fortran. Actually it more resembles the time (II) system entry
in that it returns the number of seconds since the epoch 0000 GMT Jan. 1, 1970 (as a floating-point num-
ber).
SEE ALSO
time(II)
BUGS

-1-
-

ECVT ( III ) 4/30/73 ECVT ( III )

NAME
ecvt, fcvt − output conversion
SYNOPSIS
jsr pc,ecvt
jsr pc,fcvt
char *ecvt(value, ndigit, decpt, sign)
double value;
int ndigit, *decpt, *sign;
char *fcvt(value, ndigit, decpt, sign)
...
DESCRIPTION
Ecvt is called with a floating point number in fr0.
On exit, the number has been converted into a string of ascii digits in a buffer pointed to by r0. The number
of digits produced is controlled by a global variable ndigits.
Moreover, the position of the decimal point is contained in r2: r2=0 means the d.p. is at the left hand end of
the string of digits; r2>0 means the d.p. is within or to the right of the string.
The sign of the number is indicated by r1 (0 for +; 1 for −).
The low order digit has suffered decimal rounding (i. e. may have been carried into).
From C, the value is converted and a pointer to a null-terminated string of ndigit digits is returned. The po-
sition of the decimal point is stored indirectly through decpt (negative means to the left of the returned dig-
its). If the sign of the result is negative, the word pointed to by sign is non-zero, otherwise it is zero.
Fcvt is identical to ecvt, except that the correct digit been rounded for F-style output of the number of digits
specified by ndigits.
SEE ALSO
printf (III)
BUGS

-1-
-

END ( III ) 4/28/75 END ( III )

NAME
end, etext, edata − last locations in program
SYNOPSIS
extern end;
extern etext;
extern edata;
DESCRIPTION
These names refer neither to routines nor to locations with interesting contents. Instead, their addresses co-
incide with the first address above the program text region (etext), above the initialized data region (edata),
or uninitialized data region (end). The last is the same as the program break. Values are given to these
symbols by the link editor ld (I) when, and only when, they are referred to but not defined in the set of pro-
grams loaded.
The usage of these symbols is rather specialized, but one plausible possibility is
extern end;
...
... = brk(&end+...);
(see break (II)). The problem with this is that it ignores any other subroutines which may want to extend
core for their purposes; these include sbrk (see break (II)), alloc (III), and also secret subroutines invoked
by the profile (−p) option of cc. Of course it was for the benefit of such systems that the symbols were in-
vented, and user programs, unless they are in firm control of their environment, are wise not to refer to the
absolute symbols directly.
One technique sometimes useful is to call sbrk(0), which returns the value of the current program break, in-
stead of referring to &end, which yields the program break at the instant execution started.
These symbols are accessible from assembly language if it is remembered that they should be prefixed by
‘ ’
SEE ALSO
break (II), alloc (III)
BUGS

-1-
-

EXP ( III ) 4/30/73 EXP ( III )

NAME
exp − exponential function
SYNOPSIS
jsr pc,exp
double exp(x)
double x;
DESCRIPTION
The exponential of fr0 is returned in fr0. From C, the exponential of x is returned.
DIAGNOSTICS
If the result is not representable, the c-bit is set and the largest positive number is returned. From C, no di-
agnostic is available.
Zero is returned if the result would underflow.
BUGS

-1-
-

FLOOR ( III ) 5/15/74 FLOOR ( III )

NAME
floor, ceil − floor and ceiling functions
SYNOPSIS
double floor(x)
double x;
double ceil(x)
double x;
DESCRIPTION
The floor function returns the largest integer (as a double precision number) not greater than x.
The ceil function returns the smallest integer not less than x.
BUGS

-1-
-

FMOD ( III ) 2/13/75 FMOD ( III )

NAME
fmod − floating modulo function
SYNOPSIS
double fmod(x, y)
double x, y;
DESCRIPTION
Fmod returns the number f such that x = iy + f, i is an integer, and 0 ≤ f < y.
BUGS

-1-
-

FPTRAP ( III ) 11/18/73 FPTRAP ( III )

NAME
fptrap − floating point interpreter
SYNOPSIS
sys signal; 4; fptrap
DESCRIPTION
Fptrap is a simulator of the 11/45 FP11-B floating point unit. It works by intercepting illegal instruction
traps and decoding and executing the floating point operation codes.
FILES
In systems with real floating point, there is a fake routine in /lib/liba.a with this name; when simulation is
desired, the real version should be put in liba.a
DIAGNOSTICS
A break point trap is given when a real illegal instruction trap occurs.
SEE ALSO
signal (II), cc (I) (‘−f’ option)
BUGS
Rounding mode is not interpreted. It’s slow.

-1-
-

GAMMA ( III ) 5/15/74 GAMMA ( III )

NAME
gamma − log gamma function
SYNOPSIS
jsr pc,gamma
double gamma(x)
double x;
DESCRIPTION
If x is passed (in fr0) gamma returns ln | Γ( x ) | (in fr0). The sign of Γ( x ) is returned in the external integer
signgam. The following C program might be used to calculate Γ:
y = gamma(x);
if (y > 88.)
error( );
y = exp(y);
if(signgam)
y = −y;
DIAGNOSTICS
The c-bit is set on negative integral arguments and the maximum value is returned. There is no error return
for C programs.
BUGS
No error return from C.

-1-
-

GETARG ( III ) 11/24/73 GETARG ( III )

NAME
getarg, iargc − get command arguments from Fortran
SYNOPSIS
call getarg ( i, iarray [ , isize ] )
... = iargc(dummy)
DESCRIPTION
The getarg entry fills in iarray (which is considered to be integer) with the Hollerith string representing the
i th argument to the command in which it it is called. If no isize argument is specified, at least one blank is
placed after the argument, and the last word affected is blank padded. The user should make sure that the
array is big enough.
If the isize argument is given, the argument will be followed by blanks to fill up isize words, but even if the
argument is long no more than that many words will be filled in.
The blank-padded array is suitable for use as an argument to setfil (III).
The iargc entry returns the number of arguments to the command, counting the first (file-name) argument.
SEE ALSO
exec (II), setfil (III)
BUGS

-1-
-

GETC ( III ) 4/30/72 GETC ( III )

NAME
getc, getw, fopen − buffered input
SYNOPSIS
mov $filename,r0
jsr r5,fopen; iobuf
fopen(filename, iobuf)
char *filename;
struct buf *iobuf;
jsr r5,getc; iobuf
(character in r0)
getc(iobuf)
struct buf *iobuf;
jsr r5,getw; iobuf
(word in r0)
getw(iobuf)
struct buf *iobuf;
DESCRIPTION
These routines provide a buffered input facility. Iobuf is the address of a 518(10) byte buffer area whose
contents are maintained by these routines. Its structure is
struct buf {
int fildes; /* File descriptor */
int nleft; /* Chars left in buffer */
char *nextp; /* Ptr to next character */
char buff[512]; /* The buffer */
};
Fopen may be called initially to open the file. On return, the error bit (c-bit) is set if the open failed. If
fopen is never called, get will read from the standard input file. From C, the value is negative if the open
failed.
Getc returns the next byte from the file in r0. The error bit is set on end of file or a read error. From C, the
character is returned as an integer, without sign extension; it is −1 on end-of-file or error.
Getw returns the next word in r0. Getc and getw may be used alternately; there are no odd/even problems.
Getw is may be called from C; −1 is returned on end-of-file or error, but of course is also a legitimate value.
Iobuf must be provided by the user; it must be on a word boundary.
To reuse the same buffer for another file, it is sufficient to close the original file and call fopen again.
SEE ALSO
open (II), read (II), getchar (III), putc (III)
DIAGNOSTICS
c-bit set on EOF or error; from C, negative return indicates error or EOF. Moreover, errno is set by this
routine just as it is for a system call (see introduction (II)).
BUGS

-1-
-

GETCHAR ( III ) 4/7/73 GETCHAR ( III )

NAME
getchar − read character
SYNOPSIS
getchar( )
DESCRIPTION
Getchar provides the simplest means of reading characters from the standard input for C programs. It re-
turns successive characters until end-of-file, when it returns ‘‘\0’’.
Associated with this routine is an external variable called fin, which is a structure containing a buffer such
as described under getc (III).
Generally speaking, getchar should be used only for the simplest applications; getc is better when there are
multiple input files.
SEE ALSO
getc (III)
DIAGNOSTICS
Null character returned on EOF or error.
BUGS
−1 should be returned on EOF; null is a legitimate character.

-1-
-

GETPW ( III ) 4/7/73 GETPW ( III )

NAME
getpw − get name from UID
SYNOPSIS
getpw(uid, buf)
char *buf;
DESCRIPTION
Getpw searches the password file for the (numerical) uid, and fills in buf with the corresponding line; it re-
turns non-zero if uid could not be found. The line is null-terminated.
FILES
/etc/passwd
SEE ALSO
passwd (V)
DIAGNOSTICS
non-zero return on error.
BUGS

-1-
-

HMUL ( III ) 4/7/73 HMUL ( III )

NAME
hmul − high-order product
SYNOPSIS
hmul(x, y)
DESCRIPTION
Hmul returns the high-order 16 bits of the product of x and y. (The binary multiplication operator generates
the low-order 16 bits of a product.)
BUGS

-1-
-

IERROR ( III ) 10/29/73 IERROR ( III )

NAME
ierror − catch Fortran errors
SYNOPSIS
if ( ierror ( errno ) .ne. 0 ) goto label
DESCRIPTION
Ierror provides a way of detecting errors during the running of a Fortran program. Its argument is a run-
time error number such as enumerated in fc (I).
When ierror is called, it returns a 0 value; thus the goto statement in the synopsis is not executed. Howev-
er, the routine stores inside itself the call point and invocation level. If and when the indicated error occurs,
a return is simulated from ierror with a non-zero value; thus the goto (or other statement) is executed. It is
a ghastly error to call ierror from a subroutine which has already returned when the error occurs.
This routine is essentially tailored to catching end-of-file situations. Typically it is called just before the
start of the loop which reads the input file, and the goto jumps to a graceful termination of the program.
There is a limit of 5 on the number of different error numbers which can be caught.
SEE ALSO
fc (I)
BUGS
There is no way to ignore errors.

-1-
-

LDIV ( III ) 5/7/73 LDIV ( III )

NAME
ldiv, lrem − long division
SYNOPSIS
ldiv(hidividend, lodividend, divisor)
lrem(hidividend, lodividend, divisor)
DESCRIPTION
The concatenation of the signed 16-bit hidividend and the unsigned 16-bit lodividend is divided by divisor.
The 16-bit signed quotient is returned by ldiv and the 16-bit signed remainder is returned by lrem. Divide
check and erroneous results will occur unless the magnitude of the divisor is greater than that of the high-
order dividend.
An integer division of an unsigned dividend by a signed divisor may be accomplished by
quo = ldiv(0, dividend, divisor);
and similarly for the remainder operation.
Often both the quotient and the remainder are wanted. Therefore ldiv leaves a remainder in the external cell
ldivr.
BUGS
No divide check check.

-1-
-

LOCV ( III ) 3/9/74 LOCV ( III )

NAME
locv − long output conversion
SYNOPSIS
char *locv(hi, lo)
int hi, lo;
DESCRIPTION
Locv converts a signed double-precision integer, whose parts are passed as arguments, to the equivalent
ASCII character string and returns a pointer to that string.
BUGS
Since locv returns a pointer to a static buffer containing the converted result, it cannot be used twice in the
same expression; the second result overwrites the first.

-1-
-

LOG ( III ) 4/30/72 LOG ( III )

NAME
log − natural logarithm
SYNOPSIS
jsr pc,log
double log(x)
double x;
DESCRIPTION
The natural logarithm of fr0 is returned in fr0. From C, the natural logarithm of x is returned.
DIAGNOSTICS
The error bit (c-bit) is set if the input argument is less than or equal to zero and the result is a negative num-
ber very large in magnitude. From C, there is no error indication.
BUGS

-1-
-

MONITOR ( III ) 2/11/74 MONITOR ( III )

NAME
monitor − prepare execution profile
SYNOPSIS
monitor(lowpc, highpc, buffer, bufsize)
int lowpc( ), highpc( ), buffer[ ], bufsize;
DESCRIPTION
Monitor is an interface to the system’s profile entry (II). Lowpc and highpc are the names of two functions;
buffer is the address of a (user supplied) array of bufsize integers. Monitor arranges for the system to sam-
ple the user’s program counter periodically and record the execution histogram in the buffer. The lowest
address sampled is that of lowpc and the highest is just below highpc. For the results to be significant, es-
pecially where there are small, heavily used routines, it is suggested that the buffer be no more than a few
times smaller than the range of locations sampled.
To profile the entire program, it is sufficient to use
extern etext;
...
monitor(2, &etext, buf, bufsize);
Etext is a loader-defined symbol which lies just above all the program text.
To stop execution monitoring and write the results on the file mon.out, use
monitor(0);
Then, when the program exits, prof (I) can be used to examine the results.
It is seldom necessary to call this routine directly; the −p option of cc is simpler if one is satisfied with its
default profile range and resolution.
FILES
mon.out
SEE ALSO
prof (I), profil (II), cc (I)

-1-
-

NARGS ( III ) 5/10/73 NARGS ( III )

NAME
nargs − argument count
SYNOPSIS
nargs( )
DESCRIPTION
Nargs returns the number of actual parameters supplied by the caller of the routine which calls nargs.
The argument count is accurate only when none of the actual parameters is float or double. Such parame-
ters count as four arguments instead of one.
BUGS
As indicated. Also, this routine does not work (and cannot be made to work) in programs with separated I
and D space. Altogether it is best to avoid using this routine and depend, for example, on passing an ex-
plicit argument count.

-1-
-

NLIST ( III ) 6/12/72 NLIST ( III )

NAME
nlist − get entries from name list
SYNOPSIS
nlist(filename, nl)
char *filename;
struct {
char name[8];
int type;
int value;
} nl[ ];
DESCRIPTION
Nlist examines the name list in the given executable output file and selectively extracts a list of values. The
name list consists of a list of 8-character names (null padded) each followed by two words. The list is ter-
minated with a null name. Each name is looked up in the name list of the file. If the name is found, the
type and value of the name are placed in the two words following the name. If the name is not found, the
type entry is set to −1.
This subroutine is useful for examining the system name list kept in the file /unix. In this way programs
can obtain system addresses that are up to date.
SEE ALSO
a.out (V)
DIAGNOSTICS
All type entries are set to −1 if the file cannot be found or if it is not a valid namelist.
BUGS

-1-
-

PERROR ( III ) 11/5/73 PERROR ( III )

NAME
perror, sys errlist, sys nerr, errno − system error messages
SYNOPSIS
perror(s)
char *s;
int sys nerr;
char *sys errlist[];
int errno;
DESCRIPTION
Perror produces a short error message describing the last error encountered during a call to the system from
a C program. First the argument string s is printed, then a colon, then the message and a new-line. Most
usefully, the argument string is the name of the program which incurred the error. The error number is tak-
en from the external variable errno, which is set when errors occur but not cleared when non-erroneous
calls are made.
To simplify variant formatting of messages, the vector of message strings sys errlist is provided; errno can
be used as an index in this table to get the message string without the newline. Sys nerr is the largest mes-
sage number provided for in the table; it should be checked because new error codes may be added to the
system before they are added to the table.
SEE ALSO
Introduction to System Calls
BUGS

-1-
-

POW ( III ) 4/30/73 POW ( III )

NAME
pow − floating exponentiation
SYNOPSIS
movf x,fr0
movf y,fr1
jsr pc,pow
double pow(x,y)
double x, y;

DESCRIPTION
Pow returns the value of xy (in fr0). Pow(0.0, y) is 0 for any y. Pow(−x, y) returns a result only if y is an in-
teger.
SEE ALSO
exp (III), log (III)
DIAGNOSTICS
The carry bit is set on return in case of overflow, pow(0.0, 0.0), or pow(−x, y) for non-integral y. From C
there is no diagnostic.
BUGS

-1-
-

PRINTF ( III ) 9/17/73 PRINTF ( III )

NAME
printf − formatted print
SYNOPSIS
printf(format, arg1, ...);
char *format;
DESCRIPTION
Printf converts, formats, and prints its arguments after the first under control of the first argument. The first
argument is a character string which contains two types of objects: plain characters, which are simply
copied to the output stream, and conversion specifications, each of which causes conversion and printing of
the next successive argument to printf.
Each conversion specification is introduced by the character %. Following the %, there may be
− an optional minus sign ‘‘−’’ which specifies left adjustment of the converted argument in the indicat-
ed field;
− an optional digit string specifying a field width; if the converted argument has fewer characters than
the field width it will be blank-padded on the left (or right, if the left-adjustment indicator has been
given) to make up the field width;
− an optional period ‘‘.’’ which serves to separate the field width from the next digit string;
− an optional digit string (precision) which specifies the number of digits to appear after the decimal
point, for e- and f-conversion, or the maximum number of characters to be printed from a string;
− a character which indicates the type of conversion to be applied.
The conversion characters and their meanings are
d
o
x The integer argument is converted to decimal, octal, or hexadecimal notation respectively.
f The argument is converted to decimal notation in the style ‘‘[−]ddd.ddd’’ where the number of d’s
after the decimal point is equal to the precision specification for the argument. If the precision is
missing, 6 digits are given; if the precision is explicitly 0, no digits and no decimal point are printed.
The argument should be float or double.
e The argument is converted in the style ‘‘[−]d.ddde±dd’’ where there is one digit before the decimal
point and the number after is equal to the precision specification for the argument; when the preci-
sion is missing, 6 digits are produced. The argument should be a float or double quantity.
c The argument character is printed.
s The argument is taken to be a string (character pointer) and characters from the string are printed
until a null character or until the number of characters indicated by the precision specification is
reached; however if the precision is 0 or missing all characters up to a null are printed.
l The argument is taken to be an unsigned integer which is converted to decimal and printed (the re-
sult will be in the range 0 to 65535).
If no recognizable character appears after the %, that character is printed; thus % may be printed by use of
the string %%. In no case does a non-existent or small field width cause truncation of a field; padding
takes place only if the specified field width exceeds the actual width. Characters generated by printf are
printed by calling putchar.
SEE ALSO
putchar (III)
BUGS
Very wide fields (>128 characters) fail.

-1-
-

PUTC ( III ) 6/12/72 PUTC ( III )

NAME
putc, putw, fcreat, fflush − buffered output
SYNOPSIS
mov $filename,r0
jsr r5,fcreat; iobuf
fcreat(file, iobuf)
char *file;
struct buf *iobuf;
(get byte in r0)
jsr r5,putc; iobuf
putc(c, iobuf)
int c;
struct buf *iobuf;
(get word in r0)
jsr r5,putw; iobuf
putw(w, iobuf);
int w;
struct buf *iobuf;
jsr r5,flush; iobuf
fflush(iobuf)
struct buf *iobuf;
DESCRIPTION
Fcreat creates the given file (mode 666) and sets up the buffer iobuf (size 518 bytes); putc and putw write a
byte or word respectively onto the file; flush forces the contents of the buffer to be written, but does not
close the file. The structure of the buffer is:
struct buf {
int fildes; /* File descriptor */
int nunused; /* Remaining slots */
char *xfree; /* Ptr to next free slot */
char buff[512]; /* The buffer */
};
Before terminating, a program should call flush to force out the last of the output (fflush from C).
The user must supply iobuf, which should begin on a word boundary.
To write a new file using the same buffer, it suffices to call [f]flush, close the file, and call fcreat again.
SEE ALSO
creat (II), write (II), getc (III)
DIAGNOSTICS
Fcreat sets the error bit (c-bit) if the file creation failed (from C, returns −1). Putc and putw return their
character (word) argument. In all calls errno is set appropriately to 0 or to a system error number. See in-
troduction (II).
BUGS

-1-
-

PUTCHAR ( III ) 5/10/73 PUTCHAR ( III )

NAME
putchar, flush − write character
SYNOPSIS
putchar(ch)
flush( )
DESCRIPTION
Putchar writes out its argument and returns it unchanged. Only the low-order byte is written, and only if it
is non-null. Unless other arrangements have been made, putchar writes in unbuffered fashion on the stan-
dard output file.
Associated with this routine is an external variable fout which has the structure of a buffer discussed under
putc (III). If the file descriptor part of this structure (first word) is greater than 2, output via putchar is
buffered. To achieve buffered output one may say, for example,
fout = dup(1); or
fout = creat(...);
In such a case flush must be called before the program terminates in order to flush out the buffered output.
Flush may be called at any time.
SEE ALSO
putc (III)
BUGS
The fout notion is kludgy.

-1-
-

QSORT ( III ) 2/8/75 QSORT ( III )

NAME
qsort − quicker sort
SYNOPSIS
qsort(base, nel, width, compar)
char *base;
int (*compar)( );
DESCRIPTION
Qsort is an implementation of the quicker-sort algorithm. The first argument is a pointer to the base of the
data; the second is the number of elements; the third is the width of an element in bytes; the last is the name
of the comparison routine. It is called with two arguments which are pointers to the elements being com-
pared. The routine must return an integer less than, equal to, or greater than 0 according as the first argu-
ment is to be considered less than, equal to, or greater than the second.
SEE ALSO
sort (I)
BUGS

-1-
-

RAND ( III ) 1/15/73 RAND ( III )

NAME
rand, srand − random number generator
SYNOPSIS
(seed in r0)
jsr pc,srand /to initialize
jsr pc,rand /to get a random number
srand(seed)
int seed;
rand( )

DESCRIPTION
Rand uses a multiplicative congruential random number generator to return successive pseudo-random
numbers (in r0) in the range from 0 to 215−1.
The generator is reinitialized by calling srand with 1 as argument (in r0). It can be set to a random starting
point by calling srand with whatever you like as argument, for example the low-order word of the time.
BUGS
The low-order bits are not very random.

-1-
-

RESET ( III ) 5/10/73 RESET ( III )

NAME
reset, setexit − execute non-local goto
SYNOPSIS
setexit( )
reset( )
DESCRIPTION
These routines are useful for dealing with errors and interrupts encountered in a low-level subroutine of a
program.
Setexit saves its stack environment in a static place for later use by reset.
Reset restores the environment saved by the last call of setexit. It then returns in such a way that execution
continues as if the call of setexit had just returned. All accessible data have values as of the time reset was
called.
The routine that called setexit must still be active when reset is called.
SEE ALSO
signal (II)
BUGS

-1-
-

SETFIL ( III ) 10/29/73 SETFIL ( III )

NAME
setfil − specify Fortran file name
SYNOPSIS
call setfil ( unit, hollerith-string )
DESCRIPTION
Setfil provides a primitive way to associate an integer I/O unit number with a file named by the hollerith-
string. The end of the file name is indicated by a blank. Subsequent I/O on this unit number will refer to
the file whose name is specified by the string.
Setfil should be called only before any I/O has been done on the unit, or just after doing a rewind or end-
file. It is ineffective for unit numbers 5 and 6.
SEE ALSO
fc (I)
BUGS
The exclusion of units 5 and 6 is unwarranted.

-1-
-

SIN ( III ) 3/15/72 SIN ( III )

NAME
sin, cos − trigonometric functions
SYNOPSIS
jsr pc,sin (cos)
double sin(x)
double x;
double cos(x)
double x;
DESCRIPTION
The sine (cosine) of fr0 (resp. x), measured in radians, is returned (in fr0).
The magnitude of the argument should be checked by the caller to make sure the result is meaningful.
BUGS

-1-
-

SQRT ( III ) 3/15/72 SQRT ( III )

NAME
sqrt − square root function
SYNOPSIS
jsr pc,sqrt
double sqrt(x)
double x;
DESCRIPTION
The square root of fr0 (resp. x) is returned (in fr0).
DIAGNOSTICS
The c-bit is set on negative arguments and 0 is returned. There is no error return for C programs.
BUGS
No error return from C.

-1-
-

TTYN ( III ) 1/15/73 TTYN ( III )

NAME
ttyn − return name of current typewriter
SYNOPSIS
jsr pc,ttyn
ttyn(file)

DESCRIPTION
Ttyn hunts up the last character of the name of the typewriter which is the standard input (from as) or is
specified by the argument file descriptor (from C). If n is returned, the typewriter name is then ‘‘/dev/ttyn’’.
x is returned if the indicated file does not correspond to a typewriter.
BUGS

-1-
-

CAT ( IV ) 10/27/73 CAT ( IV )

NAME
cat − phototypesetter interface
DESCRIPTION
Cat provides the interface to a Graphic Systems C/A/T phototypesetter. Bytes written on the file specify
font, size, and other control information as well as the characters to be flashed. The coding will not be de-
scribed here.
Only one process may have this file open at a time. It is write-only.
FILES
/dev/cat
SEE ALSO
troff (I), Graphic Systems specification (available on request)
BUGS

-1-
-

DC ( IV ) 5/27/74 DC ( IV )

NAME
dc − DC-11 communications interface
DESCRIPTION
The discussion of typewriter I/O given in tty (IV) applies to these devices.
The DC-11 typewriter interface operates at any of four speeds, independently settable for input and output.
The speed is selected by the same encoding used by the DH (IV) device (enumerated in stty (II)); impossi-
ble speed changes are ignored.
FILES
/dev/tty[01234567abcd] 113B Dataphones (not currently connected− see dh (IV))
SEE ALSO
tty (IV), stty (II), dh (IV)
BUGS

-1-
-

DH ( IV ) 5/27/74 DH ( IV )

NAME
dh − DH-11 communications multiplexer
DESCRIPTION
Each line attached to the DH-11 communications multiplexer behaves as described in tty (IV). Input and
output for each line may independently be set to run at any of 16 speeds; see stty (II) for the encoding.
FILES
/dev/tty[f-u]
SEE ALSO
tty (IV), stty (II)
BUGS

-1-
-

DN ( IV ) 3/20/74 DN ( IV )

NAME
dn − DN-11 ACU interface
DESCRIPTION
The dn? files are write-only. The permissible codes are:
0-9 dial 0-9
: dial *
; dial #
− 4 second delay for second dial tone
= end-of-number
The entire telephone number must be presented in a single write system call.
It is recommended that an end-of-number code be given even though not all ACU’s actually require it.
FILES
/dev/dn0 connected to 801 with dp0
/dev/dn1 not currently connected
/dev/dn2 not currently connected
SEE ALSO
dp (IV)
BUGS

-1-
-

DP ( IV ) 8/24/73 DP ( IV )

NAME
dp − DP-11 201 data-phone interface
DESCRIPTION
The dp0 file is a 201 data-phone interface. Read and write calls to dp0 are limited to a maximum of 512
bytes. Each write call is sent as a single record. Seven bits from each byte are written along with an eighth
odd parity bit. The sync must be user supplied. Each read call returns characters received from a single
record. Seven bits are returned unaltered; the eighth bit is set if the byte was not received in odd parity. A
10 second time out is set and a zero-byte record is returned if nothing is received in that time.
FILES
/dev/dp0
SEE ALSO
dn (IV), gerts (III)
BUGS

-1-
-

HP ( IV ) 2/9/75 HP ( IV )

NAME
hp − RH-11/RP04 moving-head disk
DESCRIPTION
The files hp0 ... hp7 refer to sections of RP disk drive 0. The files hp8 ... hp15 refer to drive 1 etc. This is
done since the size of a full RP drive is 170,544 blocks and internally the system is only capable of address-
ing 65536 blocks. Also since the disk is so large, this allows it to be broken up into more manageable
pieces.
The origin and size of the pseudo-disks on each drive are as follows:
disk start length
0 0 9614
1 18392 65535
2 48018 65535
3 149644 20900
4 0 40600
5 41800 40600
6 83600 40600
7 125400 40600
It is unwise for all of these files to be present in one installation, since there is overlap in addresses and pro-
tection becomes a sticky matter.
The hp files access the disk via the system’s normal buffering mechanism and may be read and written
without regard to physical disk records. There is also a ‘‘raw’’ interface which provides for direct transmis-
sion between the disk and the user’s read or write buffer. A single read or write call results in exactly one
I/O operation and therefore raw I/O is considerably more efficient when many words are transmitted. The
names of the raw RP files begin with rhp and end with a number which selects the same disk section as the
corresponding hp file.
In raw I/O the buffer must begin on a word boundary, and counts should be a multiple of 512 bytes (a disk
block). Likewise seek calls should specify a multiple of 512 bytes.
FILES
/dev/hp?, /dev/rhp?
BUGS

-1-
-

HS ( IV ) 2/9/75 HS ( IV )

NAME
hs − RH11/RS03-RS04 fixed-head disk file
DESCRIPTION
The files hs0 ... hs7 refer to RJS03 disk drives 0 through 7. The files hs8 ... hs15 refer to RJS04 disk drives
0 through 7. The RJS03 drives are each 1024 blocks long and the RJS04 drives are 2048 blocks long.
The hs files access the disk via the system’s normal buffering mechanism and may be read and written
without regard to physical disk records. There is also a ‘‘raw’’ inteface which provides for direct transmis-
sion between the disk and the user’s read or write buffer. A single read or write call results in exactly one
I/O operation and therefore raw I/O is considerably more efficient when many words are transmitted. The
names of the raw HS files begin with rhs. The same minor device considerations hold for the raw interface
as for the normal interface.
In raw I/O the buffer must begin on a word boundary, and counts should be a multiple of 512 bytes (a disk
block). Likewise seek calls should specify a multiple of 512 bytes.
FILES
/dev/hs?, /dev/rhs?
BUGS

-1-
-

HT ( IV ) 2/9/75 HT ( IV )

NAME
ht − RH-11/TU-16 magtape interface
DESCRIPTION
The files mt0, ..., mt7 refer to the DEC RH/TM/TU16 magtape. When opened for reading or writing, the
tape is rewound. When closed, it is rewound; if it was open for writing, a double end-of-file is written first.
A standard tape consists of a series of 512 byte records terminated by a double end-of-file. To the extent
possible, the system makes it possible, if inefficient, to treat the tape like any other file. Seeks have their
usual meaning and it is possible to read or write a byte at a time. Writing in very small units is inadvisable,
however, because it tends to create monstrous record gaps.
The mt files discussed above are useful when it is desired to access the tape in a way compatible with ordi-
nary files. When foreign tapes are to be dealt with, and especially when long records are to be read or writ-
ten, the ‘‘raw’’ interface is appropriate. The associated files are named rmt0, ..., rmt7. Each read or write
call reads or writes the next record on the tape. In the write case the record has the same length as the
buffer given. During a read, the record size is passed back as the number of bytes read, provided it is no
greater than the buffer size; if the record is long, an error is indicated. In raw tape I/O, the buffer must be-
gin on a word boundary and the count must be even. Seeks are ignored. An error is returned when a tape
mark is read, but another read will fetch the first record of the new tape file.
FILES
/dev/mt?, /dev/rmt?
SEE ALSO
tp (I)
BUGS
Raw I/O doesnt work yet. The magtape system is supposed to be able to take 64 drives. Such addressing
has never been tried. These bugs will be fixed when we get more experience with this device.
If any non-data error is encountered, it refuses to do anything more until closed. In raw I/O, there should be
a way to perform forward and backward record and file spacing and to write an EOF mark.

-1-
-

KL ( IV ) 5/27/74 KL ( IV )

NAME
kl − KL-11 or DL-11 asynchronous interface
DESCRIPTION
The discussion of typewriter I/O given in tty (IV) applies to these devices.
Since they run at a constant speed, attempts to change the speed via stty (II) are ignored.
The on-line console typewriter is interfaced using a KL-11 or DL-11. By appropriate switch settings dur-
ing a reboot, UNIX will come up as a single-user system with I/O on the console typewriter.
FILES
/dev/tty8 console
SEE ALSO
tty (IV), init (VIII)
BUGS
Modem control for the DL-11E is not implemented.

-1-
-

LP ( IV ) 5/27/74 LP ( IV )

NAME
lp − line printer
DESCRIPTION
Lp provides the interface to any of the standard DEC line printers. When it is opened or closed, a suitable
number of page ejects is generated. Bytes written are printed.
An internal parameter within the driver determines whether or not the device is treated as having a 96- or
64-character set. In half-ASCII mode, lower case letters are turned into upper case and certain characters
are escaped according to the following table:
{ −(
} −)
` −́
| −!
˜ −̂
The driver correctly interprets carriage returns, backspaces, tabs, and form feeds. A sequence of newlines
which extends over the end of a page is turned into a form feed. All lines are indented 8 characters. Lines
longer than 80 characters are truncated. These numbers are parameters in the driver; another parameter al-
lows indenting all printout if it is unpleasantly near the left margin.
FILES
/dev/lp
SEE ALSO
lpr (I)
BUGS
Half-ASCII mode, the indent and the maximum line length should be settable by a call analogous to stty
(II).

-1-
-

MEM ( IV ) 5/27/74 MEM ( IV )

NAME
mem, kmem, null − core memory
DESCRIPTION
Mem is a special file that is an image of the core memory of the computer. It may be used, for example, to
examine, and even to patch the system using the debugger.
A memory address is an 18-bit quantity which is used directly as a UNIBUS address. References to non-
existent locations cause errors to be returned.
Examining and patching device registers is likely to lead to unexpected results when read-only or write-on-
ly bits are present.
The file kmem is the same as mem except that kernel virtual memory rather than physical memory is ac-
cessed. In particular, the I/O area of kmem is located beginning at 160000 (octal) rather than at 760000.
The 1K region beginning at 140000 (octal) is the system’s data for the current process.
The file null returns end-of-file on read and ignores write.
FILES
/dev/mem, /dev/kmem, /dev/null

-1-
-

PC ( IV ) 10/15/73 PC ( IV )

NAME
pc − PC-11 paper tape reader/punch
DESCRIPTION
Ppt refers to the PC-11 paper tape reader or punch, depending on whether it is read or written.
When ppt is opened for writing, a 100-character leader is punched. Thereafter each byte written is punched
on the tape. No editing of the characters is performed. When the file is closed, a 100-character trailer is
punched.
When ppt is opened for reading, the process waits until tape is placed in the reader and the reader is on-line.
Then requests to read cause the characters read to be passed back to the program, again without any editing.
This means that several null leader characters will usually appear at the beginning of the file. Likewise sev-
eral nulls are likely to appear at the end. End-of-file is generated when the tape runs out.
Seek calls for this file are meaningless.
FILES
/dev/ppt
BUGS
If both the reader and the punch are open simultaneously, the trailer is sometimes not punched. Sometimes
the reader goes into a dead state in which it cannot be opened.

-1-
-

RF ( IV ) 10/15/73 RF ( IV )

NAME
rf − RF11/RS11 fixed-head disk file
DESCRIPTION
This file refers to the concatenation of all RS-11 disks.
Each disk contains 1024 256-word blocks. The length of the combined RF file is 1024×(minor+1) blocks.
That is minor device zero is taken to be 1024 blocks long; minor device one is 2048, etc.
The rf0 file accesses the disk via the system’s normal buffering mechanism and may be read and written
without regard to physical disk records. There is also a ‘‘raw’’ interface which provides for direct transmis-
sion between the disk and the user’s read or write buffer. A single read or write call results in exactly one
I/O operation and therefore raw I/O is considerably more efficient when many words are transmitted. The
name of the raw RF file is rrf0. The same minor device considerations hold for the raw interface as for the
normal interface.
In raw I/O the buffer must begin on a word boundary, and counts should be a multiple of 512 bytes (a disk
block). Likewise seek calls should specify a multiple of 512 bytes.
FILES
/dev/rf0, /dev/rrf0
BUGS
The 512-byte restrictions on the raw device are not physically necessary, but are still imposed.

-1-
-

RK ( IV ) 10/15/73 RK ( IV )

NAME
rk − RK-11/RK03 (or RK05) disk
DESCRIPTION
Rk? refers to an entire RK03 disk as a single sequentially-addressed file. Its 256-word blocks are num-
bered 0 to 4871.
Drive numbers (minor devices) of eight and larger are treated specially. Drive 8+x is the x+1 way interleav-
ing of devices rk0 to rkx. Thus blocks on rk10 are distributed alternately among rk0, rk1, and rk2.
The rk files discussed above access the disk via the system’s normal buffering mechanism and may be read
and written without regard to physical disk records. There is also a ‘‘raw’’ interface which provides for di-
rect transmission between the disk and the user’s read or write buffer. A single read or write call results in
exactly one I/O operation and therefore raw I/O is considerably more efficient when many words are trans-
mitted. The names of the raw RK files begin with rrk and end with a number which selects the same disk
as the corresponding rk file.
In raw I/O the buffer must begin on a word boundary, and counts should be a multiple of 512 bytes (a disk
block). Likewise seek calls should specify a multiple of 512 bytes.
FILES
/dev/rk?, /dev/rrk?
BUGS
Care should be taken in using the interleaved files. First, the same drive should not be accessed simultane-
ously using the ordinary name and as part of an interleaved file, because the same physical blocks have in
effect two different names; this fools the system’s buffering strategy. Second, the combined files cannot be
used for swapping or raw I/O.

-1-
-

RP ( IV ) 2/21/74 RP ( IV )

NAME
rp − RP-11/RP03 moving-head disk
DESCRIPTION
The files rp0 ... rp7 refer to sections of RP disk drive 0. The files rp8 ... rp15 refer to drive 1 etc. This is
done since the size of a full RP drive is 81200 blocks and internally the system is only capable of address-
ing 65536 blocks. Also since the disk is so large, this allows it to be broken up into more manageable
pieces.
The origin and size of the pseudo-disks on each drive are as follows:
disk start length
0 0 40600
1 40600 40600
2 0 9200
3 72000 9200
4 0 65535
5 15600 65535
6-7 unassigned
It is unwise for all of these files to be present in one installation, since there is overlap in addresses and pro-
tection becomes a sticky matter. Here is a suggestion for two useful configurations: If the root of the file
system is on some other device and the RP used as a mounted device, then rp0 and rp1, which divide the
disk into two equal size portions, is a good idea. Other things being equal, it is advantageous to have two
equal-sized portions since one can always be copied onto the other, which is occasionally useful.
If the RP is the only disk and has to contain the root and the swap area, the root can be put on rp2 and a
mountable file system on rp5. Then the swap space can be put in the unused blocks 9200 to 15600 of rp0
(or, equivalently, rp4). This arrangement puts the root file system, the swap area, and the i-list of the
mounted file system relatively near each other and thus tends to minimize head movement.
The rp files access the disk via the system’s normal buffering mechanism and may be read and written
without regard to physical disk records. There is also a ‘‘raw’’ interface which provides for direct transmis-
sion between the disk and the user’s read or write buffer. A single read or write call results in exactly one
I/O operation and therefore raw I/O is considerably more efficient when many words are transmitted. The
names of the raw RP files begin with rrp and end with a number which selects the same disk section as the
corresponding rp file.
In raw I/O the buffer must begin on a word boundary, and counts should be a multiple of 512 bytes (a disk
block). Likewise seek calls should specify a multiple of 512 bytes.
FILES
/dev/rp?, /dev/rrp?
BUGS

-1-
-

TC ( IV ) 10/15/73 TC ( IV )

NAME
tc − TC-11/TU56 DECtape
DESCRIPTION
The files tap0 ... tap7 refer to the TC-11/TU56 DECtape drives 0 to 7.
The 256-word blocks on a standard DECtape are numbered 0 to 577.
FILES
/dev/tap?
SEE ALSO
tp (I)
BUGS

-1-
-

TM ( IV ) 2/21/74 TM ( IV )

NAME
tm − TM-11/TU-10 magtape interface
DESCRIPTION
The files mt0, ..., mt7 refer to the DEC TU10/TM11 magtape. When opened for reading or writing, the tape
is rewound. When closed, it is rewound; if it was open for writing, an end-of-file is written first.
A standard tape consists of a series of 512 byte records terminated by an end-of-file. To the extent possible,
the system makes it possible, if inefficient, to treat the tape like any other file. Seeks have their usual mean-
ing and it is possible to read or write a byte at a time. Writing in very small units is inadvisable, however,
because it tends to create monstrous record gaps.
The mt files discussed above are useful when it is desired to access the tape in a way compatible with ordi-
nary files. When foreign tapes are to be dealt with, and especially when long records are to be read or writ-
ten, the ‘‘raw’’ interface is appropriate. The associated files are named rmt0, ..., rmt7. Each read or write
call reads or writes the next record on the tape. In the write case the record has the same length as the
buffer given. During a read, the record size is passed back as the number of bytes read, provided it is no
greater than the buffer size; if the record is long, an error is indicated. In raw tape I/O, the buffer must be-
gin on a word boundary and the count must be even. Seeks are ignored. An error is returned when a tape
mark is read, but another read will fetch the first record of the new tape file.
FILES
/dev/mt?, /dev/rmt?
SEE ALSO
tp (I)
BUGS
If any non-data error is encountered, it refuses to do anything more until closed. In raw I/O, there should be
a way to perform forward and backward record and file spacing and to write an EOF mark.

-1-
-

TTY ( IV ) 5/27/74 TTY ( IV )

NAME
tty − general typewriter interface
DESCRIPTION
This section describes both a particular special file, and the general nature of the typewriter interface.
The file /dev/tty is, in each process, a synonym for the control typewriter associated with that process. It is
useful for programs or Shell sequences which wish to be sure of writing messages on the typewriter no mat-
ter how output has been redirected. It can also be used for programs which demand a file name for output,
when typed output is desired and it is tiresome to find out which typewriter is currently in use.
As for typewriters in general: all of the low-speed asynchronous communications ports use the same gener-
al interface, no matter what hardware is involved. The remainder of this section discusses the common fea-
tures of the interface; the KL, DC, and DH writeups (IV) describe peculiarities of the individual devices.
When a typewriter file is opened, it causes the process to wait until a connection is established. In practice
user’s programs seldom open these files; they are opened by init and become a user’s input and output file.
The very first typewriter file open in a process becomes the control typewriter for that process. The control
typewriter plays a special role in handling quit or interrupt signals, as discussed below. The control type-
writer is inherited by a child process during a fork.
A terminal associated with one of these files ordinarily operates in full-duplex mode. Characters may be
typed at any time, even while output is occurring, and are only lost when the system’s character input
buffers become completely choked, which is rare, or when the user has accumulated the maximum allowed
number of input characters which have not yet been read by some program. Currently this limit is 256
characters. When the input limit is reached all the saved characters are thrown away without notice.
These special files have a number of modes which can be changed by use of the stty system call (II). When
first opened, the interface mode is 300 baud; either parity accepted; 10 bits/character (one stop bit); and
newline action character. Modes that can be changed by stty include the interface speed (if the hardware
permits); acceptance of even parity, odd parity, or both; a raw mode in which all characters may be read one
at a time; a carriage return (CR) mode in which CR is mapped into newline on input and either CR or line
feed (LF) cause echoing of the sequence LF-CR; mapping of upper case letters into lower case; suppression
of echoing; a variety of delays after function characters; and the printing of tabs as spaces. See getty (VIII)
for the way that terminal speed and type are detected.
Normally, typewriter input is processed in units of lines. This means that a program attempting to read will
be suspended until an entire line has been typed. Also, no matter how many characters are requested in the
read call, at most one line will be returned. It is not however necessary to read a whole line at once; any
number of characters may be requested in a read, even one, without losing information.
During input, erase and kill processing is normally done. By default, the character ‘#’ erases the last char-
acter typed, except that it will not erase beyond the beginning of a line or an EOT. By default, the character
‘@’ kills the entire line up to the point where it was typed, but not beyond an EOT. Both these characters
operate on a keystroke basis independently of any backspacing or tabbing that may have been done. Either
‘@’ or ‘#’ may be entered literally by preceding it by ‘\’; the erase or kill character remains, but the ‘\’ dis-
appears. These two characters may be changed to others.
When desired, all upper-case letters are mapped into the corresponding lower-case letter. The upper-case
letter may be generated by preceding it by ‘\’. In addition, the following escape sequences are generated on
output and accepted on input:
for use
` \´
| \!
˜ \ˆ
{ \(
} \)
In raw mode, the program reading is awakened on each character. No erase or kill processing is done; and
the EOT, quit and interrupt characters are not treated specially. The input parity bit is passed back to the
reader, but parity is still generated for output characters.

-1-
-

TTY ( IV ) 5/27/74 TTY ( IV )

The ASCII EOT (control-D) character may be used to generate an end of file from a typewriter. When an
EOT is received, all the characters waiting to be read are immediately passed to the program, without wait-
ing for a new-line, and the EOT is discarded. Thus if there are no characters waiting, which is to say the
EOT occurred at the beginning of a line, zero characters will be passed back, and this is the standard end-
of-file indication. The EOT is passed back unchanged in raw mode.
When the carrier signal from the dataset drops (usually because the user has hung up his terminal) a hangup
signal is sent to all processes with the typewriter as control typewriter. Unless other arrangements have
been made, this signal causes the processes to terminate. If the hangup signal is ignored, any read returns
with an end-of-file indication. Thus programs which read a typewriter and test for end-of-file on their input
can terminate appropriately when hung up on.
Two characters have a special meaning when typed. The ASCII DEL character (sometimes called ‘rubout’)
is not passed to a program but generates an interrupt signal which is sent to all processes with the associat-
ed control typewriter. Normally each such process is forced to terminate, but arrangements may be made
either to ignore the signal or to receive a trap to an agreed-upon location. See signal (II).
The ASCII character FS generates the quit signal. Its treatment is identical to the interrupt signal except
that unless a receiving process has made other arrangements it will not only be terminated but a core image
file will be generated. If you find it hard to type this character, try control-\ or control-shift-L.
When one or more characters are written, they are actually transmitted to the terminal as soon as previous-
ly-written characters have finished typing. Input characters are echoed by putting them in the output queue
as they arrive. When a process produces characters more rapidly than they can be typed, it will be suspend-
ed when its output queue exceeds some limit. When the queue has drained down to some threshold the pro-
gram is resumed. Even parity is always generated on output. The EOT character is not transmitted (except
in raw mode) to prevent terminals which respond to it from hanging up.
FILES
/dev/tty
SEE ALSO
dc (IV), kl (IV), dh (IV), getty (VIII), stty (I, II), gtty (I, II), signal (II)
BUGS
Half-duplex terminals are not supported. On raw-mode output, parity should be transmitted as specified in
the characters written.

-2-
-

A.OUT ( V ) 9/9/73 A.OUT ( V )

NAME
a.out − assembler and link editor output
DESCRIPTION
A.out is the output file of the assembler as and the link editor ld. Both programs make a.out executable if
there were no errors and no unresolved external references.
This file has four sections: a header, the program and data text, a symbol table, and relocation bits (in that
order). The last two may be empty if the program was loaded with the ‘‘−s’’ option of ld or if the symbols
and relocation have been removed by strip.
The header always contains 8 words:
1 A magic number (407, 410, or 411(8))
2 The size of the program text segment
3 The size of the initialized portion of the data segment
4 The size of the uninitialized (bss) portion of the data segment
5 The size of the symbol table
6 The entry location (always 0 at present)
7 Unused
8 A flag indicating relocation bits have been suppressed
The sizes of each segment are in bytes but are even. The size of the header is not included in any of the
other sizes.
When a file produced by the assembler or loader is loaded into core for execution, three logical segments
are set up: the text segment, the data segment (with uninitialized data, which starts off as all 0, following
initialized), and a stack. The text segment begins at 0 in the core image; the header is not loaded. If the
magic number (word 0) is 407, it indicates that the text segment is not to be write-protected and shared, so
the data segment is immediately contiguous with the text segment. If the magic number is 410, the data
segment begins at the first 0 mod 8K byte boundary following the text segment, and the text segment is not
writable by the program; if other processes are executing the same file, they will share the text segment. If
the magic number is 411, the text segment is again pure, write-protected, and shared, and moreover instruc-
tion and data space are separated; the text and data segment both begin at location 0. See the 11/45 hand-
book for restrictions which apply to this situation.
The stack will occupy the highest possible locations in the core image: from 177776(8) and growing down-
wards. The stack is automatically extended as required. The data segment is only extended as requested by
the break system call.
The start of the text segment in the file is 20(8); the start of the data segment is 20+St (the size of the text)
the start of the relocation information is 20+St+Sd; the start of the symbol table is 20+2(St+Sd) if the reloca-
tion information is present, 20+St+Sd if not.
The symbol table consists of 6-word entries. The first four words contain the ASCII name of the symbol,
null-padded. The next word is a flag indicating the type of symbol. The following values are possible:
00 undefined symbol
01 absolute symbol
02 text segment symbol
03 data segment symbol
37 file name symbol (produced by ld)
04 bss segment symbol
40 undefined external (.globl) symbol
41 absolute external symbol
42 text segment external symbol
43 data segment external symbol
44 bss segment external symbol
Values other than those given above may occur if the user has defined some of his own instructions.

-1-
-

A.OUT ( V ) 9/9/73 A.OUT ( V )

The last word of a symbol table entry contains the value of the symbol.
If the symbol’s type is undefined external, and the value field is non-zero, the symbol is interpreted by the
loader ld as the name of a common region whose size is indicated by the value of the symbol.
The value of a word in the text or data portions which is not a reference to an undefined external symbol is
exactly that value which will appear in core when the file is executed. If a word in the text or data portion
involves a reference to an undefined external symbol, as indicated by the relocation bits for that word, then
the value of the word as stored in the file is an offset from the associated external symbol. When the file is
processed by the link editor and the external symbol becomes defined, the value of the symbol will be
added into the word in the file.
If relocation information is present, it amounts to one word per word of program text or initialized data.
There is no relocation information if the ‘‘suppress relocation’’ flag in the header is on.
Bits 3-1 of a relocation word indicate the segment referred to by the text or data word associated with the
relocation word:
00 indicates the reference is absolute
02 indicates the reference is to the text segment
04 indicates the reference is to initialized data
06 indicates the reference is to bss (uninitialized data)
10 indicates the reference is to an undefined external symbol.
Bit 0 of the relocation word indicates if on that the reference is relative to the pc (e.g. ‘‘clr x’’); if off, that
the reference is to the actual symbol (e.g., ‘‘clr *$x’’).
The remainder of the relocation word (bits 15-4) contains a symbol number in the case of external refer-
ences, and is unused otherwise. The first symbol is numbered 0, the second 1, etc.
SEE ALSO
as (I), ld (I), strip (I), nm (I)

-2-
-

ARCHIVE ( V ) 9/10/73 ARCHIVE ( V )

NAME
ar − archive (library) file format
DESCRIPTION
The archive command ar is used to combine several files into one. Archives are used mainly as libraries to
be searched by the link-editor ld.
A file produced by ar has a magic number at the start, followed by the constituent files, each preceded by a
file header. The magic number is 177555(8) (it was chosen to be unlikely to occur anywhere else). The
header of each file is 16 bytes long:
0-7 file name, null padded on the right
8-11 modification time of the file
12 user ID of file owner
13 file mode
14-15 file size
Each file begins on a word boundary; a null byte is inserted between files if necessary. Nevertheless the
size give reflects the actual size of the file exclusive of padding.
Notice there is no provision for empty areas in an archive file.
SEE ALSO
ar (I), ld (I)
BUGS
Names are only 8 characters, not 14. More important, there isn’t enough room to store the proper mode, so
ar always extracts in mode 666.

-1-
-

ASCII ( V ) 6/12/72 ASCII ( V )

NAME
ascii − map of ASCII character set
SYNOPSIS
cat /usr/pub/ascii
DESCRIPTION
Ascii is a map of the ASCII character set, to be printed as needed. It contains:

| 000 nu l | 001 s oh | 002 s t x | 003 e t x | 004 eo t | 005 enq | 006 a ck | 007 be l |

| 010 b s | 011 h t | 012 n l | 013 v t | 014 np | 015 c r | 016 s o | 017 si |
| 020 d l e | 021 dc1 | 022 dc2 | 023 dc3 | 024 dc4 | 025 nak | 026 s yn | 027 etb|
| 030 c an | 031 em | 032 s ub | 033 e s c | 034 f s | 035 g s | 036 r s | 037 us |
| 040 s p | 041 ! | 042 " | 043 # | 044 $ | 045 % | 046 & | 047 ´ |
| 050 ( | 051 ) | 052 * | 053 + | 054 , | 055 − | 056 . | 057 / |
| 060 0 | 061 1 | 062 2 | 063 3 | 064 4 | 065 5 | 066 6 | 067 7 |
| 070 8 | 071 9 | 072 : | 073 ; | 074 < | 075 = | 076 > | 077 ? |
| 100 @ | 101 A | 102 B | 103 C | 104 D | 105 E | 106 F | 107 G |
| 110 H | 111 I | 112 J | 113 K | 114 L | 115 M | 116 N | 117 O |
| 120 P | 121 Q | 122 R | 123 S | 124 T | 125 U | 126 V | 127 W |
| 130 X | 131 Y | 132 Z | 133 [ | 134 \ | 135 ] | 136 ˆ | 137 |
| 140 ` | 141 a | 142 b | 143 c | 144 d | 145 e | 146 f | 147 g |
| 150 h | 151 i | 152 j | 153 k | 154 l | 155 m | 156 n | 157 o |
| 160 p | 161 q | 162 r | 163 s | 164 t | 165 u | 166 v | 167 w |
| 170 x | 171 y | 172 z | 173 { | 174 | | 175 } | 176 ˜ | 177 de l |

FILES
found in /usr/pub

-1-
-

CORE ( V ) 2/11/75 CORE ( V )

NAME
core − format of core image file
DESCRIPTION
UNIX writes out a core image of a terminated process when any of various errors occur. See signal (II) for
the list of reasons; the most common are memory violations, illegal instructions, bus errors, and user-gener-
ated quit signals. The core image is called ‘‘core’’ and is written in the process’s working directory (pro-
vided it can be; normal access controls apply).
The first 1024 bytes of the core image are a copy of the system’s per-user data for the process, including the
registers as they were at the time of the fault. The remainder represents the actual contents of the user’s
core area when the core image was written. If the text segment is write-protected and shared, it is not
dumped; otherwise the entire address space is dumped.
The format of the information in the first 1024 bytes is described by the user structure of the system. The
important stuff not detailed therein is the locations of the registers. Here are their offsets. The parenthe-
sized numbers for the floating registers are used if the floating-point hardware is in single precision mode,
as indicated in the status register.
fpsr 0004
fr0 0006 (0006)
fr1 0036 (0022)
fr2 0046 (0026)
fr3 0056 (0032)
fr4 0016 (0012)
fr5 0026 (0016)
r0 1772
r1 1766
r2 1750
r3 1752
r4 1754
r5 1756
sp 1764
pc 1774
ps 1776
In general the debuggers db (I) and cdb (I) are sufficient to deal with core images.
SEE ALSO
cdb (I), db (I), signal (II)

-1-
-

DIRECTORY ( V ) 9/10/73 DIRECTORY ( V )

NAME
dir − format of directories
DESCRIPTION
A directory behaves exactly like an ordinary file, save that no user may write into a directory. The fact that
a file is a directory is indicated by a bit in the flag word of its i-node entry. Directory entries are 16 bytes
long. The first word is the i-number of the file represented by the entry, if non-zero; if zero, the entry is
empty.
Bytes 2-15 represent the (14-character) file name, null padded on the right. These bytes are not cleared for
empty slots.
By convention, the first two entries in each directory are for ‘‘.’’ and ‘‘..’’. The first is an entry for the direc-
tory itself. The second is for the parent directory. The meaning of ‘‘..’’ is modified for the root directory of
the master file system and for the root directories of removable file systems. In the first case, there is no
parent, and in the second, the system does not permit off-device references. Therefore in both cases ‘‘..’’
has the same meaning as ‘‘.’’.
SEE ALSO
file system (V)

-1-
-

DUMP ( V ) 2/11/75 DUMP ( V )

NAME
dump − incremental dump tape format
DESCRIPTION
The dump and restor commands are used to write and read incremental dump magnetic tapes.
The dump tape consists of blocks of 512-bytes each. The first block has the following structure.
struct {
int isize;
int fsize;
int date[2];
int ddate[2];
int tsize;
};
Isize, and fsize are the corresponding values from the super block of the dumped file system. (See file sys-
tem (V).) Date is the date of the dump. Ddate is the incremental dump date. The incremental dump con-
tains all files modified between ddate and date. Tsize is the number of blocks per reel. This block check-
sums to the octal value 031415.
Next there are enough whole tape blocks to contain one word per file of the dumped file system. This is
isize divided by 16 rounded to the next higher integer. The first word corresponds to i-node 1, the second to
i-node 2, and so forth. If a word is zero, then the corresponding file exists, but was not dumped. (Was not
modified after ddate) If the word is −1, the file does not exist. Other values for the word indicate that the
file was dumped and the value is one more than the number of blocks it contains.
The rest of the tape contains for each dumped file a header block and the data blocks from the file. The
header contains an exact copy of the i-node (see file system (V)) and also checksums to 031415. The next-
to-last word of the block contains the tape block number, to aid in (unimplemented) recovery after tape er-
rors. The number of data blocks per file is directly specified by the control word for the file and indirectly
specified by the size in the i-node. If these numbers differ, the file was dumped with a ‘phase error’.
SEE ALSO
dump (VIII), restor (VIII), file system(V)

-1-
-

FILE SYSTEM ( V ) 2/9/75 FILE SYSTEM ( V )

NAME
fs − format of file system volume
DESCRIPTION
Every file system storage volume (e.g. RF disk, RK disk, RP disk, DECtape reel) has a common format for
certain vital information. Every such volume is divided into a certain number of 256 word (512 byte)
blocks. Block 0 is unused and is available to contain a bootstrap program, pack label, or other information.
Block 1 is the super block. Starting from its first word, the format of a super-block is
struct {
int isize;
int fsize;
int nfree;
int free[100];
int ninode;
int inode[100];
char flock;
char ilock;
char fmod;
int time[2];
};
Isize is the number of blocks devoted to the i-list, which starts just after the super-block, in block 2. Fsize
is the first block not potentially available for allocation to a file. These numbers are used by the system to
check for bad block numbers; if an ‘‘impossible’’ block number is allocated from the free list or is freed, a
diagnostic is written on the on-line console. Moreover, the free array is cleared, so as to prevent further al-
location from a presumably corrupted free list.
The free list for each volume is maintained as follows. The free array contains, in free[1], ... ,
free[nfree−1], up to 99 numbers of free blocks. Free[0] is the block number of the head of a chain of
blocks constituting the free list. The first word in each free-chain block is the number (up to 100) of free-
block numbers listed in the next 100 words of this chain member. The first of these 100 blocks is the link
to the next member of the chain. To allocate a block: decrement nfree, and the new block is free[nfree]. If
the new block number is 0, there are no blocks left, so give an error. If nfree became 0, read in the block
named by the new block number, replace nfree by its first word, and copy the block numbers in the next 100
words into the free array. To free a block, check if nfree is 100; if so, copy nfree and the free array into it,
write it out, and set nfree to 0. In any event set free[nfree] to the freed block’s number and increment nfree.
Ninode is the number of free i-numbers in the inode array. To allocate an i-node: if ninode is greater than 0,
decrement it and return inode[ninode]. If it was 0, read the i-list and place the numbers of all free inodes
(up to 100) into the inode array, then try again. To free an i-node, provided ninode is less than 100, place
its number into inode[ninode] and increment ninode. If ninode is already 100, don’t bother to enter the
freed i-node into any table. This list of i-nodes is only to speed up the allocation process; the information
as to whether the inode is really free or not is maintained in the inode itself.
Flock and ilock are flags maintained in the core copy of the file system while it is mounted and their values
on disk are immaterial. The value of fmod on disk is likewise immaterial; it is used as a flag to indicate that
the super-block has changed and should be copied to the disk during the next periodic update of file system
information.
Time is the last time the super-block of the file system was changed, and is a double-precision representa-
tion of the number of seconds that have elapsed since 0000 Jan. 1 1970 (GMT). During a reboot, the time
of the super-block for the root file system is used to set the system’s idea of the time.
I-numbers begin at 1, and the storage for i-nodes begins in block 2. Also, i-nodes are 32 bytes long, so 16
of them fit into a block. Therefore, i-node i is located in block (i + 31) / 16, and begins 32.((i + 31) (mod
16) bytes from its start. I-node 1 is reserved for the root directory of the file system, but no other i-number
has a built-in meaning. Each i-node represents one file. The format of an i-node is as follows.

-1-
-

FILE SYSTEM ( V ) 2/9/75 FILE SYSTEM ( V )

struct {
int flags; /* +0: see below */
char nlinks; /* +2: number of links to file */
char uid; /* +3: user ID of owner */
char gid; /* +4: group ID of owner */
char size0; /* +5: high byte of 24-bit size */
int size1; /* +6: low word of 24-bit size */
int addr[8]; /* +8: block numbers or device number */
int actime[2]; /* +24: time of last access */
int modtime[2]; /* +28: time of last modification */
};
The flags are as follows:
100000 i-node is allocated
060000 2-bit file type:
000000 plain file
040000 directory
020000 character-type special file
060000 block-type special file.
010000 large file
004000 set user-ID on execution
002000 set group-ID on execution
000400 read (owner)
000200 write (owner)
000100 execute (owner)
000070 read, write, execute (group)
000007 read, write, execute (others)
Special files are recognized by their flags and not by i-number. A block-type special file is basically one
which can potentially be mounted as a file system; a character-type special file cannot, though it is not nec-
essarily character-oriented. For special files the high byte of the first address word specifies the type of de-
vice; the low byte specifies one of several devices of that type. The device type numbers of block and char-
acter special files overlap.
The address words of ordinary files and directories contain the numbers of the blocks in the file (if it is
small) or the numbers of indirect blocks (if the file is large). Byte number n of a file is accessed as follows.
N is divided by 512 to find its logical block number (say b ) in the file. If the file is small (flag 010000 is 0),
then b must be less than 8, and the physical block number is addr[b].
If the file is large, b is divided by 256 to yield i. If i is less than 7, then addr[i] is the physical block num-
ber of the indirect block. The remainder from the division yields the word in the indirect block which con-
tains the number of the block for the sought-for byte.
If i is equal to 7, then the file has become extra-large (huge), and addr[7] is the address of a first indirect
block. Each word in this block is the number of a second-level indirect block; each word in the second-lev-
el indirect blocks points to a data block. Notice that extra-large files are not marked by any mode bit, but
only by having addr[7] non-zero; and that although this scheme allows for more than 256×256×512 =
33,554,432 bytes per file, the length of files is stored in 24 bits so in practice a file can be at most
16,777,216 bytes long.
For block b in a file to exist, it is not necessary that all blocks less than b exist. A zero block number either
in the address words of the i-node or in an indirect block indicates that the corresponding block has never
been allocated. Such a missing block reads as if it contained all zero words.
SEE ALSO
icheck, dcheck (VIII)

-2-
-

GREEK ( V ) 10/31/72 GREEK ( V )

NAME
greek − graphics for extended TTY-37 type-box
SYNOPSIS
cat /usr/pub/greek
DESCRIPTION
Greek gives the mapping from ascii to the ‘‘shift out’’ graphics in effect between SO and SI on model 37
Teletypes with a 128-character type-box. It contains:
alpha α A beta β B gamma γ \
GAMMA Γ G delta δ D DELTA ∆ W
epsilon ε S zeta ζ Q eta η N
THETA Θ T theta θ O lambda λ L
LAMBDA Λ E mu µ M nu ν @
xi ξ X pi π J PI Π P
rho ρ K sigma σ Y SIGMA Σ R
tau τ I phi φ U PHI Φ F
psi ψ V PSI Ψ H omega ω C
OMEGA Ω Z nabla ∇ [ not ¬
partial ∂ ] integral ∫ ˆ
SEE ALSO
ascii (VII)

-1-
-

GROUP ( V ) 2/10/75 GROUP ( V )

NAME
group − group file
DESCRIPTION
Group contains for each group the following information:
group name
encrypted password
numerical group ID
a comma separated list of all users allowed in the group
This is an ASCII file. The fields are separated by colons; Each group is separated from the next by a new-
line. If the password field is null, no password is demanded.
This file resides in directory /etc. Because of the encrypted passwords, it can and does have general read
permission and can be used, for example, to map numerical group ID’s to names.
FILES
/etc/group
SEE ALSO
newgrp (I), login (I), crypt (III), passwd (I)

-1-
-

MTAB ( V ) 1/6/74 MTAB ( V )

NAME
mtab − mounted file system table
DESCRIPTION
Mtab resides in directory /etc and contains a table of devices mounted by the mount command. Umount re-
moves entries.
Each entry is 64 bytes long; the first 32 are the null-padded name of the place where the special file is
mounted; the second 32 are the null-padded name of the special file. The special file has all its directories
stripped away; that is, everything through the last ‘‘/’’ is thrown away.
This table is present only so people can look at it. It does not matter to mount if there are duplicated entries
nor to umount if a name cannot be found.
FILES
/etc/mtab
SEE ALSO
mount (VIII), umount (VIII)
BUGS

-1-
-

PASSWD ( V ) 9/10/73 PASSWD ( V )

NAME
passwd − password file
DESCRIPTION
Passwd contains for each user the following information:
name (login name, contains no upper case)
encrypted password
numerical user ID
numerical group ID (for now, always 1)
GCOS job number, box number, optional GCOS user-id
initial working directory
program to use as Shell
This is an ASCII file. Each field within each user’s entry is separated from the next by a colon. The GCOS
field is used only when communicating with that system, and in other installations can contain any desired
information. Each user is separated from the next by a new-line. If the password field is null, no password
is demanded; if the Shell field is null, the Shell itself is used.
This file resides in directory /etc. Because of the encrypted passwords, it can and does have general read
permission and can be used, for example, to map numerical user ID’s to names.
FILES
/etc/passwd
SEE ALSO
login (I), crypt (III), passwd (I), group (V)

-1-
-

TABS ( V ) 6/15/72 TABS ( V )

NAME
tabs − set tab stops
SYNOPSIS
cat /usr/pub/tabs
DESCRIPTION
Printing this file on a suitable terminal sets tab stops every 8 columns. Suitable terminals include the Tele-
type model 37 and the GE TermiNet 300.
These tab stop settings are desirable because UNIX assumes them in calculating delays.

-1-
-

TP ( V ) 9/10/73 TP ( V )

NAME
tp − DEC/mag tape formats
DESCRIPTION
The command tp dumps files to and extracts files from DECtape and magtape. The formats of these tapes
are the same except that magtapes have larger directories.
Block zero contains a copy of a stand-alone bootstrap program. See boot procedures (VIII).
Blocks 1 through 24 for DECtape (1 through 62 for magtape) contain a directory of the tape. There are 192
(resp. 496) entries in the directory; 8 entries per block; 64 bytes per entry. Each entry has the following for-
mat:
path name 32 bytes
mode 2 bytes
uid 1 byte
gid 1 byte
unused 1 byte
size 3 bytes
time modified 4 bytes
tape address 2 bytes
unused 16 bytes
check sum 2 bytes
The path name entry is the path name of the file when put on the tape. If the pathname starts with a zero
word, the entry is empty. It is at most 32 bytes long and ends in a null byte. Mode, uid, gid, size and time
modified are the same as described under i-nodes (file system (V)). The tape address is the tape block num-
ber of the start of the contents of the file. Every file starts on a block boundary. The file occupies
(size+511)/512 blocks of continuous tape. The checksum entry has a value such that the sum of the 32
words of the directory entry is zero.
Blocks 25 (resp. 63) on are available for file storage.
A fake entry (see tp (I)) has a size of zero.
SEE ALSO
file system (V), tp (I)

-1-
-

TTYS ( V ) 2/11/75 TTYS ( V )

NAME
ttys − typewriter initialization data
DESCRIPTION
The ttys file is read by the init program and specifies which typewriter special files are to have a process
created for them which will allow people to log in. It consists of lines of 3 characters each.
The first character is either ‘0’ or ‘1’; the former causes the line to be ignored, the latter causes it to be ef-
fective. The second character is the last character in the name of a typewriter; e.g. x refers to the file
‘/dev/ttyx’. The third character is used as an argument to the getty program, which performs such tasks as
baud-rate recognition, reading the login name, and calling login. For normal lines, the character is ‘0’; oth-
er characters can be used, for example, with hard-wired terminals where speed recognition is unnecessary
or which have special characteristics. (Getty will have to be fixed in such cases.)
FILES
/etc/ttys
SEE ALSO
init (VIII), getty (VIII), login (I)

-1-
-

UTMP ( V ) 9/10/73 UTMP ( V )

NAME
utmp − user information
DESCRIPTION
This file allows one to discover information about who is currently using UNIX. The file is binary; each
entry is 16(10) bytes long. The first eight bytes contain a user’s login name or are null if the table slot is
unused. The low order byte of the next word contains the last character of a typewriter name. The next two
words contain the user’s login time. The last word is unused.
FILES
/etc/utmp
SEE ALSO
init (VIII) and login (I), which maintain the file; who (I), which interprets it.

-1-
-

WTMP ( V ) 2/22/74 WTMP ( V )

NAME
wtmp − user login history
DESCRIPTION
This file records all logins and logouts. Its format is exactly like utmp (V) except that a null user name in-
dicates a logout on the associated typewriter. Furthermore, the typewriter name ‘˜’ indicates that the system
was rebooted at the indicated time; the adjacent pair of entries with typewriter names ‘|’ and ‘}’ indicate the
system-maintained time just before and just after a date command has changed the system’s idea of the
time.
Wtmp is maintained by login (I) and init (VIII). Neither of these programs creates the file, so if it is re-
moved record-keeping is turned off. It is summarized by ac (VIII).
FILES
/usr/adm/wtmp
SEE ALSO
utmp (V), login (I), init (VIII), ac (VIII), who (I)

-1-
-

AZEL ( VI ) 6/3/74 AZEL ( VI )

NAME
azel − satellite predictions
SYNOPSIS
azel [ −d ] [ −l ] satellite1 [ −d ] [ −l ] satellite2 ...
DESCRIPTION
Azel predicts, in convenient form, the apparent trajectories of Earth satellites whose orbital elements are
given in the argument files. If a given satellite name cannot be read, an attempt is made to find it in a direc-
tory of satellites maintained by the programs’s author. The −d option causes azel to ask for a date and read
line 1 data (see below) from the standard input. The −l option causes azel to ask for the observer’s latitude,
west-longitude, and height above sea level.
For each satellite given the program types its full name, the date, and a sequence of lines each containing a
time, an azimuth, an elevation, a distance, and a visual magnitude. Each such line indicates that: at the in-
dicated time, the satellite may be seen from Murray Hill (or provided location) at the indicated azimuth and
elevation, and that its distance and apparent magnitude are as given. Predictions are printed only when the
sky is dark (sun more than 5 degrees below the horizon) and when the satellite is not eclipsed by the earth’s
shadow. Satellites which have not been seen and verified will not have had their visual magnitude level set
correctly.
All times input and output by azel are GMT (Universal Time).
The satellites for which elements are maintained are:
sla,b,e,f,k Skylab A through Skylab K. Skylab A is the laboratory; B was the rocket but it has crashed. A
and probably K have been verified.
cop Copernicus I. Never verified.
oao Orbiting Astronomical Observatory. Seen and verified.
pag Pageos I. Seen and verified; fairly dim (typically 2nd-3rd magnitude), but elements are ex-
tremely accurate.
exp19 Explorer 19; seen and verified, but quite dim (4th-5th magnitude) and fast-moving.
c103b, c156b, c184b, c206b, c220b, c461b, c500b
Various of the USSR Cosmos series; none seen.
7276a Unnamed (satellite # 72-76A); not seen.
The element files used by azel contain five lines. The first line gives a year, month number, day, hour, and
minute at which the program begins its consideration of the satellite, followed by a number of minutes and
an interval in minutes. If the year, month, and day are 0, they are taken to be the current date (taken to
change at 6 A.M. local time). The output report starts at the indicated epoch and prints the position of the
satellite for the indicated number of minutes at times separated by the indicated interval. This line is ended
by two numbers which specify options to the program governing the completeness of the report; they are
ordinarily both ‘‘1’’. The first option flag suppresses output when the sky is not dark; the second supresses
output when the satellite is eclipsed by the earth’s shadow. The next line of an element file is the full name
of the satellite. The next three are the elements themselves (including certain derivatives of the elements).
FILES
/usr/jfo/el/* − orbital element files
SEE ALSO
sky (VI)
AUTHOR
J. F. Ossanna
BUGS

-1-
-

BJ ( VI ) 3/15/72 BJ ( VI )

NAME
bj − the game of black jack
SYNOPSIS
/usr/games/bj
DESCRIPTION
Bj is a serious attempt at simulating the dealer in the game of black jack (or twenty-one) as might be found
in Reno. The following rules apply:
The bet is $2 every hand.
A player ‘natural’ (black jack) pays $3. A dealer natural loses $2. Both dealer and player naturals is
a ‘push’ (no money exchange).
If the dealer has an ace up, the player is allowed to make an ‘insurance’ bet against the chance of a
dealer natural. If this bet is not taken, play resumes as normal. If the bet is taken, it is a side bet
where the player wins $2 if the dealer has a natural and loses $1 if the dealer does not.
If the player is dealt two cards of the same value, he is allowed to ‘double’. He is allowed to play
two hands, each with one of these cards. (The bet is doubled also; $2 on each hand.)
If a dealt hand has a total of ten or eleven, the player may ‘double down’. He may double the bet ($2
to $4) and receive exactly one more card on that hand.
Under normal play, the player may ‘hit’ (draw a card) as long as his total is not over twenty-one. If
the player ‘busts’ (goes over twenty-one), the dealer wins the bet.
When the player ‘stands’ (decides not to hit), the dealer hits until he attains a total of seventeen or
more. If the dealer busts, the player wins the bet.
If both player and dealer stand, the one with the largest total wins. A tie is a push.
The machine deals and keeps score. The following questions will be asked at appropriate times. Each
question is answered by y followed by a new line for ‘yes’, or just new line for ‘no’.
? (means, ‘‘do you want a hit?’’)
Insurance?
Double down?
Every time the deck is shuffled, the dealer so states and the ‘action’ (total bet) and ‘standing’ (total won or
lost) is printed. To exit, hit the interrupt key (DEL) and the action and standing will be printed.
BUGS

-1-
-

CAL ( VI ) 11/1/73 CAL ( VI )

NAME
cal − print calendar
SYNOPSIS
cal [ month ] year
DESCRIPTION
Cal prints a calendar for the specified year. If a month is also specified, a calendar just for that month is
printed. Year can be between 1 and 9999. The month is a number between 1 and 12. The calendar pro-
duced is that for England and her colonies.
Try September 1752.
BUGS
The year is always considered to start in January even though this is historically naive.

-1-
-

CHESS ( VI ) 11/1/73 CHESS ( VI )

NAME
chess − the game of chess
SYNOPSIS
/usr/games/chess
DESCRIPTION
Chess is a computer program that plays class D chess. Moves may be given either in standard (descriptive)
notation or in algebraic notation. The symbol ‘+’ is used to specify check; ‘o-o’ and ‘o-o-o’ specify
castling. To play black, type ‘first’; to print the board, type an empty line.
Each move is echoed in the appropriate notation followed by the program’s reply.
FILES
/usr/lib/book opening ‘book’
DIAGNOSTICS
The most cryptic diagnostic is ‘eh?’ which means that the input was syntactically incorrect.
WARNING
Over-use of this program will cause it to go away.
BUGS
Pawns may be promoted only to queens.

-1-
-

COL ( VI ) 5/20/74 COL ( VI )

NAME
col − filter reverse line feeds
SYNOPSIS
col
DESCRIPTION
Col reads the standard input and writes the standard output. It performs the line overlays implied by re-
verse line feeds (ascii code ESC-7). Col is particularly useful for filtering multicolumn output made with
the ‘.rt’ command of nroff.
SEE ALSO
nroff (I)
BUGS
Can’t back up more than 102 lines.

-1-
-

CUBIC ( VI ) 11/1/73 CUBIC ( VI )

NAME
cubic − three dimensional tic-tac-toe
SYNOPSIS
/usr/games/cubic
DESCRIPTION
Cubic plays the game of three dimensional 4×4×4 tic-tac-toe. Moves are given by the three digits (each
1-4) specifying the coordinate of the square to be played.
WARNING
Too much playing of the game will cause it to disappear.
BUGS

-1-
-

FACTOR ( VI ) 1/15/73 FACTOR ( VI )

NAME
factor − discover prime factors of a number
SYNOPSIS
factor
DESCRIPTION
When factor is invoked without an argument, it waits for a number to be typed in. If you type in a positive
number less than 256 (about 7.2×1016 ) it will factor the number and print its prime factors; each one is
printed the proper number of times. Then it waits for another number. It exits if it encounters a zero or any
non-numeric character.
If factor is invoked with an argument, it factors the number as above and then exits.
Maximum time to factor is proportional to √n and occurs when n is prime or the square of a prime. It takes
1 minute to factor a prime near 1013.
DIAGNOSTICS
‘Ouch.’ for input out of range or for garbage input.
BUGS

-1-
-

FED ( VI ) 1/15/73 FED ( VI )

NAME
fed − edit form letter memory
SYNOPSIS
fed
DESCRIPTION
Fed is used to edit a form letter associative memory file, form.m, which consists of named strings. Com-
mands consist of single letters followed by a list of string names separated by a single space and ending
with a new line. The conventions of the Shell with respect to ‘*’ and ‘?’ hold for all commands but m. The
commands are:
e name ...
Fed writes the string whose name is name onto a temporary file and executes ed. On exit from the ed
the temporary file is copied back into the associative memory. Each argument is operated on separately.
Be sure to give an editor w command (without a filename) to rewrite fed’s temporary file before quitting
out of ed.
d [ name ... ]
deletes a string and its name from the memory. When called with no arguments d operates in a verbose
mode typing each string name and deleting only if a y is typed. A q response returns to fed’s command
level. Any other response does nothing.
m name1 name2 ...
(move) changes the name of name1 to name2 and removes previous string name2 if one exists. Several
pairs of arguments may be given. Literal strings are expected for the names.
n [ name ... ]
(names) lists the string names in the memory. If called with the optional arguments, it just lists those
requested.
p name ...
prints the contents of the strings with names given by the arguments.
q
returns to the system.
c[p][f]
checks the associative memory file for consistency and reports the number of free headers and blocks.
The optional arguments do the following:
p causes any unaccounted-for string to be printed.
f fixes broken memories by adding unaccounted-for headers to free storage and removing references
to released headers from associative memory.
FILES
/tmp/ftmp? temporary
form.m associative memory
SEE ALSO
form (VI), ed (I), sh (I)
WARNING
It is legal but unwise to have string names with blanks, ‘*’ or ‘?’ in them.
BUGS

-1-
-

FORM ( VI ) 6/15/72 FORM ( VI )

NAME
form − form letter generator
SYNOPSIS
form proto arg ...
DESCRIPTION
Form generates a form letter from a prototype letter, an associative memory, arguments and in a special
case, the current date.
If form is invoked with the proto argument x, the associative memory is searched for an entry with name x
and the contents filed under that name are used as the prototype. If the search fails, the message ‘[x]:’ is
typed on the console and whatever text is typed in from the console, terminated by two new lines, is used as
the prototype. If the prototype argument is missing, ‘{letter}’ is assumed.
Basically, form is a copy process from the prototype to the output file. If an element of the form [n] (where
n is a digit from 1 to 9) is encountered, the n-th arg is inserted in its place, and that argument is then res-
canned. If [0] is encountered, the current date is inserted. If the desired argument has not been given, a
message of the form ‘[n]:’ is typed. The response typed in then is used for that argument.
If an element of the form [name] or {name} is encountered, the name is looked up in the associative memo-
ry. If it is found, the contents of the memory under this name replaces the original element (again res-
canned). If the name is not found, a message of the form ‘[name]:’ is typed. The response typed in is used
for that element. The response is entered in the memory under the name if the name is enclosed in [ ]. The
response is not entered in the memory but is remembered for the duration of the letter if the name is en-
closed in {}. Brackets and braces may be nested.
In both of the above cases, the response is typed in by entering arbitrary text terminated by two new lines.
Only the first of the two new lines is passed with the text.
If one of the special characters [{]}\ is preceded by a \, it loses its special character.
If a file named ‘forma’ already exists in the user’s directory, ‘formb’ is used as the output file and so forth
to ‘formz’.
The file ‘form.m’ is created if none exists. Because form.m is operated on by the disc allocator, it should
only be changed by using fed, the form letter editor, or form.

FILES
form.m associative memory
form? output file (read only)
SEE ALSO
fed (VI), roff (I)
BUGS
An unbalanced ] or } acts as an end of file but may add a few strange entries to the associative memory.

-1-
-

GRAPH ( VI ) 2/20/75 GRAPH ( VI )

NAME
graph − draw a graph
SYNOPSIS
graph [ option ] ... | plotter
DESCRIPTION
Graph with no options takes pairs of numbers from the standard input as abscissas and ordinates of a graph.
The graph is written on the standard output to be piped to the plotter program for a particular device; see
plot (VI). These plotters exist: gsip, for the GSI and other Diablo terminals; tek, for the Tektronix 4014 ter-
minal; and vt0 for the on-line storage scope.
The following options are recognized, each as a separate argument.
a Supply abscissas automatically (they are missing from the input); spacing is given by the next argu-
ment, or is assumed to be 1 if next argument is not a number. A second optional argument is the
starting point for the automatic abscissa.
c Place character string given by next argument at each point.
d Omit connections between points. (Disconnect.)
gn Grid style:
n=0, no grid
n=1, axes only
n=2, complete grid (default).
s Save screen, don’t erase before plotting.
x Next 1 (or 2) arguments are lower (and upper) x limits.
y Next 1 (or 2) arguments are lower (and upper) y limits.
h Next argument is fraction of space for height
w Next argument is fraction of space for width.
r Next argument is fraction of space to move right before plotting.
u Next argument is fraction of space to move up before plotting.
Points are connected by straight line segments in the order they appear in input. If a specified lower limit
exceeds the upper limit, or if the automatic increment is negative, the graph is plotted upside down. Auto-
matic abscissas begin with the lower x limit, or with 0 if no limit is specified. Grid lines and automatically
determined limits fall on round values, however roundness may be subverted by giving an inappropriately
rounded lower limit. Plotting symbols specified by c are placed so that a small initial letter, such as + o x,
will fall approximately on the plotting point.
SEE ALSO
spline (VI), plot (VI)
BUGS
A limit of 1000 points is enforced silently.

-1-
-

GSI ( VI ) 3/20/74 GSI ( VI )

NAME
gsi − interpret extended character set on GSI terminal
SYNOPSIS
gsi
DESCRIPTION
Gsi interprets special characters understood by the Model 37 Teletype terminal and turns them into the es-
cape sequences understood by the GSI and other Diablo-based terminals. The things interpreted include
vertical motions and extended graphic characters. It is most often used in a pipeline like
neqn file ... | nroff | gsi
SEE ALSO
greek (V)
BUGS
Some funny characters can’t be correctly printed in column 1 because you can’t move to the left from there.

-1-
-

M6 ( VI ) 2/19/74 M6 ( VI )

NAME
m6 − general purpose macroprocessor
SYNOPSIS
m6 [ name ]
DESCRIPTION
M6 copies the standard input to the standard output, with substitutions for any macro calls that appear.
When a file name argument is given, that file is read before the standard input.
The processor is as described in the reference with these exceptions:
#def,arg1,arg2,arg3: causes arg1 to become a macro with defining text arg2 and (optional) built-in serial
number arg3.
#del,arg1: deletes the definition of macro arg1.
#end: is not implemented.
#list,arg1: sends the name of the macro designated by arg1 to the current destination without recognition
of any warning characters; arg1 is 1 for the most recently defined macro, 2 for the next most recent, and
so on. The name is taken to be empty when arg1 doesn’t make sense.
#warn,arg1,arg2: replaces the old warning character arg1 by the new warning character arg2.
#quote,arg1: sends the definition text of macro arg1 to the current destination without recognition of any
warning characters.
#serial,arg1: delivers the built-in serial number associated with macro arg1.
#source,arg1: is not implemented.
#trace,arg1: with arg1 = ‘1’ causes a reconstruction of each later call to be placed on the standard output
with a call level number; other values of arg1 turn tracing off.
The built-in ‘warn’ may be used to replace inconvenient warning characters. The example below replaces
‘#’ ‘:’ ‘<’ ‘>’ by ‘[’ ‘]’ ‘{’ ‘}’.
#warn,<#>,[:
[warn,<:>,]:
[warn,[substr,<<>>,1,1;,{]
[warn,[substr,{{>>,2,1;,}]
[now,{calls look like this}]
Every built-in function has a serial number, which specifies the action to be performed before the defining
text is expanded. The serial numbers are: 1 gt, 2 eq, 3 ge, 4 lt, 5 ne, 6 le, 7 seq, 8 sne, 9 add, 10 sub, 11
mpy, 12 div, 13 exp, 20 if, 21 def, 22 copy, 23 warn, 24 size, 25 substr, 26 go, 27 gobk, 28 del, 29 dnl, 32
quote, 33 serial, 34 list, 35 trace. Serial number 0 specifies no built-in action.
SEE ALSO
A. D. Hall, M6 Reference Manual. Computer Science Technical Report #2, Bell Laboratories, 1969.
DIAGNOSTICS
Various table overflows and ‘‘impossible’’ conditions result in comment and dump. There are no diagnos-
tics for poorly formed input.
AUTHOR
M. D. McIlroy
BUGS
Provision should be made to extend tables as needed, instead of wasting a big fixed core allocation. You
get what the PDP11 gives you for arithmetic.

-1-
-

MOO ( VI ) 11/1/73 MOO ( VI )

NAME
moo − guessing game
SYNOPSIS
/usr/games/moo
DESCRIPTION
Moo is a guessing game imported from England. The computer picks a number consisting of four distinct
decimal digits. The player guesses four distinct digits being scored on each guess. A ‘cow’ is a correct
digit in an incorrect position. A ‘bull’ is a correct digit in a correct position. The game continues until the
player guesses the number (a score of four bulls).
BUGS

-1-
-

PLOT ( VI ) 3/10/75 PLOT ( VI )

NAME
plot: tek, gsip, vt0 − graphics filters
SYNOPSIS
source | tek
source | gsip
source | vt0
DESCRIPTION
These commands produce graphical output on the Tektronix 4014 terminal, the GSI or other Diablo-mecha-
nism terminals, and the on-line storage scope respectively. They read the standard input to obtain plotting
instructions, which are usually generated by a program calling the graphics subroutines described in plot
(VII). Each instruction consists of an ASCII letter usually followed by binary information. A plotting co-
ordinate is transmitted as four bytes representing the x and y values; each value is a signed number trans-
mitted low-order byte first. The assumed plotting space is set by request. The instructions are taken from
m move: the next four bytes specify the coordinates of a point to move to. This is used before writing a la-
bel.
p point: the next four bytes specify the coordinates at which a point is drawn.
l line: the next eight bytes are taken as two pairs of coordinates specifying the endpoints of a line to be
drawn.
t label: the bytes up to a new-line are written as ASCII starting at the last point drawn or moved to.
a arc: the first four bytes specify the center, the next four specify the starting point, and the last four speci-
fy the end point of a circular arc. The least significant coordinate of the end point is used only to deter-
mine the quadrant. The arc is drawn counter-clockwise. This command is not necessarily implemented
on all (or even any) of the output devices.
c circle: The first four bytes specify the center of the circle, the next two the radius.
e erases the screen
f linemod: takes the following string as the type for all future lines. The types are ‘dotted,’ ‘solid,’ ‘long-
dashed,’ ‘shortdashed,’ and ‘dotdashed.’ This instruction is effective only with the Tektronix terminal.
d dotline: takes the first four bytes as the coordinates of the beginning of a dotted line. The next two are a
signed x-increment, and the next two are a word count. Following are the indicated number of byte-
pairs representing words. For each bit in this list of words a point is plotted which is visible if the bit is
‘1,’ invisible if not. Each point is offset rightward by the x-increment. The instruction is effective only
on the vt0 scope.

SEE ALSO
plot (VII), graph (VI)
BUGS

-1-
-

PRIMES ( VI ) 4/10/75 PRIMES ( VI )

NAME
primes − print all primes larger than somewhat
SYNOPSIS
primes
DESCRIPTION
When primes is invoked, it waits for a number to be typed in. If you type in a positive number less than 256
(about 7.2×1016 ) it will print all primes greater than or equal to this number.
DIAGNOSTICS
‘Ouch.’ for input out of range or for garbage input.
BUGS

-1-
-

QUIZ ( VI ) 9/7/74 QUIZ ( VI )

NAME
quiz − test your knowledge
SYNOPSIS
quiz [ −i file ] [ −t ] [ category1 category2 ]
DESCRIPTION
Quiz gives associative knowledge tests on various subjects. It asks items chosen from category1 and ex-
pects answers from category2. If no categories are specified, quiz gives instructions and lists the available
categories.
Quiz tells a correct answer whenever you type a bare newline. At the end of input, upon interrupt, or when
questions run out, quiz reports a score and terminates.
The −t flag specifies ‘tutorial’ mode, where missed questions are repeated later, and material is gradually
introduced as you learn.
The −i flag causes the named file to be substituted for the default index file. The lines of these files have the
syntax:
line = category newline  category ‘:’ line
category = alternate  category ‘|’ alternate
alternate = empty  alternate primary
primary = character  ‘[’ category ‘]’  option
option = ‘{’ category ‘}’
The first category on each line of an index file names an information file. The remaining categories specify
the order and contents of the data in each line of the information file. Information files have the same syn-
tax. Backslash ‘\’ is used as with sh (I) to quote syntactically significant characters or to insert transparent
newlines into a line. When either a question or its answer is empty, quiz will refrain from asking it.
FILES
/usr/lib/quiz/index
/usr/lib/quiz/*
BUGS

-1-
-

SKY ( VI ) 9/22/73 SKY ( VI )

NAME
sky − obtain ephemerides
SYNOPSIS
sky [ −l ]
DESCRIPTION
Sky predicts the apparent locations of the Sun, the Moon, the planets out to Saturn, stars of magnitude at
least 2.5, and certain other celestial objects. Sky reads the standard input to obtain a GMT time typed on
one line with blanks separating year, month number, day, hour, and minute; if the year is missing the cur-
rent year is used. If a blank line is typed the current time is used. The program prints the azimuth, ele-
vation, and magnitude of objects which are above the horizon at the ephemeris location of Murray Hill at
the indicated time. The ‘−l’ flag causes it to ask for another location.
Placing a ‘‘1’’ input after the minute entry causes the program to print out the Greenwich Sidereal Time at
the indicated moment and to print for each body its topographic right ascension and declination as well as
its azimuth and elevation. Also, instead of the magnitude, the semidiameter of the body, in seconds of arc,
is reported.
A ‘‘2’’ after the minute entry makes the coordinate system geocentric.
The effects of atmospheric extinction on magnitudes are not included; the brightest magnitudes of variable
stars are marked with ‘‘*’’.
For all bodies, the program takes into account precession and nutation of the equinox, annual (but not diur-
nal) aberration, diurnal parallax, and the proper motion of stars. In no case is refraction included.
The program takes into account perturbations of the Earth due to the Moon, Venus, Mars, and Jupiter. The
expected accuracies are: for the Sun and other stellar bodies a few tenths of seconds of arc; for the Moon
(on which particular care is lavished) likewise a few tenths of seconds. For the Sun, Moon and stars the ac-
curacy is sufficient to predict the circumstances of eclipses and occultations to within a few seconds of
time. The planets may be off by several minutes of arc.
There are lots of special options not described here, which do things like substituting named star catalogs,
smoothing nutation and aberration to aid generation of mean places of stars, and making conventional ad-
justments to the Moon to improve eclipse predictions.
For the most accurate use of the program it is necessary to know that it actually runs in Ephemeris time.
FILES
/usr/lib/startab, /usr/lib/moontab
SEE ALSO
azel (VI)
American Ephemeris and Nautical Almanac, for the appropriate years; also, the Explanatory Supplement to
the American Ephemeris and Nautical Almanac.
AUTHOR
R. Morris
BUGS

-1-
-

SNO ( VI ) 2/9/73 SNO ( VI )

NAME
sno − Snobol interpreter
SYNOPSIS
sno [ file ]
DESCRIPTION
Sno is a Snobol III (with slight differences) compiler and interpreter. Sno obtains input from the concatena-
tion of file and the standard input. All input through a statement containing the label ‘end’ is considered
program and is compiled. The rest is available to ‘syspit’.
Sno differs from Snobol III in the following ways.
There are no unanchored searches. To get the same effect:
a ** b unanchored search for b
a *x* b = x c unanchored assignment
There is no back referencing.
x = "abc"
a *x* x is an unanchored search for ‘abc’
Function declaration is different. The function declaration is done at compile time by the use of the label
‘define’. Thus there is no ability to define functions at run time and the use of the name ‘define’ is pre-
empted. There is also no provision for automatic variables other than the parameters. For example:
define f( )
or
define f(a,b,c)
All labels except ‘define’ (even ‘end’) must have a non-empty statement.
If ‘start’ is a label in the program, program execution will start there. If not, execution begins with the first
executable statement. ‘define’ is not an executable statement.
There are no builtin functions.
Parentheses for arithmetic are not needed. Normal precedence applies. Because of this, the arithmetic op-
erators ‘/’ and ‘*’ must be set off by space.
The right side of assignments must be non-empty.
Either ´ or " may be used for literal quotes.
The pseudo-variable ‘sysppt’ is not available.
SEE ALSO
Snobol III manual. (JACM; Vol. 11 No. 1; Jan 1964; pp 21)
BUGS

-1-
-

SPEAK ( VI ) 4/26/75 SPEAK ( VI )

NAME
speak − word to voice translator
SYNOPSIS
speak [ −efpsv ] [ vocabulary [ output ] ]
DESCRIPTION
Speak turns a stream of words into utterances and outputs them to a voice synthesizer, or to the specified
output. It has facilities for maintaining a vocabulary. It receives, from the standard input
− working lines: text of words separated by blanks
− phonetic lines: strings of phonemes for one word preceded and separated by commas. The phonemes
may be followed by comma-percent then a ‘replacement part’ − an ASCII string with no spaces. The
phonetic code is given in vs (V).
− empty lines
− command lines: beginning with !. The following command lines are recognized:
!r file replace coded vocabulary from file
!w file write coded vocabulary on file
!p print phonetics for working word
!l list vocabulary on standard output with phonetics
!c word copy phonetics from working word to specified word
!d print decomposition of working word into substrings
!f n turn off (or on) English preprocessing rule number n (see listing for meaning of n)
Each working line replaces its predecessor. Its first word is the ‘working word’. Each phonetic line re-
places the phonetics stored for the working word. In particular, a phonetic line of comma only deletes the
entry for the working word. Each working line, phonetic line or empty line causes the working line to be
uttered. The process terminates at the end of input.
Unknown words are pronounced by rules, and failing that, are spelled. For the builtin part of the rules, see
the reference. Spelling is done by taking each character of the word, prefixing it with ‘*’, and looking it up.
Unspellable words burp.
Words not found verbatim in the vocabulary are pronounced piecewise. First the word is bracketed by
sharps: ‘#...#’. The vocabulary is then searched for the longest fragment that matches the beginning of the
word. The phonetic part of the phonetic string is uttered, and the matched fragment is replaced by the re-
placement part of the phonetic string, if any. The process is repeated until the word is exhausted. A frag-
ment is entered into the vocabulary as a working word prefixed by ‘%’.
Speak is initialized with a coded vocabulary stored in file /usr/lib/speak.m. The vocabulary option substi-
tutes a different file for /usr/lib/speak.m. Other vocabularies, to be used with option −e, exist in
/usr/vs/latin.m and /usr/vs/polish.m.
A set of single letter options may appear in any order preceded by −. Their meanings are:
e suppress English preprocessing
f equivalent to ‘f1, f2,...’
p suppress pronunciation by rule
s suppress spelling
v suppress voice output
The following input will reconstitute a coded vocabulary, ‘speak.m’, from an ascii listing, ‘speak.v’, that
was created using !l.
(cat speak.v; echo !w speak.m)  speak −v /dev/null

FILES
/usr/lib/speak.m
SEE ALSO
M. D. McIlroy, ‘‘Synthetic English Speech by Rule,’’ Computing Science Technical Report #14, Bell Labo-
ratories, 1973

-1-
-

SPEAK ( VI ) 4/26/75 SPEAK ( VI )

vs (V), vs (IV)
BUGS
Excessively long words cause dumps.
Space is not reclaimed from changed entries; use !w and !r to effect reclamation.
!p doesn’t always work as advertised.

-2-
-

SPLINE ( VI ) 5/15/74 SPLINE ( VI )

NAME
spline − interpolate smooth curve
SYNOPSIS
spline [ option ] ...
DESCRIPTION
Spline takes pairs of numbers from the standard input as abcissas and ordinates of a function. It produces a
similar set, which is approximately equally spaced and includes the input set, on the standard output. The
cubic spline output (R. W. Hamming, Numerical Methods for Scientists and Engineers, 2nd ed., 349ff) has
two continuous derivatives, and sufficiently many points to look smooth when plotted, for example by plot
(I).
The following options are recognized, each as a separate argument.
a Supply abscissas automatically (they are missing from the input); spacing is given by the next argu-
ment, or is assumed to be 1 if next argument is not a number.
k The constant k used in the boundary value computation
y′′0 = ky′′1 , y′′n = ky′′n − 1
is set by the next argument. By default k = 0.
n Space output points so that approximately n points occur between the lower and upper x limits. (De-
fault n = 100.)
p Make output periodic, i.e. match derivatives at ends. First and last input values should normally
agree.
x Next 1 (or 2) arguments are lower (and upper) x limits. Normally these limits are calculated from the
data. Automatic abcissas start at lower limit (default 0).
SEE ALSO
plot (I)
AUTHOR
M. D. McIlroy
BUGS
A limit of 1000 input points is enforced silently.

-1-
-

TBL ( VI ) 2/2/75 TBL ( VI )

NAME
tbl − format tables for nroff or troff
SYNOPSIS
tbl [ files ] ...
DESCRIPTION
Tbl is an nroff (I) or troff(I) preprocessor for formatting tables. The input files are copied to the standard
output, except for lines between .TS and .TE command lines, which are assumed to describe tables and re-
formatted. The first line after .TS specifies the various columns: it consists of a list of column describers
separated by blanks or tabs. Each column describer is a character string made up of the letters ‘n’, ‘r’, ‘c’,
‘l’ and ‘s’, which mean:

c center within the column

r right-adjust
l left-adjust
n numerical adjustment: the units digits of numbers are aligned.
s span the previous entry over this column.
The column describer may be followed by an integer giving the number of spaces between this column and
the next; 3 is default. The describer ‘ccr5’ indicates that the first two lines in this column are centered; the
third and remaining lines are right-adjusted; and the column should be separated from the column to the
right by 5 spaces. Letting \t represent a tab (which must be typed as a genuine tab) the input
.TS
cccl sccn sscn
Household Population
Town\tHouseholds
\tNumber\tSize
Bedminster\t789\t3.26
Bernards Twp.\t3087\t3.74
Bernardsville\t2018\t3.30
Bound Brook\t3425\t3.04
Branchburg\t1644\t3.49
Bridgewater\t7897\t3.81
Far Hills\t240\t3.19
.TE
yields
Household Population
Town Households
Number Size
Bedminster 789 3.26
Bernards Twp. 3087 3.74
Bernardsville 2018 3.30
Bound Brook 3425 3.04
Branchburg 1644 3.49
Bridgewater 7897 3.81
Far Hills 240 3.19

If no arguments are given, tbl reads the standard input, so it may be used as a filter. When it is used with
eqn or neqn the tbl command should be first, to minimize the volume of data passed through pipes.
BUGS
No column describer may end with ‘s’.

-1-
-

TMG ( VI ) 10/21/72 TMG ( VI )

NAME
tmg − compiler-compiler
SYNOPSIS
tmg name
DESCRIPTION
Tmg produces a translator for the language whose parsing and translation rules are described in file name.t.
The new translator appears in a.out and may be used thus:
a.out input [ output ]
Except in rare cases input must be a randomly addressable file. If no output file is specified, the standard
output file is assumed.
FILES
name.s: assembly language version of name.t
/usr/lib/tmg: the compiler-compiler
/usr/lib/tmg[abc], /lib/libs.a: libraries
alloc.d: scratch file for table storage
SEE ALSO
A Manual for the Tmg Compiler-writing Language, internal memorandum.
DIAGNOSTICS
Syntactic errors result in "???" followed by the offending line.
Situations such as space overflow with which the Tmg processor or a Tmg-produced processor can not cope
result in a descriptive comment and a dump.
AUTHOR
M. D. McIlroy
BUGS
Footnote 1 of Section 9.2 of Tmg Manual is not enforced, causing trouble.
Restrictions (7.) against mixing bundling primitives should be lifted.
Certain hidden reserved words exist: gpar, classtab, trans, goto, alt, salt.
Octal digits include 8=10 and 9=11.

-1-
-

TTT ( VI ) 11/1/73 TTT ( VI )

NAME
ttt − the game of tic-tac-toe
SYNOPSIS
/usr/games/ttt
DESCRIPTION
Ttt is the X and O game popular in the first grade. This is a learning program that never makes the same
mistake twice.
Although it learns, it learns slowly. It must lose nearly 80 games to completely know the game.
FILES
/usr/games/ttt.k learning file
BUGS

-1-
-

UNITS ( VI ) 8/30/74 UNITS ( VI )

NAME
units − conversion program
SYNOPSIS
Units converts quantities expressed in various standard scales to their equivalents in other scales. It works
interactively in this fashion:
You have: inch
You want: cm
* 2.54000e+00
/ 3.93701e−01
A quantity is specified as a multiplicative combination of units optionally preceded by a numeric multiplier.
Powers are indicated by suffixed positive integers, division by the usual sign:
You have: 15 pounds force/in2
You want: atm
* 1.02069e+00
/ 9.79730e−01
Units only does multiplicative scale changes. Thus it can convert Kelvin to Rankine, but not Centigrade to
Fahrenheit. Most familiar units, abbreviations, and metric prefixes are recognized, together with a generous
leavening of exotica and a few constants of nature including:
pi ratio of circumference to diameter
c speed of light
e charge on an electron
g acceleration of gravity
force same as g
mole Avogadro’s number
water pressure head per unit height of water
au astronomical unit
‘Pound’ is a unit of mass. Compound names are run together, e.g. ‘lightyear’. British units that differ from
their US counterparts are prefixed thus: ‘brgallon’. For a complete list of units, ‘cat /usr/lib/units’.
FILES
/usr/lib/units
BUGS

-1-
-

WUMP ( VI ) 11/25/73 WUMP ( VI )

NAME
wump − the game of hunt-the-wumpus
SYNOPSIS
/usr/games/wump
DESCRIPTION
Wump plays the game of ‘‘Hunt the Wumpus.’’ A Wumpus is a creature that lives in a cave with several
rooms connected by tunnels. You wander among the rooms, trying to shoot the Wumpus with an arrow,
meanwhile avoiding being eaten by the Wumpus and falling into Bottomless Pits. There are also Super
Bats which are likely to pick you up and drop you in some random room.
The program asks various questions which you answer one per line; it will give a more detailed description
if you want.
This program is based on one described in People’s Computer Company, 2, 2 (November 1973).
BUGS
It will never replace Space War.

-1-
-

CR ( VII ) 1/4/75 CR ( VII )

NAME
crfork, crexit, crread, crwrite, crexch, crprior − coroutine scheme
SYNOPSIS
int crfork( [ stack, nwords ] )
int stack[];
int nwords;
crexit()
int crread(connector, buffer, nbytes)
int *connector[2];
char *buffer;
int nbytes;
crwrite(connector, buffer, nbytes)
int *connector[2];
char *buffer;
int nbytes;
crexch(conn1, conn2, i)
int *conn1[2], *conn2[2];
int i;
#define logical char *
crprior(p)
logical p;
DESCRIPTION
These functions are named by analogy to fork, exit, read, write (II). They establish and synchronize ‘corou-
tines’, which behave in many respects like a set of processes working in the same address space. The func-
tions live in /usr/lib/cr.a.
Coroutines are placed on queues to indicate their state of readiness. One coroutine is always distinguished
as ‘running’. Coroutines that are runnable but not running are registered on a ‘ready queue’. The head
member of the ready queue is started whenever no other coroutine is specifically caused to be running.
Each connector heads two queues: Connector[0] is the queue of unsatisfied crreads outstanding on the con-
nector. Connector[1] is the queue of crwrites. All queues must start empty, i.e. with heads set to zero.
Crfork is normally called with no arguments. It places the running coroutine at the head of the ready
queue, creates a new coroutine, and starts the new one running. Crfork returns immediately in the new
coroutine with value 0, and upon restarting of the old coroutine with value 1.
Crexit stops the running coroutine and does not place it in any queue.
Crread copies characters from the buffer of the crwrite at the head of the connector’s write queue to the
buffer of crread. If the write queue is empty, copying is delayed and the running coroutine is placed on the
read queue. The number of characters copied is the minimum of nbytes and the number of characters re-
maining in the write buffer, and is returned as the value of crread. After copying, the location of the write
buffer and the corresponding nbytes are updated appropriately. If zero characters remain, the coroutine of
the crwrite is moved to the head of the ready queue. If the write queue remains nonempty, the head mem-
ber of the read queue is moved to the head of the ready queue.
Crwrite queues the running coroutine on the connector’s write queue, and records the fact that nbytes (zero
or more) characters in the string buffer are available to crreads. If the read queue is not empty, its head
member is started running.
Crexch exchanges the read queues of connectors conn1 and conn2 if i=0; and it exchanges the write queues
if i=1. If a nonempty read queue that had been paired with an empty write queue becomes paired with a
nonempty write queue, crexch moves the head member of that read queue to the head of the ready queue.

-1-
-

CR ( VII ) 1/4/75 CR ( VII )

Crprior sets a priority on the running coroutine to control the queuing of crreads and crwrites. When
queued, the running coroutine will take its place before coroutines whose priorities exceed its own priority
and after others. Priorities are compared as logical, i.e. unsigned, quantities. Initially each coroutine’s pri-
ority is set as large as possible, so default queuing is FIFO.
Storage allocation. The old and new coroutine share the same activation record in the function that in-
voked crfork, so only one may return from the invoking function, and then only when the other has com-
pleted execution in that function.
The activation record for each function execution is dynamically allocated rather than stacked; a factor of 3
in running time overhead can result if function calls are very frequent. The overhead may be overcome by
providing a separate stack for each coroutine and dispensing with dynamic allocation. The base (lowest)
address and size of the new coroutine’s stack are supplied to crfork as optional arguments stack and nwords.
Stacked allocation and dynamic allocation cannot be mixed in one run. For stacked operation, obtain the
coroutine functions from /usr/lib/scr.a instead of /usr/lib/cr.a.
FILES
/usr/lib/cr.a
/usr/lib/scr.a
DIAGNOSTICS
‘rsave doesn’t work’ − an old C compilation has called ‘rsave’. It must be recompiled to work with the
coroutine scheme.
BUGS
Under /usr/lib/cr.a each function has just 12 words of anonymous stack for hard expressions and arguments
of further calls, regardless of actual need. There is no checking for stack overflow.
Under /usr/lib/scr.a stack overflow checking is not rigorous.

-2-
-

MS ( VII ) 11/6/74 MS ( VII )

NAME
ms − macros for formatting manuscripts
SYNOPSIS
nroff −ms [ options ] file ...
troff −ms [ options ] file ...
DESCRIPTION
This package of nroff and troff macro definitions provides a canned formatting facility for technical papers.
When producing 2-column output on a terminal, its output should be filtered through col (I).
The package supports three different formats: BTL technical memorandum with cover sheet, released paper
with cover sheet, and an abbreviated ‘debugging’ form without cover sheet.
The macro requests are defined in the attached Request Reference. Many nroff and troff requests are unsafe
in conjunction with this package, however the requests listed below may be used with impunity after the
first .PP.
.bp begin new page
.br break output line here
.sp n insert n spacing lines
.ls n (line spacing) n=1 single, n=2 double space
.na no alignment of right margin
Output of the eqn, neqn and tbl (I) preprocessors for equations and tables is acceptable as input.
FILES
/usr/lib/tmac.s
SEE ALSO
eqn (I), nroff (I), troff (I), tbl (VI)
BUGS

-1-
-

MS ( VII ) 11/6/74 MS ( VII )

REQUEST REFERENCE
Request Initial Cause
Form Value Break Explanation
.1C yes yes One column format on a new page.
.2C no yes Two column format.
.AB no yes Begin abstract.
.AE - yes End abstract.
.AI no yes Author’s institution follows. Suppressed in TM.
.AU x y no yes Author’s name follows. x is location and y is extension, ignored except in TM.
.B no no Boldface text follows.
.CS x... - yes Cover sheet info if TM format, suppressed otherwise. Arguments are number of text
pages, other pages, total pages, figures, tables, references.
.DA x nroff no ‘Date line’ at bottom of page is x. Default is today.
.DE - yes End displayed text. Implies .KE.
.DS x no yes Start of displayed text, to appear verbatim line-by-line. x=I for indented display (default),
x=L for left-justified on the page, x=C for centered. Implies .KS.
.EN - yes Space after equation produced by eqn or neqn.
.EQ x - yes Space before equation. Equation number is x.
.FE - yes End footnote.
.FS no no Start footnote. The note will be moved to the bottom of the page.
.HO - no ‘Bell Laboratories, Holmdel, New Jersey 07733’.
.I no no Italic text follows.
.IP x y no yes Start indented paragraph, with hanging tag x. Indentation is y ens (default 5).
.KE - yes End keep. Put kept text on next page if not enough room.
.KF no yes Start floating keep. If the kept text must be moved to the next page, float later text back
to this page.
.KS no yes Start keeping following text.
.LG no no Make letters larger.
.LP yes yes Start left-blocked paragraph.
.MH - no ‘Bell Laboratories, Murray Hill, New Jersey 07974’.
.ND troff no No date line at bottom of page.
.NH n - yes Same as .SH, with section number supplied automatically. Numbers are multilevel, like
1.2.3, where n tells what level is wanted (default is 1).
.NL yes no Make letters normal size.
.OK - yes ‘Other keywords’ for TM cover sheet follow.
.PP no yes Begin paragraph. First line indented.
.R yes no Roman text follows.
.RE - yes End relative indent level.
.RP no - Cover sheet and first page for released paper. Must precede other requests.
.RS - yes Start level of relative indentation. Following .IP’s measured from current indentation.
.SG x no yes Insert signature(s) of author(s), ignored except in TM. x is the reference line (initials of
author and typist).
.SH - yes Section head follows, font automatically bold.
.SM no no Make letters smaller.
.TL no yes Title follows.
.TM x y z no - BTL TM cover sheet and first page, x=TM number, y=(quoted list of) case number(s),
z=file number. Must precede other requests.
.WH - no ‘Bell Laboratories, Whippany, New Jersey 07981’.

-2-
-

PLOT ( VII ) 2/25/75 PLOT ( VII )

NAME
plot: openpl et al. − graphics interface
SYNOPSIS
openpl( )
erase( )
label(s)
char s[ ];
line(x1, y1, x2, y2)
circle(x, y, r)
arc(x, y, x0, y0, x1, y1)
dot(x, y, dx, n, pattern)
int pattern[ ];
move(x, y)
point(x, y)
linemod(s)
char s[ ];
space(x0, y0, x1, y1)
closepl( )

DESCRIPTION
These subroutines generate graphic output in a relatively device-independent manner. See plot (VI) for a
description of the meaning of the subroutines.
There are four libraries containing these routines, one that produces general graphics commands on the
standard output, and one each for the vt0 storage scope, the Diablo plotting terminal and the Tektronix 4014
terminal. Openpl must be used before any of the others to open the device for writing. Closepl flushes the
output.
FILES
/usr/lib/plot.a produces output for plotting filters
/usr/lib/vt0.a produces output on vt0 storage scope
/usr/lib/gsip.a produces output on Diablo terminal
/usr/lib/tek.a produces output for the Tektronix 4014 terminal
SEE ALSO
plot (VI), graph (VI)
BUGS

-1-
-

SALLOC ( VII ) 6/15/72 SALLOC ( VII )

NAME
salloc − string allocation and manipulation
SYNOPSIS
(get size in r0)
jsr pc,allocate
(header address in r1)
(get source header address in r0,
destination header address in r1)
jsr pc,copy
jsr pc,wc
(all following routines assume r1 contains header address)
jsr pc,release
(get character in r0)
jsr pc,putchar
jsr pc,lookchar
(character in r0)
jsr pc,getchar
(character in r0)
(get character in r0)
jsr pc,alterchar
(get position in r0)
jsr pc,seekchar
jsr pc,backspace
(character in r0)
(get word in r0)
jsr pc,putword
jsr pc,lookword
(word in r0)
jsr pc,getword
(word in r0)
(get word in r0)
jsr pc,alterword
jsr pc,backword
(word in r0)
jsr pc,length
(length in r0)
jsr pc,position
(position in r0)
jsr pc,rewind
jsr pc,create
jsr pc,fsfile
jsr pc,zero
DESCRIPTION
This package is a complete set of routines for dealing with almost arbitrary length strings of words and
bytes. It lives in /lib/libs.a. The strings are stored on a disk file, so the sum of their lengths can be consid-
erably larger than the available core. A small buffer cache makes for reasonable speed.

-1-
-

SALLOC ( VII ) 6/15/72 SALLOC ( VII )

For each string there is a header of four words, namely a write pointer, a read pointer and pointers to the be-
ginning and end of the block containing the string. Initially the read and write pointers point to the begin-
ning of the string. All routines that refer to a string require the header address in r1. Unless the string is
destroyed by the call, upon return r1 will point to the same string, although the string may have grown to
the extent that it had to be be moved.
Allocate obtains a string of the requested size and returns a pointer to its header in r1.
Release releases a string back to free storage.
Putchar and putword write a byte or word respectively into the string and advance the write pointer.
Lookchar and lookword read a byte or word respectively from the string but do not advance the read point-
er.
Getchar and getword read a byte or word respectively from the string and advance the read pointer.
Alterchar and alterword write a byte or word respectively into the string where the read pointer is pointing
and advance the read pointer.
Backspace and backword read the last byte or word written and decrement the write pointer.
All write operations will automatically get a larger block if the current block is exceeded. All read opera-
tions return with the error bit set if attempting to read beyond the write pointer.
Seekchar moves the read pointer to the offset specified in r0.
Length returns the current length of the string (beginning pointer to write pointer) in r0.
Position returns the current offset of the read pointer in r0.
Rewind moves the read pointer to the beginning of the string.
Create returns the read and write pointers to the beginning of the string.
Fsfile moves the read pointer to the current position of the write pointer.
Zero zeros the whole string and sets the write pointer to the beginning of the string.
Copy copies the string whose header pointer is in r0 to the string whose header pointer is in r1. Care should
be taken in using the copy instruction since r1 will be changed if the contents of the source string is bigger
than the destination string.
Wc forces the contents of the internal buffers and the header blocks to be written on disc.
An in-core version of this allocator exists in dc (I), and a permanent-file version exists in form and fed (VI).
FILES
/lib/libs.a library, accessed by ld ... -ls
alloc.d temporary file for string storage
SEE ALSO
alloc (III)
DIAGNOSTICS
‘error in copy’ − disk write error encountered in copy.
‘error in allocator’ − routine called with bad header pointer.
‘cannot open output file’ − temp file alloc.d cannot be created or opened.
‘out of space’ − no sufficiently large block or no header is available for a new or growing block.
BUGS

-2-
-

AC ( VIII ) 2/20/74 AC ( VIII )

NAME
ac − login accounting
SYNOPSIS
ac [ −w wtmp ] [ −p ] [ −d ] people
DESCRIPTION
Ac produces a printout giving connect time for each user who has logged in during the life of the current
wtmp file. A total is also produced. −w is used to specify an alternate wtmp file. −p prints individual to-
tals; without this option, only totals are printed. −d causes a printout for each midnight to midnight period.
Any people will limit the printout to only the specified login names. If no wtmp file is given,
/usr/adm/wtmp is used.
The accounting file /usr/adm/wtmp is maintained by init and login. Neither of these programs creates the
file, so if it does not exist no connect-time accounting is done. To start accounting, it should be created
with length 0. On the other hand if the file is left undisturbed it will grow without bound, so periodically
any information desired should be collected and the file truncated.
FILES
/usr/adm/wtmp
SEE ALSO
init (VIII), login (I), wtmp (V).
BUGS

-1-
-

BOOT PROCEDURES ( VIII ) 2/11/75 BOOT PROCEDURES ( VIII )

NAME
boot procedures − UNIX startup
DESCRIPTION
How to start UNIX. UNIX is started by placing it in core at location zero and transferring to zero. Since
the system is not reenterable, it is necessary to read it in from disk or tape.
The tp command places a bootstrap program on the otherwise unused block zero of the tape. The DECtape
version of this program is called tboot, the magtape version mboot. If tboot or mboot is read into location
zero and executed there, it will type ‘=’ on the console, read in a tp entry name, load that entry into core,
and transfer to zero. Thus one way to run UNIX is to maintain the system code on a tape using tp. Cau-
tion: the file /usr/mdec/tboot (DECtape) or /usr/mdec/mboot (magtape) must be present when the tape is
made! When a boot is required, execute (somehow) a program which reads in and jumps to the first block
of the tape. In response to the ‘=’ prompt, type the entry name of the system on the tape (we use plain
‘unix’). It is strongly recommended that a current version of the system be maintained in this way, even if
it is usually booted from disk.
The standard DEC ROM which loads DECtape is sufficient to read in tboot, but the magtape ROM loads
block one, not zero. If no suitable ROM is available, magtape and DECtape programs are presented below
which may be manually placed in core and executed.
The system can also be booted from a disk file with the aid of the uboot program. When read into location
0 and executed, uboot reads a single character (either p or k for RP or RK, both drive 0) to specify which
device is to be searched. Then it reads a UNIX pathname from the console, finds the corresponding file on
the given device, loads that file into core location zero, and transfers to it. Uboot operates under very se-
vere space constraints. It supplies no prompts, except that it echoes a carriage return and line feed after the
p or k. No diagnostic is provided if the indicated file cannot be found, nor is there any means of correcting
typographical errors in the file name except to start the program over. If it fails to find the file, however, it
jumps back to its start, so another try can be attempted, starting again with the p or k. Notice that uboot
will only load a file from drive 0, and the file system it searches must start at the beginning of the disk.
Uboot itself usually resides in the otherwise unused block 0 of the disk, so it can be loaded by ROM pro-
gram; mkfs can be used to put it there when the file system is created. It can also be loaded from a tp tape
as described above.
The switches. The console switches play an important role in the use and especially the booting of UNIX.
During operation, the console switches are examined 60 times per second, and the contents of the address
specified by the switches are displayed in the display register. (This is not true on the 11/40 since there is
no display register on that machine.) If the switch address is even, the address is interpreted in kernel (sys-
tem) space; if odd, the rounded-down address is interpreted in the current user space.
If any diagnostics are produced by the system, they are printed on the console only if the switches are non-
zero. Thus it is wise to have a non-zero value in the switches at all times.
During the startup of the system, the init program (VIII) reads the switches and will come up single-user if
the switches are set to 173030.
It is unwise to have a non-existent address in the switches. This causes a bus error in the system (displayed
as 177777) at the rate of 60 times per second. If there is a transfer of more than 16ms duration on a device
with a data rate faster than the bus error timeout (about 10µs) then a permanent disk non-existent-memory
error will occur.
ROM programs. Here are some programs which are suitable for installing in read-only memories, or for
manual keying into core if no ROM is present. Each program is position-independent but should be placed
well above location 0 so it will not be overwritten. Each reads a block from the beginning of a device into
core location zero. The octal words constituting the program are listed on the left.
DECtape (drive 0) from endzone:
012700 mov $tcba,r0
177346
010040 mov r0,−(r0) / use tc addr for wc
012710 mov $3,(r0) / read bn forward
000003

-1-
-

BOOT PROCEDURES ( VIII ) 2/11/75 BOOT PROCEDURES ( VIII )

105710 1: tstb (r0) / wait for ready

002376 bge 1b
112710 movb $5,(r0) / read (forward)
000005
000777 br . / loop; now halt and start at 0
DECtape (drive 0) with search:
012700 1: mov $tcba,r0
177346
010040 mov r0,−(r0) / use tc addr for wc
012740 mov $4003,−(r0) / read bn reverse
004003
005710 2: tst (r0)
002376 bge 2b / wait for error
005760 tst −2(r0) / loop if not end zone
177776
002365 bge 1b
012710 mov $3,(r0) / read bn forward
000003
105710 2: tstb (r0) / wait for ready
002376 bge 2b
112710 movb $5,(r0) / read (forward)
000005
105710 2: tstb (r0) / wait for ready
002376 bge 2b
005007 clr pc / transfer to zero
Caution: both of these DECtape programs will (literally) blow a fuse if 2 drives are dialed to zero.
Magtape from load point:
012700 mov $mtcma,r0
172526
010040 mov r0,−(r0) / usr mt addr for wc
012740 mov $60003,−(r0) / read 9−track
060003
000777 br . / loop; now halt and start at 0
RK (drive 0):
012700 mov $rkda,r0
177412
005040 clr −(r0) / rkda cleared by start
010040 mov r0,−(r0)
012740 mov $5,−(r0)
000005
105710 1: tstb (r0)
002376 bge 1b
005007 clr pc
RP (drive 0)
012700 mov $rpmr,r0
176726
005040 clr −(r0)
005040 clr −(r0)
005040 clr −(r0)
010040 mov r0,−(r0)
012740 mov $5,−(r0)
000005
105710 1: tstb (r0)
002376 bge 1b
005007 clr pc

-2-
-

BOOT PROCEDURES ( VIII ) 2/11/75 BOOT PROCEDURES ( VIII )

FILES
/unix − UNIX code
/usr/mdec/mboot − tp magtape bootstrap
/usr/mdec/tboot − tp DECtape bootstrap
/usr/mdec/uboot − file system bootstrap
SEE ALSO
tp (I), init (VIII)

-3-
-

CHGRP ( VIII ) 2/8/75 CHGRP ( VIII )

NAME
chgrp − change group
SYNOPSIS
chgrp group file ...
DESCRIPTION
The group-ID of the files is changed to group. The group may be either a decimal GID or a group name
found in the group-ID file.
Only the super-user is allowed to change the group of a file, in order to simplify as yet unimplemented ac-
counting procedures.
SEE ALSO
chown (VIII)
FILES
/etc/group
BUGS

-1-
-

CHOWN ( VIII ) 2/8/75 CHOWN ( VIII )

NAME
chown − change owner
SYNOPSIS
chown owner file ...
DESCRIPTION
The user-ID of the files is changed to owner. The owner may be either a decimal UID or a login name
found in the password file.
Only the super-user is allowed to change the owner of a file, in order to simplify as yet unimplemented ac-
counting procedures.
FILES
/etc/passwd
SEE ALSO
chgrp (VIII)
BUGS

-1-
-

CLRI ( VIII ) 10/31/73 CLRI ( VIII )

NAME
clri − clear i-node
SYNOPSIS
clri i-number [ filesystem ]
DESCRIPTION
Clri writes zeros on the 32 bytes occupied by the i-node numbered i-number. If the file system argument is
given, the i-node resides on the given device, otherwise on a default file system. The file system argument
must be a special file name referring to a device containing a file system. After clri, any blocks in the af-
fected file will show up as ‘‘missing’’ in an icheck of of the file system.
Read and write permission is required on the specified file system device. The i-node becomes allocatable.
The primary purpose of this routine is to remove a file which for some reason appears in no directory. If it
is used to zap an i-node which does appear in a directory, care should be taken to track down the entry and
remove it. Otherwise, when the i-node is reallocated to some new file, the old entry will still point to that
file. At that point removing the old entry will destroy the new file. The new entry will again point to an un-
allocated i-node, so the whole cycle is likely to be repeated again and again.
BUGS
Whatever the default file system is, it is likely to be wrong. Specify the file system explicitly.
If the file is open, clri is likely to be ineffective.

-1-
-

CRASH ( VIII ) 2/12/75 CRASH ( VIII )

NAME
crash − what to do when the system crashes
DESCRIPTION
This section gives at least a few clues about how to proceed if the system crashes. It can’t pretend to be
complete.
How to bring it back up. If the reason for the crash is not evident (see below for guidance on ‘evident’)
you may want to try to dump the system if you feel up to debugging. At the moment a dump can be taken
only on magtape. With a tape mounted and ready, stop the machine, load address 44, and start. This should
write a copy of all of core on the tape with an EOF mark. Caution: Any error is taken to mean the end of
core has been reached. This means that you must be sure the ring is in, the tape is ready, and the tape is
clean and new. If the dump fails, you can try again, but some of the registers will be lost. See below for
what to do with the tape.
In restarting after a crash, always bring up the system single-user. This is accomplished by following the
directions in boot procedures (VIII) as modified for your particular installation; a single-user system is indi-
cated by having a particular value in the switches (173030 unless you’ve changed init) as the system starts
executing. When it is running, perform a dcheck and icheck (VIII) on all file systems which could have
been in use at the time of the crash. If any serious file system problems are found, they should be repaired.
When you are satisfied with the health of your disks, check and set the date if necessary, then come up mul-
ti-user. This is most easily accomplished by changing the single-user value in the switches to something
else, then logging out by typing an EOT.
To even boot UNIX at all, three files (and the directories leading to them) must be intact. First, the initializa-
tion program /etc/init must be present and executable. If it is not, the CPU will loop in user mode at loca-
tion 6. For init to work correctly, /dev/tty8 and /bin/sh must be present. If either does not exist, the symp-
tom is best described as thrashing. Init will go into a fork/exec loop trying to create a Shell with proper
standard input and output.
If you cannot get the system to boot, a runnable system must be obtained from a backup medium. The root
file system may then be doctored as a mounted file system as described below. If there are any problems
with the root file system, it is probably prudent to go to a backup system to avoid working on a mounted file
system.
Repairing disks. The first rule to keep in mind is that an addled disk should be treated gently; it shouldn’t
be mounted unless necessary, and if it is very valuable yet in quite bad shape, perhaps it should be dumped
before trying surgery on it. This is an area where experience and informed courage count for much.
The problems reported by icheck typically fall into two kinds. There can be problems with the free list: du-
plicates in the free list, or free blocks also in files. These can be cured easily with an icheck −s. If the same
block appears in more than one file or if a file contains bad blocks, the files should be deleted, and the free
list reconstructed. The best way to delete such a file is to use clri (VIII), then remove its directory entries.
If any of the affected files is really precious, you can try to copy it to another device first.
Dcheck may report files which have more directory entries than links. Such situations are potentially dan-
gerous; clri discusses a special case of the problem. All the directory entries for the file should be removed.
If on the other hand there are more links than directory entries, there is no danger of spreading infection,
but merely some disk space that is lost for use. It is sufficient to copy the file (if it has any entries and is
useful) then use clri on its inode and remove any directory entries that do exist.
Finally, there may be inodes reported by dcheck that have 0 links and 0 entries. These occur on the root de-
vice when the system is stopped with pipes open, and on other file systems when the system stops with files
that have been deleted while still open. A clri will free the inode, and an icheck -s will recover any missing
blocks.
Why did it crash? UNIX types a message on the console typewriter when it voluntarily crashes. Here is
the current list of such messages, with enough information to provide a hope at least of the remedy. The
message has the form ‘panic: ...’, possibly accompanied by other information. Left unstated in all cases is
the possibility that hardware or software error produced the message in some unexpected way.

-1-
-

CRASH ( VIII ) 2/12/75 CRASH ( VIII )

blkdev
The getblk routine was called with a nonexistent major device as argument. Definitely hardware or
software error.
devtab
Null device table entry for the major device used as argument to getblk. Definitely hardware or soft-
ware error.
iinit
An I/O error reading the super-block for the root file system during initialization.
out of inodes
A mounted file system has no more i-nodes when creating a file. Sorry, the device isn’t available; the
icheck should tell you.
no fs
A device has disappeared from the mounted-device table. Definitely hardware or software error.
no imt
Like ‘no fs’, but produced elsewhere.
no inodes
The in-core inode table is full. Try increasing NINODE in param.h. Shouldn’t be a panic, just a user
error.
no clock
During initialization, neither the line nor programmable clock was found to exist.
swap error
An unrecoverable I/O error during a swap. Really shouldn’t be a panic, but it is hard to fix.
unlink − iget
The directory containing a file being deleted can’t be found. Hardware or software.
out of swap space
A program needs to be swapped out, and there is no more swap space. It has to be increased. This
really shouldn’t be a panic, but there is no easy fix.
out of text
A pure procedure program is being executed, and the table for such things is full. This shouldn’t be a
panic.
trap
An unexpected trap has occurred within the system. This is accompanied by three numbers: a ‘ka6’,
which is the contents of the segmentation register for the area in which the system’s stack is kept;
‘aps’, which is the location where the hardware stored the program status word during the trap; and a
‘trap type’ which encodes which trap occurred. The trap types are:
0 bus error
1 illegal instruction
2 BPT/trace
3 IOT
4 power fail
5 EMT
6 recursive system call (TRAP instruction)
7 11/70 cache parity, or programmed interrupt
10 floating point trap
11 segmentation violation
In some of these cases it is possible for octal 20 to be added into the trap type; this indicates that the proces-
sor was in user mode when the trap occurred. If you wish to examine the stack after such a trap, either
dump the system, or use the console switches to examine core; the required address mapping is described
below.

-2-
-

CRASH ( VIII ) 2/12/75 CRASH ( VIII )

Interpreting dumps. All file system problems should be taken care of before attempting to look at dumps.
The dump should be read into the file /usr/sys/core; cp (I) will do. At this point, you should execute ps
−alxk and who to print the process table and the users who were on at the time of the crash. You should
dump ( od (I)) the first 30 bytes of /usr/sys/core. Starting at location 4, the registers R0, R1, R2, R3, R4,
R5, SP and KDSA6 (KISA6 for 11/40s) are stored. If the dump had to be restarted, R0 will not be correct.
Next, take the value of KA6 (location 22(8) in the dump) multiplied by 100(8) and dump 1000(8) bytes
starting from there. This is the per-process data associated with the process running at the time of the
crash. Relabel the addresses 140000 to 141776. R5 is C’s frame or display pointer. Stored at (R5) is the
old R5 pointing to the previous stack frame. At (R5)+2 is the saved PC of the calling procedure. Trace this
calling chain until you obtain an R5 value of 141756, which is where the user’s R5 is stored. If the chain is
broken, you have to look for a plausible R5, PC pair and continue from there. Each PC should be looked
up in the system’s name list using db (I) and its ‘:’ command, to get a reverse calling order. In most cases
this procedure will give an idea of what is wrong. A more complete discussion of system debugging is im-
possible here.
SEE ALSO
clri, icheck, dcheck, boot procedures (VIII)
BUGS

-3-
-

CRON ( VIII ) 10/25/74 CRON ( VIII )

NAME
cron − clock daemon
SYNOPSIS
/etc/cron
DESCRIPTION
Cron executes commands at specified dates and times according to the instructions in the file
/usr/lib/crontab. Since cron never exits, it should only be executed once. This is best done by running cron
from the initialization process through the file /etc/rc; see init (VIII).
Crontab consists of lines of six fields each. The fields are separated by spaces or tabs. The first five are in-
teger patterns to specify the minute (0-59), hour (0-23), day of the month (1-31), month of the year (1-12),
and day of the week (1-7 with 1=monday). Each of these patterns may contain a number in the range
above; two numbers separated by a minus meaning a range inclusive; a list of numbers separated by com-
mas meaning any of the numbers; or an asterisk meaning all legal values. The sixth field is a string that is
executed by the Shell at the specified times. A percent character in this field is translated to a new-line
character. Only the first line (up to a % or end of line) of the command field is executed by the Shell. The
other lines are made available to the command as standard input.
Crontab is examined by cron every hour. Thus it could take up to an hour for entries to become effective.
If it receives a hangup signal, however, the table is examined immediately; so ‘kill −1 ...’ can be used.
FILES
/usr/lib/crontab
SEE ALSO
init(VIII), sh(I), kill (I)
DIAGNOSTICS
None − illegal lines in crontab are ignored.
BUGS
A more efficient algorithm could be used. The overhead in running cron is about one percent of the ma-
chine, exclusive of any commands executed.

-1-
-

DCHECK ( VIII ) 2/8/75 DCHECK ( VIII )

NAME
dcheck − file system directory consistency check
SYNOPSIS
dcheck [ −i numbers ] [ filesystem ]
DESCRIPTION
Dcheck reads the directories in a file system and compares the link-count in each i-node with the number of
directory entries by which it is referenced. If the file system is not specified, a set of default file systems is
checked.
The −i flag is followed by a list of i-numbers; when one of those i-numbers turns up in a directory, the num-
ber, the i-number of the directory, and the name of the entry are reported.
The program is fastest if the raw version of the special file is used, since the i-list is read in large chunks.
FILES
Currently, /dev/rrk2 and /dev/rrp0 are the default file systems.
DIAGNOSTICS
When a file turns up for which the link-count and the number of directory entries disagree, the relevant
facts are reported. Allocated files which have 0 link-count and no entries are also listed. The only danger-
ous situation occurs when there are more entries than links; if entries are removed, so the link-count drops
to 0, the remaining entries point to thin air. They should be removed. When there are more links than en-
tries, or there is an allocated file with neither links nor entries, some disk space may be lost but the situation
will not degenerate.
SEE ALSO
icheck (VIII), fs (V), clri (VIII), ncheck (VIII)
BUGS
Since dcheck is inherently two-pass in nature, extraneous diagnostics may be produced if applied to active
file systems.

-1-
-

DF ( VIII ) 1/20/73 DF ( VIII )

NAME
df − disk free
SYNOPSIS
df [ filesystem ]
DESCRIPTION
Df prints out the number of free blocks available on a file system. If the file system is unspecified, the free
space on all of the normally mounted file systems is printed.
FILES
/dev/rf?, /dev/rk?, /dev/rp?
SEE ALSO
icheck (VIII)
BUGS

-1-
-

DPD ( VIII ) 3/15/72 DPD ( VIII )

NAME
dpd − data phone daemon
SYNOPSIS
/etc/dpd
DESCRIPTION
Dpd is the 201 data phone daemon. It is designed to submit jobs to the Honeywell 6070 computer via the
GRTS interface.
Dpd uses the directory /usr/dpd. The file lock in that directory is used to prevent two daemons from be-
coming active. After the daemon has successfully set the lock, it forks and the main path exits, thus spawn-
ing the daemon. The directory is scanned for files beginning with df. Each such file is submitted as a job.
Each line of a job file must begin with a key character to specify what to do with the remainder of the line.
S directs dpd to generate a unique snumb card. This card is generated by incrementing the first word of
the file /usr/dpd/snumb and converting that to three-digit octal concatenated with the station ID.
L specifies that the remainder of the line is to be sent as a literal.
B specifies that the rest of the line is a file name. That file is to be sent as binary cards.
F is the same as B except a form feed is prepended to the file.
U specifies that the rest of the line is a file name. After the job has been transmitted, the file is un-
linked.
M is followed by a user ID; after the job is sent, the snumb number and the first line of information in
the file is mailed to the user to verify the sending of the job.
Any error encountered will cause the daemon to drop the call, wait up to 20 minutes and start over. This
means that an improperly constructed df file may cause the same job to be submitted every 20 minutes.
While waiting, the daemon checks to see that the lock file still exists. If it is gone, the daemon will exit.
FILES
/dev/dn0, /dev/dp0, /usr/dpd/*
SEE ALSO
opr (I)

-1-
-

DUMP ( VIII ) 11/24/73 DUMP ( VIII )

NAME
dump − incremental file system dump
SYNOPSIS
dump [ key [ arguments ] filesystem ]
DESCRIPTION
Dump makes an incremental file system dump on magtape of all files changed after a certain date. The key
argument specifies the date and other options about the dump. Key consists of characters from the set abc-
fiu0hds.
a Normally files larger than 1000 blocks are not incrementally dump; this flag forces them to be
dumped.
b The next argument is taken to be the maximum size of the dump tape in blocks (see s).
c If the tape overflows, increment the last character of its name and continue on that drive. (Normally
it asks you to change tapes.)
f Place the dump on the next argument file instead of the tape.
i the dump date is taken from the entry in the file /etc/dtab corresponding to the last time this file sys-
tem was dumped with the -u option.
u the date just prior to this dump is written on /etc/dtab upon successful completion of this dump. This
file contains a date for every file system dumped with this option.
0 the dump date is taken as the epoch (beginning of time). Thus this option causes an entire file system
dump to be taken.
h the dump date is some number of hours before the current date. The number of hours is taken from
the next argument in arguments.
d the dump date is some number of days before the current date. The number of days is taken from the
next argument in arguments.
s the size of the dump tape is specified in feet. The number of feet is taken from the next argument in
arguments. It is assumed that there are 9 standard UNIX records per foot. When the specified size is
reached, the dump will wait for reels to be changed. The default size is 2200 feet.
If no arguments are given, the key is assumed to be i and the file system is assumed to be /dev/rp0.
Full dumps should be taken on quiet file systems as follows:
dump 0u /dev/rp0
ncheck /dev/rp0
The ncheck will come in handy in case it is necessary to restore individual files from this dump. Incremen-
tal dumps should then be taken when desired by:
dump
When the incremental dumps get cumbersome, a new complete dump should be taken. In this way, a re-
store requires loading of the complete dump tape and only the latest incremental tape.
DIAGNOSTICS
If the dump requires more than one tape, it will ask you to change tapes. Reply with a new-line when this
has been done. If the first block on the new tape is not writable, e.g. because you forgot the write ring, you
get a chance to fix it. Generally, however, read or write failures are fatal.
FILES
/dev/mt0 magtape
/dev/rp0 default file system
/etc/dtab

-1-
-

DUMP ( VIII ) 11/24/73 DUMP ( VIII )

SEE ALSO
restor (VIII), ncheck (VIII), dump (V)
BUGS

-2-
-

GETTY ( VIII ) 2/11/75 GETTY ( VIII )

NAME
getty − set typewriter mode
SYNOPSIS
/etc/getty [ char ]
DESCRIPTION
Getty is invoked by init (VIII) immediately after a typewriter is opened following a dial-up. It reads the us-
er’s name and invokes the login command (I) with the name as argument. While reading the name getty at-
tempts to adapt the system to the speed and type of terminal being used.
Init calls getty with an argument specified by the ttys file entry for the typewriter line. Arguments other
than ‘0’ can be used to make getty treat the line specially. Normally, it sets the speed of the interface to 300
baud, specifies that raw mode is to be used (break on every character), that echo is to be suppressed, and ei-
ther parity allowed. It types the ‘‘login:’’ message, which includes the characters which put the Terminet
300 terminal into full-duplex and return the GSI terminal to non-graphic mode. Then the user’s name is
read, a character at a time. If a null character is received, it is assumed to be the result of the user pushing
the ‘‘break’’ (‘‘interrupt’’) key. The speed is then changed to 150 baud and the ‘‘login:’’ is typed again, this
time including the character sequence which puts a Teletype 37 into full-duplex. If a subsequent null char-
acter is received, the speed is changed back to 300 baud.
The user’s name is terminated by a new-line or carriage-return character. The latter results in the system
being set to treat carriage returns appropriately (see stty (II)).
The user’s name is scanned to see if it contains any lower-case alphabetic characters; if not, and if the name
is nonempty, the system is told to map any future upper-case characters into the corresponding lower-case
characters.
Finally, login is called with the user’s name as argument.
SEE ALSO
init (VIII), login (I), stty (II), ttys (V)
BUGS

-1-
-

GLOB ( VIII ) 9/19/73 GLOB ( VIII )

NAME
glob − generate command arguments
SYNOPSIS
/etc/glob command [ arguments ]
DESCRIPTION
Glob is used to expand arguments to the shell containing ‘‘*’’, ‘‘[’’, or ‘‘?’’. It is passed the argument list
containing the metacharacters; glob expands the list and calls the indicated command. The actions of glob
are detailed in the Shell writeup.
SEE
sh (I)
BUGS

-1-
-

ICHECK ( VIII ) 2/9/75 ICHECK ( VIII )

NAME
icheck − file system storage consistency check
SYNOPSIS
icheck [ −s ] [ −b numbers ] [ filesystem ]
DESCRIPTION
Icheck examines a file system, builds a bit map of used blocks, and compares this bit map against the free
list maintained on the file system. If the file system is not specified, a set of default file systems is checked.
The normal output of icheck includes a report of
The number of blocks missing; i.e. not in any file nor in the free list,
The number of special files,
The total number of files,
The number of large and huge files,
The number of directories,
The number of indirect blocks, and the number of double-indirect blocks in huge files,
The number of blocks used in files,
The number of free blocks.
The −s flag causes icheck to ignore the actual free list and reconstruct a new one by rewriting the super-
block of the file system. The file system should be dismounted while this is done; if this is not possible (for
example if the root file system has to be salvaged) care should be taken that the system is quiescent and that
it is rebooted immediately afterwards so that the old, bad in-core copy of the super-block will not continue
to be used. Notice also that the words in the super-block which indicate the size of the free list and of the i-
list are believed. If the super-block has been curdled these words will have to be patched. The −s flag caus-
es the normal output reports to be suppressed.
Following the −b flag is a list of block numbers; whenever any of the named blocks turns up in a file, a di-
agnostic is produced.
Icheck is faster if the raw version of the special file is used, since it reads the i-list many blocks at a time.
FILES
Currently, /dev/rrk2 and /dev/rrp0 are the default file systems.
SEE ALSO
dcheck (VIII), ncheck (VIII), fs (V), clri (VIII), restor(VIII)
DIAGNOSTICS
For duplicate blocks and bad blocks (which lie outside the file system) icheck announces the difficulty, the
i-number, and the kind of block involved. If a read error is encountered, the block number of the bad block
is printed and icheck considers it to contain 0. ‘‘Bad freeblock’’ means that a block number outside the
available space was encountered in the free list. ‘‘n dups in free’’ means that n blocks were found in the
free list which duplicate blocks either in some file or in the earlier part of the free list.
BUGS
Since icheck is inherently two-pass in nature, extraneous diagnostics may be produced if applied to active
file systems.
It believes even preposterous super-blocks and consequently can get core images.

-1-
-

INIT ( VIII ) 2/22/74 INIT ( VIII )

NAME
init − process control initialization
SYNOPSIS
/etc/init
DESCRIPTION
Init is invoked inside UNIX as the last step in the boot procedure. Generally its role is to create a process
for each typewriter on which a user may log in.
First, init checks to see if the console switches contain 173030. (This number is likely to vary between sys-
tems.) If so, the console typewriter /dev/tty8 is opened for reading and writing and the Shell is invoked im-
mediately. This feature is used to bring up a single-user system. When the system is brought up in this
way, the getty and login routines mentioned below and described elsewhere are not used. If the Shell termi-
nates, init starts over looking for the console switch setting.
Otherwise, init invokes a Shell, with input taken from the file /etc/rc. This command file performs house-
keeping like removing temporary files, mounting file systems, and starting daemons.
Then init reads the file /etc/ttys and forks several times to create a process for each typewriter specified in
the file. Each of these processes opens the appropriate typewriter for reading and writing. These channels
thus receive file descriptors 0 and 1, the standard input and output. Opening the typewriter will usually in-
volve a delay, since the open is not completed until someone is dialed up and carrier established on the
channel. Then /etc/getty is called with argument as specified by the last character of the ttys file line. Getty
reads the user’s name and invokes login (q.v.) to log in the user and execute the Shell.
Ultimately the Shell will terminate because of an end-of-file either typed explicitly or generated as a result
of hanging up. The main path of init, which has been waiting for such an event, wakes up and removes the
appropriate entry from the file utmp, which records current users, and makes an entry in /usr/adm/wtmp,
which maintains a history of logins and logouts. Then the appropriate typewriter is reopened and getty is
reinvoked.
Init catches the hangup signal (signal #1) and interprets it to mean that the switches should be examined as
in a reboot: if they indicate a multi-user system, the /etc/ttys file is read again. The Shell process on each
line which used to be active in ttys but is no longer there is terminated; a new process is created for each
added line; lines unchanged in the file are undisturbed. Thus it is possible to drop or add phone lines with-
out rebooting the system by changing the ttys file and sending a hangup signal to the init process: use ‘‘kill
−1 1.’’
FILES
/dev/tty?, /etc/utmp, /usr/adm/wtmp, /etc/ttys, /etc/rc
SEE ALSO
login (I), kill (I), sh (I), ttys (V), getty (VIII)

-1-
-

LPD ( VIII ) 6/1/74 LPD ( VIII )

NAME
lpd − line printer daemon
SYNOPSIS
/etc/lpd
DESCRIPTION
Lpd is the line printer daemon (spool area handler) invoked by opr. It uses the directory /usr/lpd. The file
lock in that directory is used to prevent two daemons from becoming active simultaneously. After the dae-
mon has successfully set the lock, it scans the directory for files beginning with ‘‘df.’’ Lines in each df file
specify files to be printed in the same way as is done by the data-phone daemon dpd (VIII).
FILES
/usr/lpd/* spool area
/dev/lp printer
SEE ALSO
dpd (VIII), opr (I)
BUGS

-1-
-

MKFS ( VIII ) 11/1/73 MKFS ( VIII )

NAME
mkfs − construct a file system
SYNOPSIS
/etc/mkfs special proto
DESCRIPTION
Mkfs constructs a file system by writing on the special file special according to the directions found in the
prototype file proto. The prototype file contains tokens separated by spaces or new lines. The first token is
the name of a file to be copied onto block zero as the bootstrap program (see boot procedures (VIII)). The
second token is a number specifying the size of the created file system. Typically it will be the number of
blocks on the device, perhaps diminished by space for swapping. The next token is the i-list size in blocks
(remember there are 16 i-nodes per block). The next set of tokens comprise the specification for the root
file. File specifications consist of tokens giving the mode, the user-id, the group id, and the initial contents
of the file. The syntax of the contents field depends on the mode.
The mode token for a file is a 6 character string. The first character specifies the type of the file. (The char-
acters −bcd specify regular, block special, character special and directory files respectively.) The second
character of the type is either u or − to specify set-user-id mode or not. The third is g or − for the set-
group-id mode. The rest of the mode is a three digit octal number giving the owner, group, and other read,
write, execute permissions (see chmod (I)).
Two decimal number tokens come after the mode; they specify the user and group ID’s of the owner of the
file.
If the file is a regular file, the next token is a pathname whence the contents and size are copied.
If the file is a block or character special file, two decimal number tokens follow which give the major and
minor device numbers.
If the file is a directory, mkfs makes the entries . and .. and then reads a list of names and (recursively) file
specifications for the entries in the directory. The scan is terminated with the token $.
If the prototype file cannot be opened and its name consists of a string of digits, mkfs builds a file system
with a single empty directory on it. The size of the file system is the value of proto interpreted as a decimal
number. The i-list size is the file system size divided by 43 plus the size divided by 1000. (This corre-
sponds to an average size of three blocks per file for a 4000 block file system and six blocks per file at
40,000.) The boot program is left uninitialized.
A sample prototype specification follows:
/usr/mdec/uboot
4872 55
d−−777 3 1
usr d−−777 3 1
sh −−−755 3 1 /bin/sh
ken d−−755 6 1
$
b0 b−−644 3 1 0 0
c0 c−−644 3 1 0 0
$
$
SEE ALSO
file system (V), directory (V), boot procedures (VIII)
BUGS
It is not possible to initialize a file larger than 64K bytes.
The size of the file system is restricted to 64K blocks.
There should be some way to specify links.

-1-
-

MKNOD ( VIII ) 2/11/75 MKNOD ( VIII )

NAME
mknod − build special file
SYNOPSIS
/etc/mknod name [ c ] [ b ] major minor
DESCRIPTION
Mknod makes a directory entry and corresponding i-node for a special file. The first argument is the name
of the entry. The second is b if the special file is block-type (disks, tape) or c if it is character-type (other
devices). The last two arguments are numbers specifying the major device type and the minor device (e.g.
unit, drive, or line number).
The assignment of major device numbers is specific to each system. They have to be dug out of the system
source file conf.c.
SEE ALSO
mknod (II)
BUGS

-1-
-

MOUNT ( VIII ) 10/31/73 MOUNT ( VIII )

NAME
mount − mount file system
SYNOPSIS
/etc/mount special file [ −r ]
DESCRIPTION
Mount announces to the system that a removable file system is present on the device corresponding to spe-
cial file special (which must refer to a disk or possibly DECtape). The file must exist already; it becomes
the name of the root of the newly mounted file system.
Mount maintains a table of mounted devices; if invoked without an argument it prints the table.
The optional last argument indicates that the file is to be mounted read-only. Physically write-protected and
magnetic tape file systems must be mounted in this way or errors will occur when access times are updated,
whether or not any explicit write is attempted.
SEE ALSO
mount (II), mtab (V), umount (VIII)
BUGS
Mounting file systems full of garbage will crash the system.

-1-
-

NCHECK ( VIII ) 2/8/75 NCHECK ( VIII )

NAME
ncheck − generate names from i-numbers
SYNOPSIS
ncheck [ −i numbers ] [ −a ] [ filesystem ]
DESCRIPTION
Ncheck with no argument generates a pathname vs. i-number list of all files on a set of default file systems.
The −i flag reduces the report to only those files whose i-numbers follow. −a allows printing of the names
‘.’ and ‘..’, which are ordinarily suppressed. A file system may be specified.
The full report is in no useful order, and probably should be sorted.
SEE ALSO
dcheck (VIII), icheck (VIII), sort (I)
BUGS

-1-
-

RESTOR ( VIII ) 11/24/73 RESTOR ( VIII )

NAME
restor − incremental file system restore
SYNOPSIS
restor key [ arguments ]
DESCRIPTION
Restor is used to read magtapes dumped with the dump command. The key argument specifies what is to be
done. Key is a character from the set trxw.
t The date that the tape was made and the date that was specified in the dump command are printed. A
list of all of the i-numbers on the tape is also given.
r The tape is read and loaded into the file system specified in arguments. This should not be done
lightly (see below).
x Each file on the tape is individually extracted into a file whose name is the file’s i-number. If there
are arguments, they are interpreted as i-numbers and only they are extracted.
c If the tape overflows, increment the last character of its name and continue on that drive. (Normally
it asks you to change tapes.)
f Read the dump from the next argument file instead of the tape.
i All read and checksum errors are reported, but will not cause termination.
w In conjunction with the x option, before each file is extracted, its i-number is typed out. To extract
this file, you must respond with y.
The x option is used to retrieve individual files. If the i-number of the desired file is not known, it can be
discovered by following the file system directory search algorithm. First retrieve the root directory whose
i-number is 1. List this file with ls −fi 1. This will give names and i-numbers of sub-directories. Iterating,
any file may be retrieved.
The r option should only be used to restore a complete dump tape onto a clear file system or to restore an
incremental dump tape onto this. Thus
/etc/mkfs /dev/rp0 40600
restor r /dev/rp0
is a typical sequence to restore a complete dump. Another restor can be done to get an incremental dump
in on top of this.
A dump followed by a mkfs and a restor is used to change the size of a file system.
FILES
/dev/mt0
SEE ALSO
ls (I), dump (VIII), mkfs (VIII), clri (VIII)
DIAGNOSTICS
There are various diagnostics involved with reading the tape and writing the disk. There are also diagnos-
tics if the i-list or the free list of the file system is not large enough to hold the dump.
If the dump extends over more than one tape, it may ask you to change tapes. Reply with a new-line when
the next tape has been mounted.
BUGS
There is redundant information on the tape that could be used in case of tape reading problems. Unfortu-
nately, restor’s approach is to exit if anything is wrong.

-1-
-

SA ( VIII ) 6/1/74 SA ( VIII )

NAME
sa − Shell accounting
SYNOPSIS
sa [ −abcjlnrstuv ] [ file ]
DESCRIPTION
When a user logs in, if the Shell is able to open the file /usr/adm/sha, then as each command completes the
Shell writes at the end of this file the name of the command, the user, system, and real time consumed, and
the user ID. Sa reports on, cleans up, and generally maintains this and other accounting files. To turn ac-
counting on and off, the accounting file must be created or destroyed externally.
Sa is able to condense the information in /usr/adm/sha into a summary file /usr/adm/sht which contains a
count of the number of times each command was called and the time resources consumed. This condensa-
tion is desirable because on a large system sha can grow by 100 blocks per day. The summary file is read
before the accounting file, so the reports include all available information.
If a file name is given as the last argument, that file will be treated as the accounting file; sha is the default.
There are zillions of options:

a Place all command names containing unprintable characters and those used only once under the name
‘‘***other.’’
b Sort output by sum of user and system time divided by number of calls. Default sort is by sum of user
and system times.
c Besides total user, system, and real time for each command print percentage of total time over all com-
mands.
j Instead of total minutes time for each category, give seconds per call.
l Separate system and user time; normally they are combined.
n Sort by number of calls.
r Reverse order of sort.
s Merge accounting file into summary file /usr/adm/sht when done.
t For each command report ratio of real time to the sum of user and system times.
u Superseding all other flags, print for each command in the accounting file the day of the year, time, day
of the week, user ID and command name.
v If the next character is a digit n, then type the name of each command used n times or fewer. Await a
reply from the typewriter; if it begins with ‘‘y’’, add the command to the category ‘‘**junk**.’’ This is
used to strip out garbage.
FILES
/usr/adm/sha accounting
/usr/adm/sht summary
SEE ALSO
ac (VIII)
BUGS

-1-
-

SU ( VIII ) 10/31/73 SU ( VIII )

NAME
su − become privileged user
SYNOPSIS
su
DESCRIPTION
Su allows one to become the super-user, who has all sorts of marvelous (and correspondingly dangerous)
powers. In order for su to do its magic, the user must supply a password. If the password is correct, su will
execute the Shell with the UID set to that of the super-user. To restore normal UID privileges, type an end-
of-file to the super-user Shell.
The password demanded is that of the entry ‘‘root’’ in the system’s password file.
To remind the super-user of his responsibilities, the Shell substitutes ‘#’ for its usual prompt ‘%’.
SEE ALSO
sh (I)

-1-
-

SYNC ( VIII ) 11/1/73 SYNC ( VIII )

NAME
sync − update the super block
SYNOPSIS
sync
DESCRIPTION
Sync executes the sync system primitive. If the system is to be stopped, sync must be called to insure file
system integrity. See sync (II) for details.
SEE ALSO
sync (II)
BUGS

-1-
-

UMOUNT ( VIII ) 10/31/73 UMOUNT ( VIII )

NAME
umount − dismount file system
SYNOPSIS
/etc/umount special
DESCRIPTION
Umount announces to the system that the removable file system previously mounted on special file special
is to be removed.
SEE ALSO
mount (VIII), umount (II), mtab (V)
FILES
/etc/mtab mounted device table
DIAGNOSTICS
It complains if the special file is not mounted or if it is busy. The file system is busy if there is an open file
on it or if someone has his current directory there.
BUGS

-1-
-

UPDATE ( VIII ) 11/1/73 UPDATE ( VIII )

NAME
update − periodically update the super block
SYNOPSIS
update
DESCRIPTION
Update is a program that executes the sync primitive every 30 seconds. This insures that the file system is
fairly up to date in case of a crash. This command should not be executed directly, but should be executed
out of the initialization shell command file. See sync (II) for details.
SEE ALSO
sync (II), init (VIII)
BUGS
With update running, if the CPU is halted just as the sync is executed, a file system can be damaged. This
is partially due to DEC hardware that writes zeros when NPR requests fail. A fix would be to have sync
temporarily increment the system time by at least 30 seconds to trigger the execution of update. This
would give 30 seconds grace to halt the CPU.

-1-
-

WALL ( VIII ) 4/10/75 WALL ( VIII )

NAME
wall − write to all users
SYNOPSIS
/etc/wall
DESCRIPTION
Wall reads its standard input until an end-of-file. It then sends this message to all currently logged in users
preceded by ‘‘Broadcast Message ...’’. It is used to warn all users, typically prior to shutting down the sys-
tem.
The sender should be super-user to override any protections the users may have invoked.
FILES
/dev/tty?
SEE ALSO
mesg (I), write (I)
DIAGNOSTICS
‘‘Cannot send to ...’’ when the open on a user’s tty file fails.
BUGS

-1-
DOCUMENTS FOR USE WITH THE
UNIX TIME-SHARING SYSTEM

Sixth Edition

The enclosed UNIX documentation is supplied

in accordance with the Software Agreement
you have with the Western Electric Company.
 CONTENTS

1. Setting Up UNIX − Sixth Edition

2. The UNIX Time-Sharing System

3. C Reference Manual

4. Programming in C − A Tutorial

5. UNIX Assembler Reference Manual

6. A Tutorial Introduction to the ED Text Editor

7. UNIX for Beginners

8. RATFOR − A Preprocessor for a Rational Fortran

9. YACC − Yet Another Compiler-Compiler

10. NROFF Users’ Manual

11. The UNIX I/O System

12. A Manual for the Tmg Compiler-writing Language

13. On the Security of UNIX

14. The M6 Macro Processor

15. A System for Typesetting Mathematics

16. DC − An Interactive Desk Calculator

17. BC − An Arbitrary Precision Desk-Calculator Language

18. The Portable C Library (on UNIX)

19. UNIX Summary

 SETTING UP UNIX − Sixth Edition

Enclosed are:
1. ‘UNIX Programmer’s Manual,’ Sixth Edition.
2. Documents with the following titles:
Setting Up UNIX − Sixth Edition
The UNIX Time-Sharing System
C Reference Manual
Programming in C − A Tutorial
UNIX Assembler Reference Manual
A Tutorial Introduction to the ED Text Editor
UNIX for Beginners
RATFOR − A Preprocessor for a Rational Fortran
YACC − Yet Another Compiler-Compiler
NROFF Users’ Manual
The UNIX I/O System
A Manual for the Tmg Compiler-writing Language
On the Security of UNIX
The M6 Macro Processor
A System for Typesetting Mathematics
DC − An Interactive Desk Calculator
BC − An Arbitrary Precision Desk-Calculator Language
The Portable C Library (on UNIX)
UNIX Summary

3. The UNIX software on magtape or disk pack.

If you are set up to do it, it might be a good idea immediately to make a copy of the disk or tape to
guard against disaster. The tape contains 12100 512-byte records followed by a single file mark; only the
first 4000 512-byte blocks on the disk are significant.
The system as distributed corresponds to three fairly full RK packs. The first contains the binary ver-
sion of all programs, and the source for the operating system itself; the second contains all remaining
source programs; the third contains manuals intended to be printed using the formatting programs roff or
nroff. The ‘binary’ disk is enough to run the system, but you will almost certainly want to modify some
source programs.

Making a Disk From Tape

If your system is on magtape, perform the following bootstrap procedure to obtain a disk with the
binaries.
1. Mount magtape on drive 0 at load point.
2. Mount formatted disk pack on drive 0.
3. Key in and execute at 100000
 -2-

TU10 TU16
012700 (to be added)
172526
010040
012740
060003
000777

The tape should move and the CPU loop. (The TU10 code is not the DEC bulk ROM for tape; it
reads block 0, not block 1.)
4. Halt and restart the CPU at 0. The tape should rewind. The console should type ‘=’.
5. Copy the magtape to disk by the following. This assumes TU10 and RK05; see 6 below for other
devices. The machine’s printouts are shown in italic (the ‘=’ signs should be considered italic). Ter-
minate each line you type by carriage return or line-feed.
= tmrk
disk offset
0
tape offset
100 (See 6 below)
count
1 (The tape should move)
= tmrk
disk offset
1
tape offset
101 (See 7 below)
count
3999 (The tape moves lots more)
=
To explain: the tmrk program copies tape to disk with the given offsets and counts. Its first use copies
a bootstrap program to disk block 0; the second use copies the file system itself onto the disk. You
may get back to ‘=’ level by starting at 137000.
6. If you have TU16 tape say ‘htrk’ instead of ‘tmrk’ in the above example. If you have an RP03 disk,
say ‘tmrp’ or ‘htrp’, and use a 99 instead of 100 tape offset. If you have an RP04 disk, use ‘tmhp’ or
‘hthp’ instead or ‘tmrk’, and use a 98 instead of 100 tape offset. The different offsets load bootstrap
programs appropriate to the disk they will live on.
7. This procedure generates the ‘binary’ disk; the ‘source’ disk may be generated on another RK pack
by using a tape offset of 4101 instead of 101. The ‘document’ disk is at offset 8101 instead of 101.
Unless you have only a single RK drive, it is probably wise to wait on generating these disks. Better
tools are available using UNIX itself.

Booting UNIX
Once the UNIX ‘binary’ disk is obtained, the system is booted by keying in and executing one of the
following programs at 100000. These programs correspond to the DEC bulk ROMs for disks, since they
read in and execute block 0 at location 0.
 -3-

RK05 RP03 RP04

012700 012700 (to be added)
177414 176726
005040 005040
005040 005040
010040 005040
012740 010040
000005 012740
105710 000005
002376 105710
005007 002376
005007
Now follow the indicated dialog, where ‘@’ and ‘#’ are prompts:
@ rkunix (or ‘rpunix’ or ‘hpunix’)
mem = xxx
login: root
#
The mem message gives the memory available to user programs in .1K units. Most of the UNIX software
will run with 120 (for 12K words), but some things require much more.
UNIX is now running, and the ‘UNIX Programmer’s manual’ applies; references below of the form
X-Y mean the subsection named X in section Y of the manual. The ‘#’ is the prompt from the UNIX Shell,
and indicates you are logged in as the super-user. The only valid user names are ‘root’ and ‘bin’. The root
is the super-user and bin is the owner of nearly every file in the file system.
Before UNIX is turned up completely, a few configuration dependent exercises must be performed.
At this point, it would be wise to read all of the manuals and to augment this reading with hand to hand
combat. It might be instructive to examine the Shell run files mentioned below.

Reconfiguration
The UNIX system running is configured to run on an 11/40 with the given disk, TU10 magtape and
TU56 DECtape. This is almost certainly not the correct configuration. Print (cat-I) the file /usr/sys/run.
This file is a set of Shell commands that will completely recompile the system source, install it in the cor-
rect libraries and build the three configurations for rk, rp and hp.
Using the Shell file as a guide, compile (cc-I) and rename (mv-I) the configuration program
‘mkconf ’. Run the configuration program and type into it a list of the controllers on your system. Choose
from:
 -4-

pc (PC11)
lp (LP11)
rf (RS11)
hs (RS03/RS04)
tc (TU56)
rk (RK03/RK05)
tm (TU10)
rp (RP03)
hp (RP04)
ht (TU16)
dc* (DC11)
kl* (KL11/DL11-ABC)
dl* (DL11-E)
dp (DP11)
dn (DN11)
dh (DH11)
dhdm (DM11-BB)
The devices marked with * should be preceded by a number specifying how many. (The console typewriter
is automatically included; don’t count it in the kl specification.) Mkconf will generate the two files l.s (trap
vectors) and c.c (configuration table). Take a careful look at l.s to make sure that all the devices that you
have are assembled in the correct interrupt vectors. If your configuration is non-standard, you will have to
modify l.s to fit your configuration.
In the run Shell file, the 11/45 code is commented out. If you have an 11/45 you must also edit (ed-I)
the file /usr/sys/conf/m45.s to set the assembly flag fpp to reflect if you have the FP11-B floating point unit.
The main difference between an 11/40 and an 11/45 (or 11/70) system is that in the former instruction
restart after a segmentation violation caused by overflowing a user stack must be handled by software,
while in the latter machines there is hardware help. As mentioned above, the 11/45 and 11/70 systems
include conditionally-enabled code to save the status of the floating point unit when switching users. The
source for such things is in one of the two files m40.s and m45.s.
Another difference is that in 11/45 and 11/70 systems the instruction and data spaces are separated
inside UNIX itself. Since the layout of addresses in the system is somewhat peculiar, and not directly sup-
ported by the link-editor ld, the sysfix program has to be run before the loaded output file can be booted.
There are certain magic numbers and configuration parameters imbedded in various device drivers
that you may want to change. The device addresses of each device are defined in each driver. In case you
have any non-standard device addresses, just change the address and recompile. (The device drivers are in
the directory /usr/sys/dmr.)
The DC11 driver is set to run 14 lines. This can be changed in dc.c.
The DH11 driver will only handle a single DH with a full complement of 16 lines. If you have less,
you may want to edit dh.c.
The DN11 driver will handle 3 DN’s. Edit dn.c.
The DP11 driver can only handle a single DP. This cannot be easily changed.
The KL/DL driver is set up to run a single DL11-A, -B, or -C (the console) and no DL11-E’s. To
change this, edit kl.c to have NKL11 reflect the total number of DL11-ABC’s and NDL11 to reflect the
number of DL11-E’s. So far as the driver is concerned, the difference between the devices is their
addresses.
The line printer driver is set up to print the 96 character set on 80 column paper (LP11-H) with
indenting. Edit lp.c.
All of the disk and tape drivers (rf.c, rk.c, rp.c, tm.c, tc.c, hs.c, hp.c, ht.c) are set up to run 8 drives
and should not need to be changed. The big disk drivers (rp.c and hp.c) have partition tables in them which
you may want to experiment with.
 -5-

After all the corrections have been made, use /usr/sys/run as a guide to recompile the changed
drivers, install them in /usr/sys/lib2 and to assemble the trap vectors (l.s), configuration table (c.c) and
machine language assist (m40.s or m45.s). After all this, link edit the objects (ld-I) and if you have an
11/45, sysfix the result. The final object file (a.out) should be renamed /unix and booted. See Boot Proce-
dures-VIII for a discussion of booting. (Note: remember, before booting, always perform a sync-VIII to
force delayed output to the disk.)

Special Files
Next you must put in all of the special files in the directory /dev using mknod-VIII. Print the config-
uration file c.c created above. This is the major device switch of each device class (block and character).
There is one line for each device configured in your system and a null line for place holding for those
devices not configured. The block special devices are put in first by executing the following generic com-
mand for each disk or tape drive. (Note that some of these files already exist in the directory /dev. Exam-
ine each file with ls-I with −l flag to see if the file should be removed.)
/etc/mknod /dev/NAME b MAJOR MINOR
The NAME is selected from the following list:
c.c NAME device
rf rf0 RS fixed head disk
tc tap0 TU56 DECtape
rk rk0 RK03 RK05 moving head disk
tm mt0 TU10 TU16 magtape
rp rp0 RP moving head disk
hs hs0 RS03 RS04 fixed head disk
hp hp0 RP04 moving head disk
The major device number is selected by counting the line number (from zero) of the device’s entry in the
block configuration table. Thus the first entry in the table bdevsw would be major device zero.
The minor device is the drive number, unit number or partition as described under each device in sec-
tion IV. The last digit of the name (all given as 0 in the table above) should reflect the minor device num-
ber. For tapes where the unit is dial selectable, a special file may be made for each possible selection.
The same goes for the character devices. Here the names are arbitrary except that devices meant to
be used for teletype access should be named /dev/ttyX, where X is any character. The files tty8 (console),
mem, kmem, null are already correctly configured.
The disk and magtape drivers provide a ‘raw’ interface to the device which provides direct transmis-
sion between the user’s core and the device and allows reading or writing large records. The raw device
counts as a character device, and should have the name of the corresponding standard block special file
with ‘r’ prepended. Thus the raw magtape files would be called /dev/rmtX.
When all the special files have been created, care should be taken to change the access modes
(chmod-I) on these files to appropriate values.

The Source Disk

You should now extract the source disk. This can be done as described above or the UNIX command
dd-I may be used. The disk image begins at block 4100 on the tape, so the command
dd if=/dev/mt0 of=/dev/rk1 count=4000 skip=4100
might be used to extract the disk to RK drive 1.
This disk should be mounted (mount-VIII) on /usr/source; it contains directories of source code. In
each directory is a Shell file run that will recompile all the source in the directory. These run files should be
consulted whenever you need to recompile.
 -6-

Floating Point
UNIX only supports the 11/45 FP11-B floating point unit. For machines without this hardware, there
is a user subroutine available that will catch illegal instruction traps and interpret floating point operations.
(See fptrap-III.) The system as delivered has this code included in all commands that have floating point.
This code is never used if the FP hardware is available and therefore does not need to be changed. The
penalty is a little bit of disk space and loading time for the few floating commands.
The C compiler in /usr/source/c probably should be changed if floating point is available. The fpp
flag in c0t.s should be set and C should be recompiled and reloaded and installed. This allows floating
point C programs to be compiled without the −f flag and prevents the floating point interpreter from getting
into new floating programs. (See /usr/source/c/run.)

Time Conversion
If your machine is not in the Eastern time zone, you must edit (ed-I) the subroutine
/usr/source/s4/ctime.c to reflect your local time. The variable ‘timezone’ should be changed to reflect the
time difference between local time and GMT. For EST, this is 5*60*60; for PST it would be 8*60*60.
This routine also contains the names of the standard and Daylight Savings time zone; so ‘EST’ and ‘EDT’
might be changed to ‘PST’ and ‘PDT’ respectively. Notice that these two names are in upper case and
escapes may be needed (tty-IV). Finally, there is a ‘daylight’ flag; when it is 1 it causes the time to shift to
Daylight Savings automatically between the last Sundays in April and October (or other algorithms in 1974
and 1975). Normally this will not have to be reset. After ctime.c has been edited it should be compiled and
installed in its library. (See /usr/source/s4/run.) Then you should (at your leisure) recompile and reinstall
all programs performing time conversion. These include: (in s1) date, dump, ls, cron, (in s2) mail, pr,
restor, who, sa and tp.

Disk Layout
If there are to be more file systems mounted than just the root, use mkfs-VIII to create the new file
system and put its mounting in the file /etc/rc (see init-VIII and mount-VIII). (You might look at /etc/rc
anyway to see what has been provided for you.)
There are two considerations in deciding how to adjust the arrangement of things on your disks: the
most important is making sure there is adequate space for what is required; secondarily, throughput should
be maximized. The RK disk (or its image) as distributed has 4000 blocks for file storage, and the remain-
der of the disk (872 blocks) is set aside for swap space. In our own system, which allows 14 simultaneous
users, this amount of swap space is not quite enough, so we use 1872 blocks for this purpose; it is large
enough so running out of swap space never occurs.
Many common system programs (C, the editor, the assembler etc.) create intermediate files in the
/tmp directory, so the file system where this is stored also should be made large enough to accommodate
most high-water marks. In an idle state, we have about 900 free blocks on the file system where /tmp
resides, and hit the bottom every few days or so. (This causes a momentary disruption, but not a crash, as
swap-space runout does.) All the programs that create files in /tmp try to take care to delete them, but most
are not immune to events like being hung up upon, and can leave dregs. The directory should be examined
every so often and the old files deleted.
Exhaustion of user-file space is certain to occur now and then; the only mechanisms for controlling
this phenomenon are occasional use of du-I and threatening messages of the day and personal letters.
The efficiency with which UNIX is able to use the CPU is largely dictated by the configuration of
disk controllers. For general time-sharing applications, the best strategy is to try to split user files, the root
directory (including the /tmp directory) and the swap area among three controllers. In our own system, for
example, we have user files on an RP, the root on an RF fixed-head disk, and swap on an RK. This is best
for us since the RK has a faster transfer rate than the rather slow RF, and in swapping the transfer rate rather
than access time is the dominant influence on throughput.
Once you have decided how to make best use of your hardware, the question is how to initialize it. If
you have the equipment, the best way to move a file system is to dump it (dump-VIII) to magtape, use
mkfs-VIII to create the new file system, and restore the tape. If you don’t have magtape, dump accepts an
 -7-

argument telling where to put the dump; you might use another disk or DECtape. Sometimes a file system
has to be increased in logical size without copying. The super-block of the device has a word giving the
highest address which can be allocated. For relatively small increases, this word can be patched using the
debugger (db-I) and the free list reconstructed using icheck-VIII. The size should not be increased very
greatly by this technique, however, since although the allocatable space will increase the maximum number
of files will not (that is, the i-list size can’t be changed). Read and understand the description given in file
system-VI before playing around in this way.
If you have only an RP disk, see section rp-IV for some suggestions on how to lay out the informa-
tion on it. The file systems distributed on tape, containing the binary, the source, and the manuals, are each
only 4000 blocks long. Perhaps the simplest way to integrate the latter two into a large file system is to
extract the tape into the upper part of the RP, dump it, and restore it into an empty, non-overlapping file sys-
tem structure. If you have to merge a file system into another, existing one, the best bet is to use ncheck-
VIII to get a list of names, then edit this list into a sequence of mkdir and cp commands which will serve as
input to the Shell. (But notice that owner information is lost.)

New Users
Install new users by editing the password file /etc/passwd (passwd-V). You’ll have to make current
directories for the new users and change their owners to the newly installed name. Login as each user to
make sure the password file is correctly edited. For example:
ed /etc/passwd
$a
joe::10:1::/usr/joe:
.
w
q
mkdir /usr/joe
chown joe /usr/joe
login joe
ls −la
login root
This will make a new login entry for joe. His default current directory is /usr/joe which has been created.
The delivered password file has the user ken in it to be used as a prototype.

Multiple Users
If UNIX is to support simultaneous access from more than just the console teletype, the file /etc/ttys
(ttys-V) has to be edited. For some historical reason tty8 is the name of the console typewriter. To add new
typewriters be sure the device is configured and the special file exists, then set the first character of the
appropriate line of /etc/ttys to 1 (or add a new line). Note that init.c will have to be recompiled if there are
to be more than 20 typewriters. Also note that if the special file is inaccessible when init tries to create a
process for it, the system will thrash trying and retrying to open it.

File System Health

Periodically (say every day or so) and always after a crash, you should check all the file systems for
consistency (icheck, dcheck-VIII). It is quite important to execute sync (VIII) before rebooting or taking
the machine down. This is done automatically every 30 seconds by the update program (VIII) when a mul-
tiple-user system is running, but you should do it anyway to make sure.
Dumping of the file system should be done regularly, since once the system is going it is very easy to
become complacent. Just remember that our RP controller has failed three times, each time in such a way
that all information on the disk was wiped out without any error status from the controller. Complete and
incremental dumps are easily done with the dump command (VIII) but restoration of individual files is
painful. Dumping of files by name is best done by tp (I) but the number of files is limited. Finally if there
are enough drives entire disks can be copied using cp-I, or preferably with dd-I using the raw special files
 -8-

and an appropriate block size. Note that there is no stand-alone program with UNIX that will restore any of
these formats. Unless some action has been taken to prevent destruction of a running version of UNIX, you
can find yourself stranded even though you have backup.
 -9-

Odds and Ends

The programs dump, icheck, dcheck, ncheck, and df (source in /usr/source/s1 and /usr/source/s2)
should be changed to reflect your default mounted file system devices. Print the first few lines of these pro-
grams and the changes will be obvious.
If you would like to share any UNIX compatible software with others, please let us know about it. If
you find bugs in the software or the documentation, again let us know.
Lastly, there is a UNIX users’ group forming. To get on their mailing list, send your name(s) and
address to:
Prof. Melvin Ferentz
Physics Dept.
Brooklyn College of CUNY
Brooklyn, N.Y. 11210

Good luck.
Ken Thompson
Dennis Ritchie
-

The UNIX Time-Sharing System

Dennis M. Ritchie
Ken Thompson

Bell Laboratories
Murray Hill, N. J. 07974

ABSTRACT

UNIX is a general-purpose, multi-user, interactive operating system for the Digital Equipment Corporation
PDP-11/40, 11/45 and 11/70 computers. It offers a number of features seldom found even in larger operating sys-
tems, including
1. A hierarchical file system incorporating demountable volumes,
2. Compatible file, device, and inter-process I/O,
3. The ability to initiate asynchronous processes,
4. System command language selectable on a per-user basis,
5. Over 100 subsystems including a dozen languages.
This paper discusses the nature and implementation of the file system and of the user command interface.

1. Introduction
There have been three versions of UNIX. The earliest version (circa 1969-70) ran on the Digital Equipment Corpo-
ration PDP-7 and -9 computers. The second version ran on the unprotected PDP-11/20 computer. This paper de-
scribes only the PDP-11/40, /45 and /70 1 system, since it is more modern and many of the differences between it and
older UNIX systems result from redesign of features found to be deficient or lacking.
Since PDP-11 UNIX became operational in February, 1971, about 100 installations have been put into service; they
are generally smaller than the system described here. Most of them are engaged in applications such as the prepara-
tion and formatting of patent applications and other textual material, the collection and processing of trouble data
from various switching machines within the Bell System, and recording and checking telephone service orders. Our
own installation is used mainly for research in operating systems, languages, computer networks, and other topics in
computer science, and also for document preparation.

Copyright © 1974, Association for Computing Machinery, Inc. General permission to republish, but not for profit, all or part of this material is
granted provided that ACM’s copyright notice is given and that reference is made to the publication, to its date of issue, and to the fact that reprint-
ing privileges were granted by permission of the Association for Computing Machinery.
This is a revised version of an article appearing in the Communications of the ACM, Volume 17, Number 7 (July 1974) pp. 365-375. That arti-
cle is a revised version of a paper presented at the Fourth ACM Symposium on Operating Systems Principles, IBM Thomas J. Watson Research
Center, Yorktown Heights, New York, October 15-17, 1973.
-

UNIX Time-Sharing System - 2

Perhaps the most important achievement of UNIX is to demonstrate that a powerful operating system for interac-
tive use need not be expensive either in equipment or in human effort: UNIX can run on hardware costing as little as
$40,000, and less than two man-years were spent on the main system software. Yet UNIX contains a number of fea-
tures seldom offered even in much larger systems. Hopefully, however, the users of UNIX will find that the most im-
portant characteristics of the system are its simplicity, elegance, and ease of use.
Besides the system proper, the major programs available under UNIX are

assembler,
text editor based on QED 2 ,
linking loader,
symbolic debugger,
compiler for a language resembling BCPL 3 with types and structures (C),
interpreter for a dialect of BASIC,
phototypesetting and equation setting programs
Fortran compiler,
Snobol interpreter,
top-down compiler-compiler (TMG 4 ),
bottom-up compiler-compiler (YACC),
form letter generator,
macro processor (M6 5 ),
permuted index program.
There is also a host of maintenance, utility, recreation and novelty programs. All of these programs were written lo-
cally. It is worth noting that the system is totally self-supporting. All UNIX software is maintained under UNIX; like-
wise, this paper and all other UNIX documents were generated and formatted by the UNIX editor and text formatting
program.

2. Hardware and software environment

The PDP-11/45 on which our UNIX installation is implemented is a 16-bit word (8-bit byte) computer with 112K
bytes of core memory; UNIX occupies 53K bytes. This system, however, includes a very large number of device
drivers and enjoys a generous allotment of space for I/O buffers and system tables; a minimal system capable of run-
ning the software mentioned above can require as little as 64K bytes of core altogether.
Our PDP-11 has a 1M byte fixed-head disk, used for file system storage and swapping, four moving-head disk
drives which each provide 2.5M bytes on removable disk cartridges, and a single moving-head disk drive which uses
removable 40M byte disk packs. There are also a high-speed paper tape reader-punch, nine-track magnetic tape, and
DECtape (a variety of magnetic tape facility in which individual records may be addressed and rewritten). Besides
the console typewriter, there are 30 variable-speed communications interfaces attached to 100-series datasets and a
201 dataset interface used primarily for spooling printout to a communal line printer. There are also several one-of-
a-kind devices including a Picturephone® interface, a voice response unit, a voice synthesizer, a phototypesetter, a
digital switching network, and a satellite PDP-11/20 which generates vectors, curves, and characters on a Tektronix
611 storage-tube display.
The greater part of UNIX software is written in the above-mentioned C language 6 . Early versions of the operating
system were written in assembly language, but during the summer of 1973, it was rewritten in C. The size of the
new system is about one third greater than the old. Since the new system is not only much easier to understand and
to modify but also includes many functional improvements, including multiprogramming and the ability to share
reentrant code among several user programs, we considered this increase in size quite acceptable.
-

UNIX Time-Sharing System - 3

3. The File system

The most important role of UNIX is to provide a file system. From the point of view of the user, there are three
kinds of files: ordinary disk files, directories, and special files.

3.1 Ordinary files

A file contains whatever information the user places on it, for example symbolic or binary (object) programs. No
particular structuring is expected by the system. Files of text consist simply of a string of characters, with lines de-
marcated by the new-line character. Binary programs are sequences of words as they will appear in core memory
when the program starts executing. A few user programs manipulate files with more structure; for example, the as-
sembler generates, and the loader expects, an object file in a particular format. However, the structure of files is con-
trolled by the programs which use them, not by the system.

3.2 Directories
Directories provide the mapping between the names of files and the files themselves, and thus induce a structure
on the file system as a whole. Each user has a directory of his own files; he may also create subdirectories to contain
groups of files conveniently treated together. A directory behaves exactly like an ordinary file except that it cannot
be written on by unprivileged programs, so that the system controls the contents of directories. However, anyone
with appropriate permission may read a directory just like any other file.
The system maintains several directories for its own use. One of these is the root directory. All files in the system
can be found by tracing a path through a chain of directories until the desired file is reached. The starting point for
such searches is often the root. Another system directory contains all the programs provided for general use; that is,
all the commands. As will be seen, however, it is by no means necessary that a program reside in this directory for it
to be executed.
Files are named by sequences of 14 or fewer characters. When the name of a file is specified to the system, it may
be in the form of a path name, which is a sequence of directory names separated by slashes ‘‘ / ’’ and ending in a file
name. If the sequence begins with a slash, the search begins in the root directory. The name / alpha / beta / gamma
causes the system to search the root for directory alpha, then to search alpha for beta, finally to find gamma in beta.
Gamma may be an ordinary file, a directory, or a special file. As a limiting case, the name ‘‘/ ’’ refers to the root it-
self.
A path name not starting with ‘‘/ ’’ causes the system to begin the search in the user’s current directory. Thus, the
name alpha / beta specifies the file named beta in subdirectory alpha of the current directory. The simplest kind of
name, for example alpha, refers to a file which itself is found in the current directory. As another limiting case, the
null file name refers to the current directory.
The same non-directory file may appear in several directories under possibly different names. This feature is
called linking; a directory entry for a file is sometimes called a link. UNIX differs from other systems in which link-
ing is permitted in that all links to a file have equal status. That is, a file does not exist within a particular directory;
the directory entry for a file consists merely of its name and a pointer to the information actually describing the file.
Thus a file exists independently of any directory entry, although in practice a file is made to disappear along with the
last link to it.
Each directory always has at least two entries. The name ‘‘ . ’’ in each directory refers to the directory itself. Thus
a program may read the current directory under the name ‘‘ . ’’ without knowing its complete path name. The name
‘‘ . . ’’ by convention refers to the parent of the directory in which it appears, that is, to the directory in which it was
created.
The directory structure is constrained to have the form of a rooted tree. Except for the special entries ‘‘ . ’’ and
‘‘ . . ’’, each directory must appear as an entry in exactly one other, which is its parent. The reason for this is to sim-
plify the writing of programs which visit subtrees of the directory structure, and more important, to avoid the separa-
tion of portions of the hierarchy. If arbitrary links to directories were permitted, it would be quite difficult to detect
when the last connection from the root to a directory was severed.
-

UNIX Time-Sharing System - 4

3.3 Special files

Special files constitute the most unusual feature of the UNIX file system. Each I/O device supported by UNIX is as-
sociated with at least one such file. Special files are read and written just like ordinary disk files, but requests to read
or write result in activation of the associated device. An entry for each special file resides in directory / dev, although
a link may be made to one of these files just like an ordinary file. Thus, for example, to punch paper tape, one may
write on the file / dev / ppt. Special files exist for each communication line, each disk, each tape drive, and for physi-
cal core memory. Of course, the active disks and the core special file are protected from indiscriminate access.
There is a threefold advantage in treating I/O devices this way: file and device I/O are as similar as possible; file
and device names have the same syntax and meaning, so that a program expecting a file name as a parameter can be
passed a device name; finally, special files are subject to the same protection mechanism as regular files.

3.4 Removable file systems

Although the root of the file system is always stored on the same device, it is not necessary that the entire file sys-
tem hierarchy reside on this device. There is a mount system request which has two arguments: the name of an ex-
isting ordinary file, and the name of a special file whose associated storage volume (e. g. disk pack) should have the
structure of an independent file system containing its own directory hierarchy. The effect of mount is to cause refer-
ences to the heretofore ordinary file to refer instead to the root directory of the file system on the removable volume.
In effect, mount replaces a leaf of the hierarchy tree (the ordinary file) by a whole new subtree (the hierarchy stored
on the removable volume). After the mount, there is virtually no distinction between files on the removable volume
and those in the permanent file system. In our installation, for example, the root directory resides on the fixed-head
disk, and the large disk drive, which contains user’s files, is mounted by the system initialization program; the four
smaller disk drives are available to users for mounting their own disk packs. A mountable file system is generated
by writing on its corresponding special file. A utility program is available to create an empty file system, or one may
simply copy an existing file system.
There is only one exception to the rule of identical treatment of files on different devices: no link may exist be-
tween one file system hierarchy and another. This restriction is enforced so as to avoid the elaborate bookkeeping
which would otherwise be required to assure removal of the links when the removable volume is finally dismounted.
In particular, in the root directories of all file systems, removable or not, the name ‘‘ . . ’’ refers to the directory itself
instead of to its parent.

3.5 Protection
Although the access control scheme in UNIX is quite simple, it has some unusual features. Each user of the system
is assigned a unique user identification number. When a file is created, it is marked with the user ID of its owner.
Also given for new files is a set of seven protection bits. Six of these specify independently read, write, and execute
permission for the owner of the file and for all other users.
If the seventh bit is on, the system will temporarily change the user identification of the current user to that of the
creator of the file whenever the file is executed as a program. This change in user ID is effective only during the ex-
ecution of the program which calls for it. The set-user-ID feature provides for privileged programs which may use
files inaccessible to other users. For example, a program may keep an accounting file which should neither be read
nor changed except by the program itself. If the set-user-identification bit is on for the program, it may access the
file although this access might be forbidden to other programs invoked by the given program’s user. Since the actual
user ID of the invoker of any program is always available, set-user-ID programs may take any measures desired to
satisfy themselves as to their invoker’s credentials. This mechanism is used to allow users to execute the carefully-
written commands which call privileged system entries. For example, there is a system entry invokable only by the
‘‘super-user’’ (below) which creates an empty directory. As indicated above, directories are expected to have entries
for ‘‘ . ’’ and ‘‘ . . ’’. The command which creates a directory is owned by the super-user and has the set-user-ID bit
set. After it checks its invoker’s authorization to create the specified directory, it creates it and makes the entries for
‘‘ . ’’ and ‘‘ . . ’’.
Since anyone may set the set-user-ID bit on one of his own files, this mechanism is generally available without ad-
ministrative intervention. For example, this protection scheme easily solves the MOO accounting problem posed in
-

UNIX Time-Sharing System - 5

[7].
The system recognizes one particular user ID (that of the ‘‘super-user’’) as exempt from the usual constraints on
file access; thus (for example) programs may be written to dump and reload the file system without unwanted inter-
ference from the protection system.

3.6 I/O calls

The system calls to do I/O are designed to eliminate the differences between the various devices and styles of ac-
cess. There is no distinction between ‘‘random’’ and ‘‘sequential’’ I/O, nor is any logical record size imposed by the
system. The size of an ordinary file is determined by the highest byte written on it; no predetermination of the size
of a file is necessary or possible.
To illustrate the essentials of I/O in UNIX, some of the basic calls are summarized below in an anonymous lan-
guage which will indicate the required parameters without getting into the complexities of machine language pro-
gramming. Each call to the system may potentially result in an error return, which for simplicity is not represented
in the calling sequence.
To read or write a file assumed to exist already, it must be opened by the following call:
filep = open ( name, flag )
Name indicates the name of the file. An arbitrary path name may be given. The flag argument indicates whether the
file is to be read, written, or ‘‘updated,’’ that is read and written simultaneously.
The returned value filep is called a file descriptor. It is a small integer used to identify the file in subsequent calls
to read, write or otherwise manipulate the file.
To create a new file or completely rewrite an old one, there is a create system call which creates the given file if it
does not exist, or truncates it to zero length if it does exist. Create also opens the new file for writing and, like open,
returns a file descriptor.
There are no user-visible locks in the file system, nor is there any restriction on the number of users who may
have a file open for reading or writing. Although it is possible for the contents of a file to become scrambled when
two users write on it simultaneously, in practice difficulties do not arise. We take the view that locks are neither nec-
essary nor sufficient, in our environment, to prevent interference between users of the same file. They are unneces-
sary because we are not faced with large, single-file data bases maintained by independent processes. They are in-
sufficient because locks in the ordinary sense, whereby one user is prevented from writing on a file which another
user is reading, cannot prevent confusion when, for example, both users are editing a file with an editor which makes
a copy of the file being edited.
It should be said that the system has sufficient internal interlocks to maintain the logical consistency of the file
system when two users engage simultaneously in such inconvenient activities as writing on the same file, creating
files in the same directory, or deleting each other’s open files.
Except as indicated below, reading and writing are sequential. This means that if a particular byte in the file was
the last byte written (or read), the next I/O call implicitly refers to the first following byte. For each open file there is
a pointer, maintained by the system, which indicates the next byte to be read or written. If n bytes are read or writ-
ten, the pointer advances by n bytes.
Once a file is open, the following calls may be used.
n = read ( filep, buffer, count )
n = write ( filep, buffer, count )
Up to count bytes are transmitted between the file specified by filep and the byte array specified by buffer. The re-
turned value n is the number of bytes actually transmitted. In the write case, n is the same as count except under ex-
ceptional conditions like I/O errors or end of physical medium on special files; in a read, however, n may without er-
ror be less than count. If the read pointer is so near the end of the file that reading count characters would cause
reading beyond the end, only sufficient bytes are transmitted to reach the end of the file; also, typewriter-like devices
-

UNIX Time-Sharing System - 6

never return more than one line of input. When a read call returns with n equal to zero, it indicates the end of the
file. For disk files this occurs when the read pointer becomes equal to the current size of the file. It is possible to
generate an end-of-file from a typewriter by use of an escape sequence which depends on the device used.
Bytes written on a file affect only those implied by the position of the write pointer and the count; no other part of
the file is changed. If the last byte lies beyond the end of the file, the file is grown as needed.
To do random (direct access) I/O it is only necessary to move the read or write pointer to the appropriate location
in the file.
location = seek ( filep, offset, base )
The pointer associated with filep is moved to a position offset bytes from the beginning of the file, from the current
position of the pointer, or from the end of the file, depending on base. Offset may be negative. For some devices
(e.g. paper tape and typewriters) seek calls are ignored. The actual offset from the beginning of the file to which the
pointer was moved is returned in location.

3.6.1 Other I/O calls

There are several additional system entries having to do with I/O and with the file system which will not be dis-
cussed. For example: close a file, get the status of a file, change the protection mode or the owner of a file, create a
directory, make a link to an existing file, delete a file.

4. Implementation of the file system

As mentioned in §3.2 above, a directory entry contains only a name for the associated file and a pointer to the file
itself. This pointer is an integer called the i-number (for index number) of the file. When the file is accessed, its i-
number is used as an index into a system table (the i-list ) stored in a known part of the device on which the directory
resides. The entry thereby found (the file’s i-node ) contains the description of the file:

1. its owner;
2. its protection bits;
3. the physical disk or tape addresses for the file contents;
4. its size;
5. time of last modification;
6. the number of links to the file; that is, the number of times it appears in a directory;
7. a bit indicating whether the file is a directory;
8. a bit indicating whether the file is a special file;
9. a bit indicating whether the file is ‘‘large’’ or ‘‘small.’’

The purpose of an open or create system call is to turn the path name given by the user into an i-number by search-
ing the explicitly or implicitly named directories. Once a file is open, its device, i-number, and read/write pointer are
stored in a system table indexed by the file descriptor returned by the open or create. Thus the file descriptor sup-
plied during a subsequent call to read or write the file may be easily related to the information necessary to access
the file.
When a new file is created, an i-node is allocated for it and a directory entry is made which contains the name of
the file and the i-node number. Making a link to an existing file involves creating a directory entry with the new
name, copying the i-number from the original file entry, and incrementing the link-count field of the i-node. Remov-
ing (deleting) a file is done by decrementing the link-count of the i-node specified by its directory entry and erasing
the directory entry. If the link-count drops to 0, any disk blocks in the file are freed and the i-node is deallocated.
The space on all fixed or removable disks which contain a file system is divided into a number of 512-byte blocks
logically addressed from 0 up to a limit which depends on the device. There is space in the i-node of each file for
eight device addresses. A small (non-special) file fits into eight or fewer blocks; in this case the addresses of the
blocks themselves are stored. For large (non-special) files, seven of the eight device addresses may point to indirect
blocks each containing 256 addresses for the data blocks of the file. If required, the eighth word is the address of a
-

UNIX Time-Sharing System - 7

double-indirect block containing 256 more addresses of indirect blocks. Thus files may conceptually grow to
(7+256).256.512 bytes; actually they are restricted to 16,777,216 ( 224 ) bytes. Once opened, a small file (size 1 to 8
blocks) can be accessed directly. A large file (size 9 to 32768 blocks) requires one additional access to read below
logical block 1792 (7.256) and two additional references above 1792.
The foregoing discussion applies to ordinary files. When an I/O request is made to a file whose i-node indicates
that it is special, the last seven device address words are immaterial, and the first is interpreted as a pair of bytes
which constitute an internal device name. These bytes specify respectively a device type and subdevice number.
The device type indicates which system routine will deal with I/O on that device; the subdevice number selects, for
example, a disk drive attached to a particular controller or one of several similar typewriter interfaces.
In this environment, the implementation of the mount system call (§3.4) is quite straightforward. Mount main-
tains a system table whose argument is the i-number and device name of the ordinary file specified during the mount,
and whose corresponding value is the device name of the indicated special file. This table is searched for each (i-
number, device)-pair which turns up while a path name is being scanned during an open or create; if a match is
found, the i-number is replaced by 1 (which is the i-number of the root directory on all file systems), and the device
name is replaced by the table value.
To the user, both reading and writing of files appear to be synchronous and unbuffered. That is, immediately after
return from a read call the data are available, and conversely after a write the user’s workspace may be reused. In
fact the system maintains a rather complicated buffering mechanism which reduces greatly the number of I/O opera-
tions required to access a file. Suppose a write call is made specifying transmission of a single byte. UNIX will
search its buffers to see whether the affected disk block currently resides in core memory; if not, it will be read in
from the device. Then the affected byte is replaced in the buffer and an entry is made in a list of blocks to be writ-
ten. The return from the write call may then take place, although the actual I/O may not be completed until a later
time. Conversely, if a single byte is read, the system determines whether the secondary storage block in which the
byte is located is already in one of the system’s buffers; if so, the byte can be returned immediately. If not, the block
is read into a buffer and the byte picked out.
The system recognizes when a program has made accesses to sequential blocks of a file, and asynchronously pre-
reads the next block. This significantly reduces the running time of most programs while adding little to system
overhead.
A program which reads or writes files in units of 512 bytes has an advantage over a program which reads or writes
a single byte at a time, but the gain is not immense; it comes mainly from the avoidance of system overhead. A pro-
gram which is used rarely or which does no great volume of I/O may quite reasonably read and write in units as
small as it wishes.
The notion of the i-list is an unusual feature of UNIX. In practice, this method of organizing the file system has
proved quite reliable and easy to deal with. To the system itself, one of its strengths is the fact that each file has a
short, unambiguous name which is related in a simple way to the protection, addressing, and other information need-
ed to access the file. It also permits a quite simple and rapid algorithm for checking the consistency of a file system,
for example verification that the portions of each device containing useful information and those free to be allocated
are disjoint and together exhaust the space on the device. This algorithm is independent of the directory hierarchy,
since it need only scan the linearly-organized i-list. At the same time the notion of the i-list induces certain peculiar-
ities not found in other file system organizations. For example, there is the question of who is to be charged for the
space a file occupies, since all directory entries for a file have equal status. Charging the owner of a file is unfair in
general, since one user may create a file, another may link to it, and the first user may delete the file. The first user is
still the owner of the file, but it should be charged to the second user. The simplest reasonably fair algorithm seems
to be to spread the charges equally among users who have links to a file. The current version of UNIX avoids the is-
sue by not charging any fees at all.
-

UNIX Time-Sharing System - 8

4.1 Efficiency of the file system

To provide an indication of the overall efficiency of UNIX and of the file system in particular, timings were made
of the assembly of a 8848-line program. The assembly was run alone on the machine; the total clock time was 32
seconds, for a rate of 276 lines per second. The time was divided as follows: 66% assembler execution time, 21%
system overhead, 13% disk wait time. We will not attempt any interpretation of these figures nor any comparison
with other systems, but merely note that we are generally satisfied with the overall performance of the system.

5. Processes and images

An image is a computer execution environment. It includes a core image, general register values, status of open
files, current directory and the like. An image is the current state of a pseudo-computer.
A process is the execution of an image. While the processor is executing on behalf of a process, the image must
reside in core; during the execution of other processes it remains in core unless the appearance of an active, higher-
priority process forces it to be swapped out to the fixed-head disk.
The user-core part of an image is divided into three logical segments. The program text segment begins at loca-
tion 0 in the virtual address space. During execution, this segment is write-protected and a single copy of it is shared
among all processes executing the same program. At the first 8K byte boundary above the program text segment in
the virtual address space begins a non-shared, writable data segment, the size of which may be extended by a system
call. Starting at the highest address in the virtual address space is a stack segment, which automatically grows
downward as the hardware’s stack pointer fluctuates.

5.1 Processes
Except while UNIX is bootstrapping itself into operation, a new process can come into existence only by use of the
fork system call:
processid = fork ( label )
When fork is executed by a process, it splits into two independently executing processes. The two processes have in-
dependent copies of the original core image, and share any open files. The new processes differ only in that one is
considered the parent process: in the parent, control returns directly from the fork, while in the child, control is
passed to location label. The processid returned by the fork call is the identification of the other process.
Because the return points in the parent and child process are not the same, each image existing after a fork may
determine whether it is the parent or child process.

5.2 Pipes
Processes may communicate with related processes using the same system read and write calls that are used for
file system I/O. The call
filep = pipe ( )
returns a file descriptor filep and creates an inter-process channel called a pipe. This channel, like other open files, is
passed from parent to child process in the image by the fork call. A read using a pipe file descriptor waits until an-
other process writes using the file descriptor for the same pipe. At this point, data are passed between the images of
the two processes. Neither process need know that a pipe, rather than an ordinary file, is involved.
Although inter-process communication via pipes is a quite valuable tool (see §6.2), it is not a completely general
mechanism, since the pipe must be set up by a common ancestor of the processes involved.

5.3 Execution of programs

Another major system primitive is invoked by
execute ( file, arg1, arg2, . . . , argn )
which requests the system to read in and execute the program named by file, passing it string arguments arg1 , arg2 ,
-

UNIX Time-Sharing System - 9

. . . , argn. All the code and data in the process using execute is replaced from the file, but open files, current directo-
ry, and inter-process relationships are unaltered. Only if the call fails, for example because file could not be found or
because its execute-permission bit was not set, does a return take place from the execute primitive; it resembles a
‘‘jump’’ machine instruction rather than a subroutine call.

5.4 Process synchronization

Another process control system call
processid = wait ( )
causes its caller to suspend execution until one of its children has completed execution. Then wait returns the pro-
cessid of the terminated process. An error return is taken if the calling process has no descendants. Certain status
from the child process is also available.

5.5 Termination
Lastly,
exit ( status )
terminates a process, destroys its image, closes its open files, and generally obliterates it. When the parent is notified
through the wait primitive, the indicated status is available to the parent. Processes may also terminate as a result of
various illegal actions or user-generated signals (§7 below).

6. The Shell
For most users, communication with UNIX is carried on with the aid of a program called the Shell. The Shell is a
command line interpreter: it reads lines typed by the user and interprets them as requests to execute other programs.
In simplest form, a command line consists of the command name followed by arguments to the command, all sepa-
rated by spaces:
command arg1 arg2 . . . argn
The Shell splits up the command name and the arguments into separate strings. Then a file with name command is
sought; command may be a path name including the ‘‘/’’ character to specify any file in the system. If command is
found, it is brought into core and executed. The arguments collected by the Shell are accessible to the command.
When the command is finished, the Shell resumes its own execution, and indicates its readiness to accept another
command by typing a prompt character.
If file command cannot be found, the Shell prefixes the string / bin/ to command and attempts again to find the
file. Directory / bin contains all the commands intended to be generally used.

6.1 Standard I/O

The discussion of I/O in §3 above seems to imply that every file used by a program must be opened or created by
the program in order to get a file descriptor for the file. Programs executed by the Shell, however, start off with two
open files which have file descriptors 0 and 1. As such a program begins execution, file 1 is open for writing, and is
best understood as the standard output file. Except under circumstances indicated below, this file is the user’s type-
writer. Thus programs which wish to write informative or diagnostic information ordinarily use file descriptor 1.
Conversely, file 0 starts off open for reading, and programs which wish to read messages typed by the user usually
read this file.
The Shell is able to change the standard assignments of these file descriptors from the user’s typewriter printer
and keyboard. If one of the arguments to a command is prefixed by ‘‘>’’, file descriptor 1 will, for the duration of
the command, refer to the file named after the ‘‘>’’. For example,
ls
ordinarily lists, on the typewriter, the names of the files in the current directory. The command
-

UNIX Time-Sharing System - 10

ls >there
creates a file called there and places the listing there. Thus the argument ‘‘>there’’ means, ‘‘place output on there.’’
On the other hand,
ed
ordinarily enters the editor, which takes requests from the user via his typewriter. The command
ed <script
interprets script as a file of editor commands; thus ‘‘<script’’ means, ‘‘take input from script.’’
Although the file name following ‘‘<’’ or ‘‘>’’ appears to be an argument to the command, in fact it is interpreted
completely by the Shell and is not passed to the command at all. Thus no special coding to handle I/O redirection is
needed within each command; the command need merely use the standard file descriptors 0 and 1 where appropri-
ate.

6.2 Filters
An extension of the standard I/O notion is used to direct output from one command to the input of another. A se-
quence of commands separated by vertical bars causes the Shell to execute all the commands simultaneously and to
arrange that the standard output of each command be delivered to the standard input of the next command in the se-
quence. Thus in the command line
ls | pr −2 | opr
ls lists the names of the files in the current directory; its output is passed to pr, which paginates its input with dated
headings. The argument ‘‘−2’’ means double column. Likewise the output from pr is input to opr. This command
spools its input onto a file for off-line printing.
This procedure could have been carried out more clumsily by
ls >temp1
pr −2 <temp1 >temp2
opr <temp2
followed by removal of the temporary files. In the absence of the ability to redirect output and input, a still clumsier
method would have been to require the ls command to accept user requests to paginate its output, to print in multi-
column format, and to arrange that its output be delivered off-line. Actually it would be surprising, and in fact un-
wise for efficiency reasons, to expect authors of commands such as ls to provide such a wide variety of output op-
tions.
A program such as pr which copies its standard input to its standard output (with processing) is called a filter.
Some filters which we have found useful perform character transliteration, sorting of the input, and encryption and
decryption.

6.3 Command Separators; Multitasking

Another feature provided by the Shell is relatively straightforward. Commands need not be on different lines; in-
stead they may be separated by semicolons.
ls; ed
will first list the contents of the current directory, then enter the editor.
A related feature is more interesting. If a command is followed by ‘‘&’’, the Shell will not wait for the command
to finish before prompting again; instead, it is ready immediately to accept a new command. For example,
as source >output &
causes source to be assembled, with diagnostic output going to output; no matter how long the assembly takes, the
Shell returns immediately. When the Shell does not wait for the completion of a command, the identification of the
-

UNIX Time-Sharing System - 11

process running that command is printed. This identification may be used to wait for the completion of the com-
mand or to terminate it. The ‘‘&’’ may be used several times in a line:
as source >output & ls >files &
does both the assembly and the listing in the background. In the examples above using ‘‘&’’, an output file other
than the typewriter was provided; if this had not been done, the outputs of the various commands would have been
intermingled.
The Shell also allows parentheses in the above operations. For example
( date; ls ) >x &
prints the current date and time followed by a list of the current directory onto the file x. The Shell also returns im-
mediately for another request.

6.4 The Shell as a Command; Command Files

The Shell is itself a command, and may be called recursively. Suppose file tryout contains the lines
as source
mv a.out testprog
testprog
The mv command causes the file a.out to be renamed testprog. A.out is the (binary) output of the assembler, ready to
be executed. Thus if the three lines above were typed on the console, source would be assembled, the resulting pro-
gram renamed testprog, and testprog executed. When the lines are in tryout, the command
sh <tryout
would cause the Shell sh to execute the commands sequentially.
The Shell has further capabilities, including the ability to substitute parameters and to construct argument lists
from a specified subset of the file names in a directory. It is also possible to execute commands conditionally on
character string comparisons or on existence of given files and to perform transfers of control within filed command
sequences.

6.5 Implementation of the Shell

The outline of the operation of the Shell can now be understood. Most of the time, the Shell is waiting for the us-
er to type a command. When the new-line character ending the line is typed, the Shell’s read call returns. The Shell
analyzes the command line, putting the arguments in a form appropriate for execute. Then fork is called. The child
process, whose code of course is still that of the Shell, attempts to perform an execute with the appropriate argu-
ments. If successful, this will bring in and start execution of the program whose name was given. Meanwhile, the
other process resulting from the fork, which is the parent process, wait s for the child process to die. When this hap-
pens, the Shell knows the command is finished, so it types its prompt and reads the typewriter to obtain another com-
mand.
Given this framework, the implementation of background processes is trivial; whenever a command line contains
‘‘&’’, the Shell merely refrains from waiting for the process which it created to execute the command.
Happily, all of this mechanism meshes very nicely with the notion of standard input and output files. When a pro-
cess is created by the fork primitive, it inherits not only the core image of its parent but also all the files currently
open in its parent, including those with file descriptors 0 and 1. The Shell, of course, uses these files to read com-
mand lines and to write its prompts and diagnostics, and in the ordinary case its children_the command pro-
grams_inherit them automatically. When an argument with ‘‘<’’ or ‘‘>’’ is given however, the offspring process, just
before it performs execute, makes the standard I/O file descriptor 0 or 1 respectively refer to the named file. This is
easy because, by agreement, the smallest unused file descriptor is assigned when a new file is open ed (or create d); it
is only necessary to close file 0 (or 1) and open the named file. Because the process in which the command program
runs simply terminates when it is through, the association between a file specified after ‘‘<’’ or ‘‘>’’ and file descrip-
tor 0 or 1 is ended automatically when the process dies. Therefore the Shell need not know the actual names of the
-

UNIX Time-Sharing System - 12

files which are its own standard input and output, since it need never reopen them.
Filters are straightforward extensions of standard I/O redirection with pipes used instead of files.
In ordinary circumstances, the main loop of the Shell never terminates. (The main loop includes that branch of
the return from fork belonging to the parent process; that is, the branch which does a wait, then reads another com-
mand line.) The one thing which causes the Shell to terminate is discovering an end-of-file condition on its input
file. Thus, when the Shell is executed as a command with a given input file, as in
sh <comfile
the commands in comfile will be executed until the end of comfile is reached; then the instance of the Shell invoked
by sh will terminate. Since this Shell process is the child of another instance of the Shell, the wait executed in the
latter will return, and another command may be processed.

6.6 Initialization
The instances of the Shell to which users type commands are themselves children of another process. The last
step in the initialization of UNIX is the creation of a single process and the invocation (via execute) of a program
called init. The role of init is to create one process for each typewriter channel which may be dialed up by a user.
The various subinstances of init open the appropriate typewriters for input and output. Since when init was invoked
there were no files open, in each process the typewriter keyboard will receive file descriptor 0 and the printer file de-
scriptor 1. Each process types out a message requesting that the user log in and waits, reading the typewriter, for a
reply. At the outset, no one is logged in, so each process simply hangs. Finally someone types his name or other
identification. The appropriate instance of init wakes up, receives the log-in line, and reads a password file. If the
user name is found, and if he is able to supply the correct password, init changes to the user’s default current directo-
ry, sets the process’s user ID to that of the person logging in, and performs an execute of the Shell. At this point the
Shell is ready to receive commands and the logging-in protocol is complete.
Meanwhile, the mainstream path of init (the parent of all the subinstances of itself which will later become Shells)
does a wait. If one of the child processes terminates, either because a Shell found an end of file or because a user
typed an incorrect name or password, this path of init simply recreates the defunct process, which in turn reopens the
appropriate input and output files and types another login message. Thus a user may log out simply by typing the
end-of-file sequence in place of a command to the Shell.

6.7 Other programs as Shell

The Shell as described above is designed to allow users full access to the facilities of the system, since it will
invoke the execution of any program with appropriate protection mode. Sometimes, however, a different interface to
the system is desirable, and this feature is easily arranged.
Recall that after a user has successfully logged in by supplying his name and password, init ordinarily invokes the
Shell to interpret command lines. The user’s entry in the password file may contain the name of a program to be in-
voked after login instead of the Shell. This program is free to interpret the user’s messages in any way it wishes.
For example, the password file entries for users of a secretarial editing system specify that the editor ed is to be
used instead of the Shell. Thus when editing system users log in, they are inside the editor and can begin work im-
mediately; also, they can be prevented from invoking UNIX programs not intended for their use. In practice, it has
proved desirable to allow a temporary escape from the editor to execute the formatting program and other utilities.
Several of the games (e.g., chess, blackjack, 3D tic-tac-toe) available on UNIX illustrate a much more severely re-
stricted environment. For each of these an entry exists in the password file specifying that the appropriate game-
playing program is to be invoked instead of the Shell. People who log in as a player of one of the games find them-
selves limited to the game and unable to investigate the presumably more interesting offerings of UNIX as a whole.
-

UNIX Time-Sharing System - 13

7. Traps
The PDP-11 hardware detects a number of program faults, such as references to non-existent memory, unimple-
mented instructions, and odd addresses used where an even address is required. Such faults cause the processor to
trap to a system routine. When an illegal action is caught, unless other arrangements have been made, the system
terminates the process and writes the user’s image on file core in the current directory. A debugger can be used to
determine the state of the program at the time of the fault.
Programs which are looping, which produce unwanted output, or about which the user has second thoughts may
be halted by the use of the interrupt signal, which is generated by typing the ‘‘delete’’ character. Unless special ac-
tion has been taken, this signal simply causes the program to cease execution without producing a core image file.
There is also a quit signal which is used to force a core image to be produced. Thus programs which loop unex-
pectedly may be halted and the core image examined without prearrangement.
The hardware-generated faults and the interrupt and quit signals can, by request, be either ignored or caught by
the process. For example, the Shell ignores quits to prevent a quit from logging the user out. The editor catches in-
terrupts and returns to its command level. This is useful for stopping long printouts without losing work in progress
(the editor manipulates a copy of the file it is editing). In systems without floating point hardware, unimplemented
instructions are caught and floating point instructions are interpreted.

8. Perspective
Perhaps paradoxically, the success of UNIX is largely due to the fact that it was not designed to meet any prede-
fined objectives. The first version was written when one of us (Thompson), dissatisfied with the available computer
facilities, discovered a little-used PDP-7 and set out to create a more hospitable environment. This essentially per-
sonal effort was sufficiently successful to gain the interest of the remaining author and others, and later to justify the
acquisition of the PDP-11/20, specifically to support a text editing and formatting system. When in turn the 11/20
was outgrown, UNIX had proved useful enough to persuade management to invest in the PDP-11/45. Our goals
throughout the effort, when articulated at all, have always concerned themselves with building a comfortable rela-
tionship with the machine and with exploring ideas and inventions in operating systems. We have not been faced
with the need to satisfy someone else’s requirements, and for this freedom we are grateful.
Three considerations which influenced the design of UNIX are visible in retrospect.
First: since we are programmers, we naturally designed the system to make it easy to write, test, and run pro-
grams. The most important expression of our desire for programming convenience was that the system was arranged
for interactive use, even though the original version only supported one user. We believe that a properly-designed
interactive system is much more productive and satisfying to use than a ‘‘batch’’ system. Moreover such a system is
rather easily adaptable to non-interactive use, while the converse is not true.
Second: there have always been fairly severe size constraints on the system and its software. Given the partially
antagonistic desires for reasonable efficiency and expressive power, the size constraint has encouraged not only
economy but a certain elegance of design. This may be a thinly disguised version of the ‘‘salvation through suffer-
ing’’ philosophy, but in our case it worked.
Third: nearly from the start, the system was able to, and did, maintain itself. This fact is more important than it
might seem. If designers of a system are forced to use that system they quickly become aware of its functional and
superficial deficiencies and are strongly motivated to correct them before it is too late. Since all source programs
were always available and easily modified on-line, we were willing to revise and rewrite the system and its software
when new ideas were invented, discovered, or suggested by others.
The aspects of UNIX discussed in this paper exhibit clearly at least the first two of these design considerations.
The interface to the file system, for example, is extremely convenient from a programming standpoint. The lowest
possible interface level is designed to eliminate distinctions between the various devices and files and between direct
and sequential access. No large ‘‘access method’’ routines are required to insulate the programmer from the system
calls; in fact all user programs either call the system directly or use a small library program, only tens of instructions
long, which buffers a number of characters and reads or writes them all at once.
-

UNIX Time-Sharing System - 14

Another important aspect of programming convenience is that there are no ‘‘control blocks’’ with a complicated
structure partially maintained by and depended on by the file system or other system calls. Generally speaking, the
contents of a program’s address space are the property of the program, and we have tried to avoid placing restric-
tions on the data structures within that address space.
Given the requirement that all programs should be usable with any file or device as input or output, it is also desir-
able from a space-efficiency standpoint to push device-dependent considerations into the operating system itself.
The only alternatives seem to be to load routines for dealing with each device with all programs, which is expensive
in space, or to depend on some means of dynamically linking to the routine appropriate to each device when it is ac-
tually needed, which is expensive either in overhead or in hardware.
Likewise, the process control scheme and command interface have proved both convenient and efficient. Since
the Shell operates as an ordinary, swappable user program, it consumes no wired-down space in the system proper,
and it may be made as powerful as desired at little cost. In particular, given the framework in which the Shell ex-
ecutes as a process which spawns other processes to perform commands, the notions of I/O redirection, background
processes, command files, and user-selectable system interfaces all become essentially trivial to implement.

8.1 Influences
The success of UNIX lies not so much in new inventions but rather in the full exploitation of a carefully selected
set of fertile ideas, and especially in showing that they can be keys to the implementation of a small yet powerful op-
erating system.
The fork operation, essentially as we implemented it, was present in the Berkeley time sharing system 8 . On a
number of points we were influenced by Multics, which suggested the particular form of the I/O system calls 9 and
both the name of the Shell and its general functions. The notion that the Shell should create a process for each com-
mand was also suggested to us by the early design of Multics, although in that system it was later dropped for effi-
ciency reasons. A similar scheme is used by TENEX 10 .

9. Statistics
The following numbers are presented to suggest the scale of our operation. Those of our users not involved in
document preparation tend to use the system for program development, especially language work. There are few
important ‘‘applications’’ programs.
Overall, we have
100 user population
14 maximum simultaneous users
380 directories
4800 files
66300 512-byte secondary storage blocks used
There is a ‘‘background’’ process that runs at the lowest possible priority; it is used to soak up any idle CPU time.
It has been used to produce a million-digit approximation to the constant e−2, and is now solving all rook-and-pawn
vs. rook chess endgames. Not counting this background work, we average daily
2400 commands
5.5 CPU hours
100 connect hours
32 different users
100 logins

Acknowledgements. We are grateful to R.H. Canaday, L.L. Cherry, and L.E. McMahon for their contributions to
UNIX.We are particularly appreciative of the inventiveness, thoughtful criticism, and constant support of R. Morris,
M.D. McIlroy, and J.F. Ossanna.
-

UNIX Time-Sharing System - 15

References
1. Digital Equipment Corporation. PDP-11/40 Processor Handbook (1972), PDP-11/45 Processor Handbook
(1971), and PDP-11/70 Processor Handbook (1975).
2. Deutsch, L.P., and Lampson, B.W. An online editor. Comm. ACM 10, 12 (Dec. 1967), 793-799, 803.
3. Richards, M. BCPL: A tool for compiler writing and system programming. Proc. AFIPS 1969 SJCC, Vol. 34,
AFIPS Press, Montvale, N.J., pp. 557-566.
4. McClure, R.M. TMG—A syntax directed compiler. Proc. ACM 20th Nat. Conf., ACM, 1965, New York, pp.
262-274.
5. Hall, A.D. The M6 macroprocessor. Computing Science Tech. Rep. #2, Bell Telephone Laboratories, 1969.
6. Ritchie, D.M. C reference manual. Unpublished memorandum, Bell Telephone Laboratories (1973).
7. Aleph-null. Computer Recreations. Software Practice and Experience 1, 2 (Apr.-June 1971), 201-204.
8. Deutch, L.P. and Lampson, B.W. SDS 930 time-sharing system preliminary reference manual. Doc.
30.10.10, Project GENIE, Univ. Cal. at Berkeley (Apr. 1965).
9. Feiertag, R.J., and Organick, E.I. The Multics input-output system. Proc. Third Symposium on Operating
Systems Principles. Oct. 18-20, 1971, ACM, New York, pp. 35-41.
10. Bobrow, D.G., Burchfiel, J.D., Murphy, D.L., and Tomlinson, R.S. TENEX, a paged time sharing system for
the PDP-10. Comm. ACM 15., 3 (March 1972) 135-143.
-

C Reference Manual

Dennis M. Ritchie
Bell Telephone Laboratories
Murray Hill, New Jersey 07974

1. Introduction
C is a computer language based on the earlier language B [1]. The languages and their compilers differ in two
major ways: C introduces the notion of types, and defines appropriate extra syntax and semantics; also, C on the
PDP-11 is a true compiler, producing machine code where B produced interpretive code.
Most of the software for the UNIX time-sharing system [2] is written in C, as is the operating system itself. C is al-
so available on the HIS 6070 computer at Murray Hill and and on the IBM System/370 at Holmdel [3]. This paper is
a manual only for the C language itself as implemented on the PDP-11. However, hints are given occasionally in the
text of implementation-dependent features.
The UNIX Programmer’s Manual [4] describes the library routines available to C programs under UNIX, and also
the procedures for compiling programs under that system. ‘‘The GCOS C Library’’ by Lesk and Barres [5] describes
routines available under that system as well as compilation procedures. Many of these routines, particularly the ones
having to do with I/O, are also provided under UNIX. Finally, ‘‘Programming in C− A Tutorial,’’ by B. W. Kernighan
[6], is as useful as promised by its title and the author’s previous introductions to allegedly impenetrable subjects.

2. Lexical conventions
There are six kinds of tokens: identifiers, keywords, constants, strings, expression operators, and other separators.
In general blanks, tabs, newlines, and comments as described below are ignored except as they serve to separate to-
kens. At least one of these characters is required to separate otherwise adjacent identifiers, constants, and certain op-
erator-pairs.
If the input stream has been parsed into tokens up to a given character, the next token is taken to include the
longest string of characters which could possibly constitute a token.

2.1 Comments
The characters /* introduce a comment, which terminates with the characters */.

2.2 Identifiers (Names)

An identifier is a sequence of letters and digits; the first character must be alphabetic. The underscore ‘‘ ’’ counts
as alphabetic. Upper and lower case letters are considered different. No more than the first eight characters are sig-
nificant, and only the first seven for external identifiers.

2.3 Keywords
The following identifiers are reserved for use as keywords, and may not be used otherwise:
-

C Reference Manual - 2

int break
char continue
float if
double else
struct for
auto do
extern while
register switch
static case
goto default
return entry
sizeof
The entry keyword is not currently implemented by any compiler but is reserved for future use.

2.3 Constants
There are several kinds of constants, as follows:

2.3.1 Integer constants

An integer constant is a sequence of digits. An integer is taken to be octal if it begins with 0, decimal otherwise.
The digits 8 and 9 have octal value 10 and 11 respectively.

2.3.2 Character constants

A character constant is 1 or 2 characters enclosed in single quotes ‘‘ ´ ’’. Within a character constant a single
quote must be preceded by a back-slash ‘‘\’’. Certain non-graphic characters, and ‘‘\’’ itself, may be escaped accord-
ing to the following table:
BS \b
NL \n
CR \r
HT \t
ddd \ddd
\ \\
The escape ‘‘\ddd’’ consists of the backslash followed by 1, 2, or 3 octal digits which are taken to specify the value
of the desired character. A special case of this construction is ‘‘\0’’ (not followed by a digit) which indicates a null
character.
Character constants behave exactly like integers (not, in particular, like objects of character type). In conformity
with the addressing structure of the PDP-11, a character constant of length 1 has the code for the given character in
the low-order byte and 0 in the high-order byte; a character constant of length 2 has the code for the first character in
the low byte and that for the second character in the high-order byte. Character constants with more than one char-
acter are inherently machine-dependent and should be avoided.

2.3.3 Floating constants

A floating constant consists of an integer part, a decimal point, a fraction part, an e, and an optionally signed inte-
ger exponent. The integer and fraction parts both consist of a sequence of digits. Either the integer part or the frac-
tion part (not both) may be missing; either the decimal point or the e and the exponent (not both) may be missing.
Every floating constant is taken to be double-precision.

2.4 Strings
A string is a sequence of characters surrounded by double quotes ‘‘ " ’’. A string has the type array-of-characters
(see below) and refers to an area of storage initialized with the given characters. The compiler places a null byte
( \0 ) at the end of each string so that programs which scan the string can find its end. In a string, the character ‘‘ " ’’
must be preceded by a ‘‘\’’ ; in addition, the same escapes as described for character constants may be used.
-

C Reference Manual - 3

3. Syntax notation
In the syntax notation used in this manual, syntactic categories are indicated by italic type, and literal words and
characters in gothic. Alternatives are listed on separate lines. An optional terminal or non-terminal symbol is in-
dicated by the subscript ‘‘opt,’’ so that
{ expressionopt }
would indicate an optional expression in braces.

4. What’s in a Name?
C bases the interpretation of an identifier upon two attributes of the identifier: its storage class and its type. The
storage class determines the location and lifetime of the storage associated with an identifier; the type determines the
meaning of the values found in the identifier’s storage.
There are four declarable storage classes: automatic, static, external, and register. Automatic variables are local to
each invocation of a function, and are discarded on return; static variables are local to a function, but retain their val-
ues independently of invocations of the function; external variables are independent of any function. Register vari-
ables are stored in the fast registers of the machine; like automatic variables they are local to each function and dis-
appear on return.
C supports four fundamental types of objects: characters, integers, single-, and double-precision floating-point
numbers.
Characters (declared, and hereinafter called, char) are chosen from the ASCII set; they occupy the right-
most seven bits of an 8-bit byte. It is also possible to interpret chars as signed, 2’s complement 8-bit num-
bers.
Integers (int) are represented in 16-bit 2’s complement notation.
Single precision floating point (float) quantities have magnitude in the range approximately 10±38 or 0;
their precision is 24 bits or about seven decimal digits.
Double-precision floating-point (double) quantities have the same range as floats and a precision of 56
bits or about 17 decimal digits.

Besides the four fundamental types there is a conceptually infinite class of derived types constructed from the fun-
damental types in the following ways:
arrays of objects of most types;
functions which return objects of a given type;
pointers to objects of a given type;
structures containing objects of various types.
In general these methods of constructing objects can be applied recursively.

5. Objects and lvalues

An object is a manipulatable region of storage; an lvalue is an expression referring to an object. An obvious ex-
ample of an lvalue expression is an identifier. There are operators which yield lvalues: for example, if E is an ex-
pression of pointer type, then *E is an lvalue expression referring to the object to which E points. The name ‘‘lval-
ue’’ comes from the assignment expression ‘‘E1 = E2’’ in which the left operand E1 must be an lvalue expression.
The discussion of each operator below indicates whether it expects lvalue operands and whether it yields an lvalue.

6. Conversions
A number of operators may, depending on their operands, cause conversion of the value of an operand from one
type to another. This section explains the result to be expected from such conversions.
-

C Reference Manual - 4

6.1 Characters and integers

A char object may be used anywhere an int may be. In all cases the char is converted to an int by propa-
gating its sign through the upper 8 bits of the resultant integer. This is consistent with the two’s complement repre-
sentation used for both characters and integers. (However, the sign-propagation feature disappears in other imple-
mentations.)

6.2 Float and double

All floating arithmetic in C is carried out in double-precision; whenever a float appears in an expression it is
lengthened to double by zero-padding its fraction. When a double must be converted to float, for example by
an assignment, the double is rounded before truncation to float length.

6.3 Float and double; integer and character

All ints and chars may be converted without loss of significance to float or double. Conversion of
float or double to int or char takes place with truncation towards 0. Erroneous results can be expected if the
magnitude of the result exceeds 32,767 (for int) or 127 (for char).

6.4 Pointers and integers

Integers and pointers may be added and compared; in such a case the int is converted as specified in the discus-
sion of the addition operator.
Two pointers to objects of the same type may be subtracted; in this case the result is converted to an integer as
specified in the discussion of the subtraction operator.

7. Expressions
The precedence of expression operators is the same as the order of the major subsections of this section (highest
precedence first). Thus the expressions referred to as the operands of + (§7.4) are those expressions defined in
§§7.1—7.3. Within each subsection, the operators have the same precedence. Left- or right-associativity is speci-
fied in each subsection for the operators discussed therein. The precedence and associativity of all the expression
operators is summarized in an appendix.
Otherwise the order of evaluation of expressions is undefined. In particular the compiler considers itself free to
compute subexpressions in the order it believes most efficient, even if the subexpressions involve side effects.

7.1 Primary expressions

Primary expressions involving . , −>, subscripting, and function calls group left to right.

7.1.1 identifier
An identifier is a primary expression, provided it has been suitably declared as discussed below. Its type is speci-
fied by its declaration. However, if the type of the identifier is ‘‘array of . . .’’, then the value of the identifier-expres-
sion is a pointer to the first object in the array, and the type of the expression is ‘‘pointer to . . .’’. Moreover, an array
identifier is not an lvalue expression.
Likewise, an identifier which is declared ‘‘function returning . . .’’, when used except in the function-name position
of a call, is converted to ‘‘pointer to function returning . . .’’.

7.1.2 constant
A decimal, octal, character, or floating constant is a primary expression. Its type is int in the first three cases,
double in the last.

7.1.3 string
A string is a primary expression. Its type is originally ‘‘array of char’’; but following the same rule as in §7.1.1
for identifiers, this is modified to ‘‘pointer to char’’ and the result is a pointer to the first character in the string.

7.1.4 ( expression )
A parenthesized expression is a primary expression whose type and value are identical to those of the unadorned
expression. The presence of parentheses does not affect whether the expression is an lvalue.
-

C Reference Manual - 5

7.1.5 primary-expression [ expression ]

A primary expression followed by an expression in square brackets is a primary expression. The intuitive mean-
ing is that of a subscript. Usually, the primary expression has type ‘‘pointer to . . .’’, the subscript expression is int,
and the type of the result is ‘‘ . . . ’’. The expression ‘‘E1[E2]’’ is identical (by definition) to ‘‘* ( ( E1 ) + ( E2 ) ) ’’.
All the clues needed to understand this notation are contained in this section together with the discussions in §§
7.1.1, 7.2.1, and 7.4.1 on identifiers, *, and + respectively; §14.3 below summarizes the implications.

7.1.6 primary-expression ( expression-listopt )

A function call is a primary expression followed by parentheses containing a possibly empty, comma-separated
list of expressions which constitute the actual arguments to the function. The primary expression must be of type
‘‘function returning . . .’’, and the result of the function call is of type ‘‘ . . . ’’. As indicated below, a hitherto unseen
identifier followed immediately by a left parenthesis is contextually declared to represent a function returning an in-
teger; thus in the most common case, integer-valued functions need not be declared.
Any actual arguments of type float are converted to double before the call; any of type char are converted
to int.
In preparing for the call to a function, a copy is made of each actual parameter; thus, all argument-passing in C is
strictly by value. A function may change the values of its formal parameters, but these changes cannot possibly af-
fect the values of the actual parameters. On the other hand, it is perfectly possible to pass a pointer on the under-
standing that the function may change the value of the object to which the pointer points.
Recursive calls to any function are permissible.

7.1.7 primary-lvalue . member-of-structure

An lvalue expression followed by a dot followed by the name of a member of a structure is a primary expression.
The object referred to by the lvalue is assumed to have the same form as the structure containing the structure mem-
ber. The result of the expression is an lvalue appropriately offset from the origin of the given lvalue whose type is
that of the named structure member. The given lvalue is not required to have any particular type.
Structures are discussed in §8.5.

7.1.8 primary-expression −> member-of-structure

The primary-expression is assumed to be a pointer which points to an object of the same form as the structure of
which the member-of-structure is a part. The result is an lvalue appropriately offset from the origin of the pointed-to
structure whose type is that of the named structure member. The type of the primary-expression need not in fact be
pointer; it is sufficient that it be a pointer, character, or integer.
Except for the relaxation of the requirement that E1 be of pointer type, the expression ‘‘E1−>MOS’’ is exactly
equivalent to ‘‘(*E1).MOS’’.

7.2 Unary operators

Expressions with unary operators group right-to-left.

7.2.1 * expression
The unary * operator means indirection: the expression must be a pointer, and the result is an lvalue referring to
the object to which the expression points. If the type of the expression is ‘‘pointer to . . .’’, the type of the result is
‘‘ . . . ’’.

7.2.2 & lvalue-expression

The result of the unary & operator is a pointer to the object referred to by the lvalue-expression. If the type of the
lvalue-expression is ‘‘ . . . ’’, the type of the result is ‘‘pointer to . . .’’.

7.2.3 − expression
The result is the negative of the expression, and has the same type. The type of the expression must be char,
int, float, or double.
-

C Reference Manual - 6

7.2.4 ! expression
The result of the logical negation operator ! is 1 if the value of the expression is 0, 0 if the value of the expres-
sion is non-zero. The type of the result is int. This operator is applicable only to ints or chars.

7.2.5
˜ expression
The operator yields the one’s complement of its operand. The type of the expression must be int or char, and
˜
the result is int.

7.2.6 ++ lvalue-expression
The object referred to by the lvalue expression is incremented. The value is the new value of the lvalue expres-
sion and the type is the type of the lvalue. If the expression is int or char, it is incremented by 1; if it is a pointer
to an object, it is incremented by the length of the object. ++ is applicable only to these types. (Not, for example, to
float or double.)

7.2.7 −− lvalue-expression
The object referred to by the lvalue expression is decremented analogously to the ++ operator.

7.2.8 lvalue-expression ++
The result is the value of the object referred to by the lvalue expression. After the result is noted, the object re-
ferred to by the lvalue is incremented in the same manner as for the prefix ++ operator: by 1 for an int or char, by
the length of the pointed-to object for a pointer. The type of the result is the same as the type of the lvalue-expres-
sion.

7.2.9 lvalue-expression −−
The result of the expression is the value of the object referred to by the the lvalue expression. After the result is
noted, the object referred to by the lvalue expression is decremented in a way analogous to the postfix ++ operator.

7.2.10 sizeof expression

The sizeof operator yields the size, in bytes, of its operand. When applied to an array, the result is the total
number of bytes in the array. The size is determined from the declarations of the objects in the expression. This ex-
pression is semantically an integer constant and may be used anywhere a constant is required. Its major use is in
communication with routines like storage allocators and I/O systems.

7.3 Multiplicative operators

The multiplicative operators *, /, and % group left-to-right.

7.3.1 expression * expression

The binary * operator indicates multiplication. If both operands are int or char, the result is int; if one is
int or char and one float or double, the former is converted to double, and the result is double; if both
are float or double, the result is double. No other combinations are allowed.

7.3.2 expression / expression

The binary / operator indicates division. The same type considerations as for multiplication apply.

7.3.3 expression % expression

The binary % operator yields the remainder from the division of the first expression by the second. Both operands
must be int or char, and the result is int. In the current implementation, the remainder has the same sign as the
dividend.

7.4 Additive operators

The additive operators + and − group left-to-right.
-

C Reference Manual - 7

7.4.1 expression + expression

The result is the sum of the expressions. If both operands are int or char, the result is int. If both are float
or double, the result is double. If one is char or int and one is float or double, the former is converted to
double and the result is double. If an int or char is added to a pointer, the former is converted by multiplying
it by the length of the object to which the pointer points and the result is a pointer of the same type as the original
pointer. Thus if P is a pointer to an object, the expression ‘‘P+1’’ is a pointer to another object of the same type as
the first and immediately following it in storage.
No other type combinations are allowed.

7.4.2 expression − expression

The result is the difference of the operands. If both operands are int, char, float, or double, the same type
considerations as for + apply. If an int or char is subtracted from a pointer, the former is converted in the same
way as explained under + above.
If two pointers to objects of the same type are subtracted, the result is converted (by division by the length of the
object) to an int representing the number of objects separating the pointed-to objects. This conversion will in gen-
eral give unexpected results unless the pointers point to objects in the same array, since pointers, even to objects of
the same type, do not necessarily differ by a multiple of the object-length.

7.5 Shift operators

The shift operators << and >> group left-to-right.

7.5.1 expression << expression

7.5.2 expression >> expression
Both operands must be int or char, and the result is int. The second operand should be non-negative. The
value of ‘‘E1<<E2’’ is E1 (interpreted as a bit pattern 16 bits long) left-shifted E2 bits; vacated bits are 0-filled. The
value of ‘‘E1>>E2’’ is E1 (interpreted as a two’s complement, 16-bit quantity) arithmetically right-shifted E2 bit po-
sitions. Vacated bits are filled by a copy of the sign bit of E1. [Note: the use of arithmetic rather than logical shift
does not survive transportation between machines.]

7.6 Relational operators

The relational operators group left-to-right, but this fact is not very useful; ‘‘a<b<c’’ does not mean what it seems
to.

7.6.1 expression < expression

7.6.2 expression > expression
7.6.3 expression <= expression
7.6.4 expression >= expression
The operators < (less than), > (greater than), <= (less than or equal to) and >= (greater than or equal to) all yield 0
if the specified relation is false and 1 if it is true. Operand conversion is exactly the same as for the + operator ex-
cept that pointers of any kind may be compared; the result in this case depends on the relative locations in storage of
the pointed-to objects. It does not seem to be very meaningful to compare pointers with integers other than 0.

7.7 Equality operators

7.7.1 expression == expression
7.7.2 expression != expression
The == (equal to) and the != (not equal to) operators are exactly analogous to the relational operators except for
their lower precedence. (Thus ‘‘a<b == c<d’’ is 1 whenever a<b and c<d have the same truth-value).

7.8 expression & expression

The & operator groups left-to-right. Both operands must be int or char; the result is an int which is the bit-
wise logical and function of the operands.
-

C Reference Manual - 8

7.9 expression ˆ expression

The ˆ operator groups left-to-right. The operands must be int or char; the result is an int which is the bit-
wise exclusive or function of its operands.

7.10 expression expression

The operator groups left-to-right. The operands must be int or char; the result is an int which is the bit-wise
inclusive or of its operands.

7.11 expression && expression

The && operator returns 1 if both its operands are non-zero, 0 otherwise. Unlike &, && guarantees left-to-right
evaluation; moreover the second operand is not evaluated if the first operand is 0.
The operands need not have the same type, but each must have one of the fundamental types or be a pointer.

7.12 expression expression

The operator returns 1 if either of its operands is non-zero, and 0 otherwise. Unlike , guarantees left-to-right
evaluation; moreover, the second operand is not evaluated if the value of the first operand is non-zero.
The operands need not have the same type, but each must have one of the fundamental types or be a pointer.

7.13 expression ? expression : expression

Conditional expressions group left-to-right. The first expression is evaluated and if it is non-zero, the result is the
value of the second expression, otherwise that of third expression. If the types of the second and third operand are
the same, the result has their common type; otherwise the same conversion rules as for + apply. Only one of the sec-
ond and third expressions is evaluated.

7.14 Assignment operators

There are a number of assignment operators, all of which group right-to-left. All require an lvalue as their left
operand, and the type of an assignment expression is that of its left operand. The value is the value stored in the left
operand after the assignment has taken place.

7.14.1 lvalue = expression

The value of the expression replaces that of the object referred to by the lvalue. The operands need not have the
same type, but both must be int, char, float, double, or pointer. If neither operand is a pointer, the assign-
ment takes place as expected, possibly preceded by conversion of the expression on the right.
When both operands are int or pointers of any kind, no conversion ever takes place; the value of the expression
is simply stored into the object referred to by the lvalue. Thus it is possible to generate pointers which will cause ad-
dressing exceptions when used.

7.14.2 lvalue =+ expression

7.14.3 lvalue =− expression
7.14.4 lvalue =* expression
7.14.5 lvalue =/ expression
7.14.6 lvalue =% expression
7.14.7 lvalue =>> expression
7.14.8 lvalue =<< expression
7.14.9 lvalue =& expression
7.14.10 lvalue =ˆ expression
7.14.11 lvalue = expression
The behavior of an expression of the form ‘‘E1 =op E2’’ may be inferred by taking it as equivalent to
‘‘E1 = E1 op E2’’; however, E1 is evaluated only once. Moreover, expressions like ‘‘i =+ p’’ in which a pointer is
added to an integer, are forbidden.
-

C Reference Manual - 9

7.15 expression , expression

A pair of expressions separated by a comma is evaluated left-to-right and the value of the left expression is dis-
carded. The type and value of the result are the type and value of the right operand. This operator groups left-to-
right. It should be avoided in situations where comma is given a special meaning, for example in actual arguments
to function calls (§7.1.6) and lists of initializers (§10.2).

8. Declarations
Declarations are used within function definitions to specify the interpretation which C gives to each identifier;
they do not necessarily reserve storage associated with the identifier. Declarations have the form
declaration:
decl-specifiers declarator-listopt ;
The declarators in the declarator-list contain the identifiers being declared. The decl-specifiers consist of at most
one type-specifier and at most one storage class specifier.
decl-specifiers:
type-specifier
sc-specifier
type-specifier sc-specifier
sc-specifier type-specifier
8.1 Storage class specifiers
The sc-specifiers are:
sc-specifier:
auto
static
extern
register
The auto, static, and register declarations also serve as definitions in that they cause an appropriate
amount of storage to be reserved. In the extern case there must be an external definition (see below) for the given
identifiers somewhere outside the function in which they are declared.
There are some severe restrictions on register identifiers: there can be at most 3 register identifiers in any
function, and the type of a register identifier can only be int, char, or pointer (not float, double, struc-
ture, function, or array). Also the address-of operator & cannot be applied to such identifiers. Except for these re-
strictions (in return for which one is rewarded with faster, smaller code), register identifiers behave as if they were
automatic. In fact implementations of C are free to treat register as synonymous with auto.
If the sc-specifier is missing from a declaration, it is generally taken to be auto.

8.2 Type specifiers

The type-specifiers are
type-specifier:
int
char
float
double
struct { type-decl-list }
struct identifier { type-decl-list }
struct identifier
The struct specifier is discussed in §8.5. If the type-specifier is missing from a declaration, it is generally taken
to be int.
-

C Reference Manual - 10

8.3 Declarators
The declarator-list appearing in a declaration is a comma-separated sequence of declarators.
declarator-list:
declarator
declarator , declarator-list
The specifiers in the declaration indicate the type and storage class of the objects to which the declarators refer.
Declarators have the syntax:
declarator:
identifier
* declarator
declarator ( )
declarator [ constant-expressionopt ]
( declarator )
The grouping in this definition is the same as in expressions.

8.4 Meaning of declarators

Each declarator is taken to be an assertion that when a construction of the same form as the declarator appears in
an expression, it yields an object of the indicated type and storage class. Each declarator contains exactly one identi-
fier; it is this identifier that is declared.
If an unadorned identifier appears as a declarator, then it has the type indicated by the specifier heading the decla-
ration.
If a declarator has the form

*D
for D a declarator, then the contained identifier has the type ‘‘pointer to . . .’’, where ‘‘ . . . ’’ is the type which the
identifier would have had if the declarator had been simply D.
If a declarator has the form
D( )
then the contained identifier has the type ‘‘function returning ...’’, where ‘‘ . . . ’’ is the type which the identifier would
have had if the declarator had been simply D.
A declarator may have the form
D[constant-expression]
or
D[ ]
In the first case the constant expression is an expression whose value is determinable at compile time, and whose
type is int. in the second the constant 1 is used. (Constant expressions are defined precisely in §15.) Such a
declarator makes the contained identifier have type ‘‘array.’’ If the unadorned declarator D would specify a non-ar-
ray of type ‘‘. . .’’, then the declarator ‘‘D[ i ]’’ yields a 1-dimensional array with rank i of objects of type ‘‘. . .’’. If
the unadorned declarator D would specify an n -dimensional array with rank i1 × i2 × . . . × in, then the declarator
‘‘D[ in+1 ]’’ yields an (n +1 ) -dimensional array with rank i1 × i2 × . . . × in × in+1.
An array may be constructed from one of the basic types, from a pointer, from a structure, or from another array
(to generate a multi-dimensional array).
Finally, parentheses in declarators do not alter the type of the contained identifier except insofar as they alter the
binding of the components of the declarator.
Not all the possibilities allowed by the syntax above are actually permitted. The restrictions are as follows: func-
tions may not return arrays, structures or functions, although they may return pointers to such things; there are no ar-
rays of functions, although there may be arrays of pointers to functions. Likewise a structure may not contain a
function, but it may contain a pointer to a function.
-

C Reference Manual - 11

As an example, the declaration

int i, *ip, f ( ), *fip( ), (*pfi) ( );
declares an integer i, a pointer ip to an integer, a function f returning an integer, a function fip returning a pointer to
an integer, and a pointer pfi to a function which returns an integer. Also
float fa[17], *afp[17];
declares an array of float numbers and an array of pointers to float numbers. Finally,
static int x3d[3][5][7];
declares a static three-dimensional array of integers, with rank 3×5×7. In complete detail, x3d is an array of three
items: each item is an array of five arrays; each of the latter arrays is an array of seven integers. Any of the expres-
sions ‘‘x3d’’, ‘‘x3d[ i ]’’, ‘‘x3d[ i ][ j ]’’, ‘‘x3d[ i ][ j ][ k ]’’ may reasonably appear in an expression. The first three
have type ‘‘array’’, the last has type int.

8.5 Structure declarations

Recall that one of the forms for a structure specifier is
struct { type-decl-list }
The type-decl-list is a sequence of type declarations for the members of the structure:
type-decl-list:
type-declaration
type-declaration type-decl-list
A type declaration is just a declaration which does not mention a storage class (the storage class ‘‘member of struc-
ture’’ here being understood by context).
type-declaration:
type-specifier declarator-list ;
Within the structure, the objects declared have addresses which increase as their declarations are read left-to-right.
Each component of a structure begins on an addressing boundary appropriate to its type. On the PDP-11 the only re-
quirement is that non-characters begin on a word boundary; therefore, there may be 1-byte, unnamed holes in a
structure, and all structures have an even length in bytes.
Another form of structure specifier is
struct identifier { type-decl-list }
This form is the same as the one just discussed, except that the identifier is remembered as the structure tag of the
structure specified by the list. A subsequent declaration may then be given using the structure tag but without the
list, as in the third form of structure specifier:
struct identifier
Structure tags allow definition of self-referential structures; they also permit the long part of the declaration to be
given once and used several times. It is however absurd to declare a structure which contains an instance of itself, as
distinct from a pointer to an instance of itself.
A simple example of a structure declaration, taken from §16.2 where its use is illustrated more fully, is
struct tnode {
char tword[20];
int count;
struct tnode *left;
struct tnode *right;
};
which contains an array of 20 characters, an integer, and two pointers to similar structures. Once this declaration has
-

C Reference Manual - 12

been given, the following declaration makes sense:

struct tnode s, *sp;
which declares s to be a structure of the given sort and sp to be a pointer to a structure of the given sort.
The names of structure members and structure tags may be the same as ordinary variables, since a distinction can
be made by context. However, names of tags and members must be distinct. The same member name can appear in
different structures only if the two members are of the same type and if their origin with respect to their structure is
the same; thus separate structures can share a common initial segment.

9. Statements
Except as indicated, statements are executed in sequence.

9.1 Expression statement

Most statements are expression statements, which have the form
expression ;
Usually expression statements are assignments or function calls.

9.2 Compound statement

So that several statements can be used where one is expected, the compound statement is provided:
compound-statement:
{ statement-list }

statement-list:
statement
statement statement-list
9.3 Conditional statement
The two forms of the conditional statement are
if ( expression ) statement
if ( expression ) statement else statement
In both cases the expression is evaluated and if it is non-zero, the first substatement is executed. In the second case
the second substatement is executed if the expression is 0. As usual the ‘‘else’’ ambiguity is resolved by connecting
an else with the last encountered elseless if.

9.4 While statement

The while statement has the form
while ( expression ) statement
The substatement is executed repeatedly so long as the value of the expression remains non-zero. The test takes
place before each execution of the statement.

9.5 Do statement
The do statement has the form
do statement while ( expression ) ;
The substatement is executed repeatedly until the value of the expression becomes zero. The test takes place after
each execution of the statement.
-

C Reference Manual - 13

9.6 For statement

The for statement has the form
for ( expression-1opt ; expression-2opt ; expression-3opt ) statement
This statement is equivalent to
expression-1;
while ( expression-2 ) {
statement
expression-3 ;
}
Thus the first expression specifies initialization for the loop; the second specifies a test, made before each iteration,
such that the loop is exited when the expression becomes 0; the third expression typically specifies an incrementa-
tion which is performed after each iteration.
Any or all of the expressions may be dropped. A missing expression-2 makes the implied while clause equiv-
alent to ‘‘while( 1 )’’; other missing expressions are simply dropped from the expansion above.

9.7 Switch statement

The switch statement causes control to be transferred to one of several statements depending on the value of an
expression. It has the form
switch ( expression ) statement
The expression must be int or char. The statement is typically compound. Each statement within the statement
may be labelled with case prefixes as follows:
case constant-expression :
where the constant expression must be int or char. No two of the case constants in a switch may have the same
value. Constant expressions are precisely defined in §15.
There may also be at most one statement prefix of the form
default :
When the switch statement is executed, its expression is evaluated and compared with each case constant in an un-
defined order. If one of the case constants is equal to the value of the expression, control is passed to the statement
following the matched case prefix. If no case constant matches the expression, and if there is a default prefix,
control passes to the prefixed statement. In the absence of a default prefix none of the statements in the switch is
executed.
Case or default prefixes in themselves do not alter the flow of control.

9.8 Break statement

The statement
break ;
causes termination of the smallest enclosing while, do, for, or switch statement; control passes to the state-
ment following the terminated statement.

9.9 Continue statement

The statement
continue ;
causes control to pass to the loop-continuation portion of the smallest enclosing while, do, or for statement; that
is to the end of the loop. More precisely, in each of the statements
-

C Reference Manual - 14

while ( . . . ) { do { for ( . . . ) {
... ... ...
contin: ; contin: ; contin: ;
} } while ( . . . ) ; }
a continue is equivalent to ‘‘goto contin’’.

9.10 Return statement

A function returns to its caller by means of the return statement, which has one of the forms
return ;
return ( expression ) ;
In the first case no value is returned. In the second case, the value of the expression is returned to the caller of the
function. If required, the expression is converted, as if by assignment, to the type of the function in which it appears.
Flowing off the end of a function is equivalent to a return with no returned value.

9.11 Goto statement

Control may be transferred unconditionally by means of the statement
goto expression ;
The expression should be a label (§§9.12, 14.4) or an expression of type ‘‘pointer to int’’ which evaluates to a la-
bel. It is illegal to transfer to a label not located in the current function unless some extra-language provision has
been made to adjust the stack correctly.

9.12 Labelled statement

Any statement may be preceded by label prefixes of the form
identifier :
which serve to declare the identifier as a label. More details on the semantics of labels are given in §14.4 below.

9.13 Null statement

The null statement has the form
;
A null statement is useful to carry a label just before the ‘‘}’’ of a compound statement or to supply a null body to a
looping statement such as while.

10. External definitions

A C program consists of a sequence of external definitions. External definitions may be given for functions, for
simple variables, and for arrays. They are used both to declare and to reserve storage for objects. An external defi-
nition declares an identifier to have storage class extern and a specified type. The type-specifier (§8.2) may be
empty, in which case the type is taken to be int.

10.1 External function definitions

Function definitions have the form
function-definition:
type-specifieropt function-declarator function-body
A function declarator is similar to a declarator for a ‘‘function returning ...’’ except that it lists the formal parameters
of the function being defined.
function-declarator:
declarator ( parameter-listopt )

parameter-list:
-

C Reference Manual - 15

identifier
identifier , parameter-list
The function-body has the form
function-body:
type-decl-list function-statement
The purpose of the type-decl-list is to give the types of the formal parameters. No other identifiers should be de-
clared in this list, and formal parameters should be declared only here.
The function-statement is just a compound statement which may have declarations at the start.
function-statement:
{ declaration-listopt statement-list }
A simple example of a complete function definition is
int max ( a, b, c)
int a, b, c;
{
int m;
m = ( a > b )? a : b ;
return ( m > c? m : c ) ;
}
Here ‘‘int’’ is the type-specifier; ‘‘max(a, b, c)’’ is the function-declarator; ‘‘int a, b, c;’’ is the type-decl-list for the
formal parameters; ‘‘{ . . . }’’ is the function-statement.
C converts all float actual parameters to double, so formal parameters declared float have their declaration
adjusted to read double. Also, since a reference to an array in any context (in particular as an actual parameter) is
taken to mean a pointer to the first element of the array, declarations of formal parameters declared ‘‘array of ...’’ are
adjusted to read ‘‘pointer to ...’’. Finally, because neither structures nor functions can be passed to a function, it is
useless to declare a formal parameter to be a structure or function (pointers to structures or functions are of course
permitted).
A free return statement is supplied at the end of each function definition, so running off the end causes control,
but no value, to be returned to the caller.

10.2 External data definitions

An external data definition has the form
data-definition:
externopt type-specifieropt init-declarator-listopt ;
The optional extern specifier is discussed in § 11.2. If given, the init-declarator-list is a comma-separated list of
declarators each of which may be followed by an initializer for the declarator.
init-declarator-list:
init-declarator
init-declarator , init-declarator-list

init-declarator:
declarator initializeropt
Each initializer represents the initial value for the corresponding object being defined (and declared).
initializer:
constant
{ constant-expression-list }
-

C Reference Manual - 16

constant-expression-list:
constant-expression
constant-expression , constant-expression-list
Thus an initializer consists of a constant-valued expression, or comma-separated list of expressions, inside braces.
The braces may be dropped when the expression is just a plain constant. The exact meaning of a constant expres-
sion is discussed in §15. The expression list is used to initialize arrays; see below.
The type of the identifier being defined should be compatible with the type of the initializer: a double constant
may initialize a float or double identifier; a non-floating-point expression may initialize an int, char, or
pointer.
An initializer for an array may contain a comma-separated list of compile-time expressions. The length of the ar-
ray is taken to be the maximum of the number of expressions in the list and the square-bracketed constant in the ar-
ray’s declarator. This constant may be missing, in which case 1 is used. The expressions initialize successive mem-
bers of the array starting at the origin (subscript 0) of the array. The acceptable expressions for an array of type ‘‘ar-
ray of ...’’ are the same as those for type ‘‘...’’. As a special case, a single string may be given as the initializer for an
array of chars; in this case, the characters in the string are taken as the initializing values.
Structures can be initialized, but this operation is incompletely implemented and machine-dependent. Basically
the structure is regarded as a sequence of words and the initializers are placed into those words. Structure initializa-
tion, using a comma-separated list in braces, is safe if all the members of the structure are integers or pointers but is
otherwise ill-advised.
The initial value of any externally-defined object not explicitly initialized is guaranteed to be 0.

11. Scope rules

A complete C program need not all be compiled at the same time: the source text of the program may be kept in
several files, and precompiled routines may be loaded from libraries. Communication among the functions of a pro-
gram may be carried out both through explicit calls and through manipulation of external data.
Therefore, there are two kinds of scope to consider: first, what may be called the lexical scope of an identifier,
which is essentially the region of a program during which it may be used without drawing ‘‘undefined identifier’’ di-
agnostics; and second, the scope associated with external identifiers, which is characterized by the rule that refer-
ences to the same external identifier are references to the same object.

11.1 Lexical scope

C is not a block-structured language; this may fairly be considered a defect. The lexical scope of names declared
in external definitions extends from their definition through the end of the file in which they appear. The lexical
scope of names declared at the head of functions (either as formal parameters or in the declarations heading the
statements constituting the function itself) is the body of the function.
It is an error to redeclare identifiers already declared in the current context, unless the new declaration specifies
the same type and storage class as already possessed by the identifiers.

11.2 Scope of externals

If a function declares an identifier to be extern, then somewhere among the files or libraries constituting the
complete program there must be an external definition for the identifier. All functions in a given program which re-
fer to the same external identifier refer to the same object, so care must be taken that the type and extent specified in
the definition are compatible with those specified by each function which references the data.
In PDP-11 C, it is explicitly permitted for (compatible) external definitions of the same identifier to be present in
several of the separately-compiled pieces of a complete program, or even twice within the same program file, with
the important limitation that the identifier may be initialized in at most one of the definitions. In other operating sys-
tems, however, the compiler must know in just which file the storage for the identifier is allocated, and in which file
the identifier is merely being referred to. In the implementations of C for such systems, the appearance of the ex-
tern keyword before an external definition indicates that storage for the identifiers being declared will be allocat-
ed in another file. Thus in a multi-file program, an external data definition without the extern specifier must ap-
pear in exactly one of the files. Any other files which wish to give an external definition for the identifier must in-
clude the extern in the definition. The identifier can be initialized only in the file where storage is allocated.
In PDP-11 C none of this nonsense is necessary and the extern specifier is ignored in external definitions.
-

C Reference Manual - 17

12. Compiler control lines

When a line of a C program begins with the character #, it is interpreted not by the compiler itself, but by a pre-
processor which is capable of replacing instances of given identifiers with arbitrary token-strings and of inserting
named files into the source program. In order to cause this preprocessor to be invoked, it is necessary that the very
first line of the program begin with #. Since null lines are ignored by the preprocessor, this line need contain no oth-
er information.

12.1 Token replacement

A compiler-control line of the form
# define identifier token-string
(note: no trailing semicolon) causes the preprocessor to replace subsequent instances of the identifier with the given
string of tokens (except within compiler control lines). The replacement token-string has comments removed from
it, and it is surrounded with blanks. No rescanning of the replacement string is attempted. This facility is most valu-
able for definition of ‘‘manifest constants’’, as in
# define tabsize 100
...
int table[tabsize];

12.2 File inclusion

Large C programs often contain many external data definitions. Since the lexical scope of external definitions ex-
tends to the end of the program file, it is good practice to put all the external definitions for data at the start of the
program file, so that the functions defined within the file need not repeat tedious and error-prone declarations for
each external identifier they use. It is also useful to put a heavily used structure definition at the start and use its
structure tag to declare the auto pointers to the structure used within functions. To further exploit this technique
when a large C program consists of several files, a compiler control line of the form
# include "filename "
results in the replacement of that line by the entire contents of the file filename.

13. Implicit declarations

It is not always necessary to specify both the storage class and the type of identifiers in a declaration. Sometimes
the storage class is supplied by the context: in external definitions, and in declarations of formal parameters and
structure members. In a declaration inside a function, if a storage class but no type is given, the identifier is assumed
to be int; if a type but no storage class is indicated, the identifier is assumed to be auto. An exception to the latter
rule is made for functions, since auto functions are meaningless (C being incapable of compiling code into the
stack). If the type of an identifier is ‘‘function returning ...’’, it is implicitly declared to be extern.
In an expression, an identifier followed by ( and not currently declared is contextually declared to be ‘‘function
returning int’’.
Undefined identifiers not followed by ( are assumed to be labels which will be defined later in the function.
(Since a label is not an lvalue, this accounts for the ‘‘Lvalue required’’ error message sometimes noticed when an
undeclared identifier is used.) Naturally, appearance of an identifier as a label declares it as such.
For some purposes it is best to consider formal parameters as belonging to their own storage class. In practice, C
treats parameters as if they were automatic (except that, as mentioned above, formal parameter arrays and floats
are treated specially).

14. Types revisited

This section summarizes the operations which can be performed on objects of certain types.
-

C Reference Manual - 18

14.1 Structures
There are only two things that can be done with a structure: pick out one of its members (by means of the . or −>
operators); or take its address (by unary &). Other operations, such as assigning from or to it or passing it as a pa-
rameter, draw an error message. In the future, it is expected that these operations, but not necessarily others, will be
allowed.

14.2 Functions
There are only two things that can be done with a function: call it, or take its address. If the name of a function
appears in an expression not in the function-name position of a call, a pointer to the function is generated. Thus, to
pass one function to another, one might say
int f( );
...
g( f );
Then the definition of g might read
g ( funcp )
int (*funcp) ( );
{
...
(*funcp) ( );
...
}
Notice that f was declared explicitly in the calling routine since its first appearance was not followed by ( .

14.3 Arrays, pointers, and subscripting

Every time an identifier of array type appears in an expression, it is converted into a pointer to the first member of
the array. Because of this conversion, arrays are not lvalues. By definition, the subscript operator [ ] is interpreted
in such a way that ‘‘E1[E2]’’ is identical to ‘‘*( ( E1) + (E2 ) )’’. Because of the conversion rules which apply to +, if
E1 is an array and E2 an integer, then E1[E2] refers to the E2-th member of E1. Therefore, despite its asymmetric
appearance, subscripting is a commutative operation.
A consistent rule is followed in the case of multi-dimensional arrays. If E is an n -dimensional array of rank
i × j × . . . ×k, then E appearing in an expression is converted to a pointer to an (n−1)-dimensional array with rank
j × . . . ×k. If the * operator, either explicitly or implicitly as a result of subscripting, is applied to this pointer, the re-
sult is the pointed-to (n−1)-dimensional array, which itself is immediately converted into a pointer.
For example, consider
int x[3][5];
Here x is a 3×5 array of integers. When x appears in an expression, it is converted to a pointer to (the first of three)
5-membered arrays of integers. In the expression ‘‘x[ i ]’’, which is equivalent to ‘‘*(x+i)’’, x is first converted to a
pointer as described; then i is converted to the type of x, which involves multiplying i by the length the object to
which the pointer points, namely 5 integer objects. The results are added and indirection applied to yield an array
(of 5 integers) which in turn is converted to a pointer to the first of the integers. If there is another subscript the
same argument applies again; this time the result is an integer.
It follows from all this that arrays in C are stored row-wise (last subscript varies fastest) and that the first subscript
in the declaration helps determine the amount of storage consumed by an array but plays no other part in subscript
calculations.

14.4 Labels
Labels do not have a type of their own; they are treated as having type ‘‘array of int’’. Label variables should be
declared ‘‘pointer to int’’; before execution of a goto referring to the variable, a label (or an expression deriving
from a label) should be assigned to the variable.
Label variables are a bad idea in general; the switch statement makes them almost always unnecessary.
-

C Reference Manual - 19

15. Constant expressions

In several places C requires expressions which evaluate to a constant: after case, as array bounds, and in ini-
tializers. In the first two cases, the expression can involve only integer constants, character constants, and sizeof
expressions, possibly connected by the binary operators
+ − * / % & ˆ << >>
or by the unary operators
−
˜
Parentheses can be used for grouping, but not for function calls.
A bit more latitude is permitted for initializers; besides constant expressions as discussed above, one can also ap-
ply the unary & operator to external scalars, and to external arrays subscripted with a constant expression. The unary
& can also be applied implicitly by appearance of unsubscripted external arrays. The rule here is that initializers
must evaluate either to a constant or to the address of an external identifier plus or minus a constant.

16. Examples.
These examples are intended to illustrate some typical C constructions as well as a serviceable style of writing C
programs.

16.1 Inner product

This function returns the inner product of its array arguments.
double inner ( v1, v2, n )
double v1 [ ] , v2 [ ] ;
{
double sum ;
int i ;
sum = 0.0 ;
for ( i=0 ; i<n ; i++ )
sum =+ v1 [ i ] * v2 [ i ] ;
return ( sum ) ;
}
The following version is somewhat more efficient, but perhaps a little less clear. It uses the facts that parameter ar-
rays are really pointers, and that all parameters are passed by value.
double inner ( v1, v2, n )
double *v1, *v2 ;
{
double sum ;
sum = 0.0 ;
while ( n−− )
sum =+ *v1++ * *v2++ ;
return ( sum ) ;
}
The declarations for the parameters are really exactly the same as in the last example. In the first case array declara-
tions ‘‘ [ ] ’’ were given to emphasize that the parameters would be referred to as arrays; in the second, pointer dec-
larations were given because the indirection operator and ++ were used.

16.2 Tree and character processing

Here is a complete C program ( courtesy of R. Haight ) which reads a document and produces an alphabetized list
of words found therein together with the number of occurrences of each word. The method keeps a binary tree of
words such that the left descendant tree for each word has all the words lexicographically smaller than the given
word, and the right descendant has all the larger words. Both the insertion and the printing routine are recursive.
The program calls the library routines getchar to pick up characters and exit to terminate execution. Printf is
-

C Reference Manual - 20

called to print the results according to a format string. A version of printf is given below ( §16.3 ) .
Because all the external definitions for data are given at the top, no extern declarations are necessary within the
functions. To stay within the rules, a type declaration is given for each non-integer function when the function is
used before it is defined. However, since all such functions return pointers which are simply assigned to other point-
ers, no actual harm would result from leaving out the declarations; the supposedly int function values would be as-
signed without error or complaint.
# define nwords 100 /* number of different words */
# define wsize 20 /* max chars per word */
struct tnode { /* the basic structure */
char tword [ wsize ] ;
int count ;
struct tnode *left ;
struct tnode *right ;
};
struct tnode space [ nwords ] ; /* the words themselves */
int nnodes nwords ; /* number of remaining slots */
struct tnode *spacep space ; /* next available slot */
struct tnode *freep ; /* free list */
/*
* The main routine reads words until end-of-file ( ´\0´ returned from "getchar" )
* "tree" is called to sort each word into the tree.
*/
main ( )
{
struct tnode *top, *tree ( ) ;
char c, word [ wsize ] ;
int i ;
i = top = 0 ;
while ( c=getchar ( ) )
if ( ´a´<=c && c<=´z´ ´A´<=c && c <=´Z´ ) {
if ( i<wsize−1 )
word [ i++ ] = c ;
} else
if ( i ) {
word [ i++ ] = ´\0´ ;
top = tree ( top, word ) ;
i = 0;
}
tprint ( top ) ;
}
/*
* The central routine. If the subtree pointer is null, allocate a new node for it.
* If the new word and the node´s word are the same, increase the node´s count.
* Otherwise, recursively sort the word into the left or right subtree according
* as the argument word is less or greater than the node´s word.
*/
struct tnode *tree ( p, word )
struct tnode *p ;
char word [ ] ;
{
struct tnode *alloc ( ) ;
int cond ;
/* Is pointer null? */
if ( p==0 ) {
p = alloc ( ) ;
-

C Reference Manual - 21

copy ( word, p−>tword ) ;

p−>count = 1 ;
p−>right = p−>left = 0 ;
return ( p ) ;
}
/* Is word repeated? */
if ( ( cond=compar ( p−>tword, word ) ) == 0 ) {
p−>count++ ;
return ( p ) ;
}
/* Sort into left or right */
if ( cond<0 )
p−>left = tree ( p−>left, word ) ;
else
p−>right = tree ( p−>right, word ) ;
return ( p ) ;
}
/*
* Print the tree by printing the left subtree, the given node, and the right subtree
*/
tprint ( p )
struct tnode *p ;
{
while ( p ) {
tprint ( p−>left ) ;
printf ( "%d: %s\n", p−>count, p−>tword ) ;
p = p−>right ;
}
}
/*
* String comparison: return number ( >, =, < ) 0
* according as s1 ( >, =, < ) s2.
*/
compar ( s1, s2 )
char *s1, *s2 ;
{
int c1, c2 ;
while ( ( c1 = *s1++ ) == ( c2 = *s2++ ) )
if ( c1==´\0´ )
return ( 0 ) ;
return ( c2−c1 ) ;
}
/*
* String copy: copy s1 into s2 until the null
* character appears.
*/
copy ( s1, s2 )
char *s1, *s2 ;
{
while ( *s2++ = *s1++ ) ;
}
/*
* Node allocation: return pointer to a free node.
* Bomb out when all are gone. Just for fun, there
* is a mechanism for using nodes that have been
* freed, even though no one here calls "free."
*/
struct tnode *alloc ( )
-

C Reference Manual - 22

{
struct tnode *t ;
if ( freep ) {
t = freep ;
freep = freep−>left ;
return ( t ) ;
}
if ( −−nnodes < 0 ) {
printf ( "Out of space\n" ) ;
exit ( ) ;
}
return ( spacep++ ) ;
}
/*
* The uncalled routine which puts a node on the free list.
*/
free ( p )
struct tnode *p ;
{
p−>left = freep ;
freep = p ;
}
To illustrate a slightly different technique of handling the same problem, we will repeat fragments of this example
with the tree nodes treated explicitly as members of an array. The fundamental change is to deal with the subscript
of the array member under discussion, instead of a pointer to it. The struct declaration becomes
struct tnode {
char tword [ wsize ] ;
int count;
int left;
int right;
};
and alloc becomes
alloc ( )
{
int t;
t = −−nnodes;
if ( t<=0 ) {
printf ( "Out of space\n" ) ;
exit ( ) ;
}
return ( t ) ;
}
The free stuff has disappeared because if we deal with exclusively with subscripts some sort of map has to be kept,
which is too much trouble.
Now the tree routine returns a subscript also, and it becomes:
tree ( p, word )
char word [ ] ;
{
int cond;
if ( p==0 ) {
p = alloc ( ) ;
copy ( word, space [ p ] .tword ) ;
-

C Reference Manual - 23

space [ p ] .count = 1;
space [ p ] .right = space [ p ] .left = 0;
return ( p ) ;
}
if ( ( cond=compar ( space [ p ] .tword, word ) ) == 0 ) {
space [ p ] .count++;
return ( p ) ;
}
if ( cond<0 )
space [ p ] .left = tree ( space [ p ] .left, word ) ;
else
space [ p ] .right = tree ( space [ p ] .right, word ) ;
return ( p ) ;
}
The other routines are changed similarly. It must be pointed out that this version is noticeably less efficient than the
first because of the multiplications which must be done to compute an offset in space corresponding to the sub-
scripts.
The observation that subscripts ( like ‘‘a [ i ] ’’ ) are less efficient than pointer indirection ( like ‘‘*ap’’ ) holds true
independently of whether or not structures are involved. There are of course many situations where subscripts are
indispensable, and others where the loss in efficiency is worth a gain in clarity.

16.3 Formatted output

Here is a simplified version of the printf routine, which is available in the C library. It accepts a string ( character
array ) as first argument, and prints subsequent arguments according to specifications contained in this format string.
Most characters in the string are simply copied to the output; two-character sequences beginning with ‘‘%’’ specify
that the next argument should be printed in a style as follows:
%d decimal number
%o octal number
%c ASCII character, or 2 characters if upper character is not null
%s string ( null-terminated array of characters )
%f floating-point number
The actual parameters for each function call are laid out contiguously in increasing storage locations; therefore, a
function with a variable number of arguments may take the address of ( say ) its first argument, and access the re-
maining arguments by use of subscripting ( regarding the arguments as an array ) or by indirection combined with
pointer incrementation.
If in such a situation the arguments have mixed types, or if in general one wishes to insist that an lvalue should be
treated as having a given type, then struct declarations like those illustrated below will be useful. It should be ev-
ident, though, that such techniques are implementation dependent.
Printf depends as well on the fact that char and float arguments are widened respectively to int and dou-
ble, so there are effectively only two sizes of arguments to deal with. Printf calls the library routines putchar to
write out single characters and ftoa to dispose of floating-point numbers.
printf ( fmt, args )
char fmt [ ] ;
{
char *s ;
struct { char **charpp ; };
struct { double *doublep ; };
int *ap, x, c ;
ap = &args ; /* argument pointer */
for ( ; ; ) {
while ( ( c = *fmt++ ) != ´%´ ) {
if ( c == ´\0´ )
return ;
-

C Reference Manual - 24

putchar ( c ) ;
}
switch ( c = *fmt++ ) {
/* decimal */
case ´d ´:
x = *ap++ ;
if ( x < 0 ) {
x = −x ;
if ( x<0 ) { /* is − infinity */
printf ( "−32768" ) ;
continue ;
}
putchar ( ´−´ ) ;
}
printd ( x ) ;
continue ;
/* octal */
case ´o´:
printo ( *ap++ ) ;
continue ;
/* float, double */
case ´f ´:
/* let ftoa do the real work */
ftoa ( *ap.doublep++ ) ;
continue ;
/* character */
case ´c´:
putchar ( *ap++ ) ;
continue ;
/* string */
case ´s´:
s = *ap.charpp++ ;
while ( c = *s++ )
putchar ( c ) ;
continue ;
}
putchar ( c ) ;
}
}
/*
* Print n in decimal ; n must be non-negative
*/
printd ( n )
{
int a ;
if ( a=n/10 )
printd ( a ) ;
putchar ( n%10 + ´0´ ) ;
}
/*
* Print n in octal, with exactly 1 leading 0
*/
printo ( n )
{
if ( n )
printo ( ( n>>3 ) &017777 ) ;
putchar ( ( n&07 ) +´0´ ) ;
}
-

C Reference Manual - 25

REFERENCES

1. Johnson, S. C., and Kernighan, B. W. ‘‘The Programming Language B.’’ Comp. Sci. Tech. Rep. #8., Bell Labo-
ratories, 1972.
2. Ritchie, D. M., and Thompson, K. L. ‘‘The UNIX Time-sharing System.’’ C. ACM 7, 17, July, 1974, pp.
365-375.
3. Peterson, T. G., and Lesk, M. E. ‘‘A User’s Guide to the C Language on the IBM 370.’’ Internal Memoran-
dum, Bell Laboratories, 1974.
4. Thompson, K. L., and Ritchie, D. M. UNIX Programmer’s Manual. Bell Laboratories, 1973.
5. Lesk, M. E., and Barres, B. A. ‘‘The GCOS C Library.’’ Internal memorandum, Bell Laboratories, 1974.
6. Kernighan, B. W. ‘‘Programming in C− A Tutorial.’’ Unpublished internal memorandum, Bell Laboratories,
1974.
-

C Reference Manual - 26

APPENDIX 1
Syntax Summary

1. Expressions.

expression:
primary
* expression
& expression
− expression
! expression
˜ expression
++ lvalue
−− lvalue
lvalue ++
lvalue −−
sizeof expression
expression binop expression
expression ? expression : expression
lvalue asgnop expression
expression , expression
primary:
identifier
constant
string
( expression )
primary ( expression-listopt )
primary [ expression ]
lvalue . identifier
primary —> identifier
lvalue:
identifier
primary [ expression ]
lvalue . identifier
primary —> identifier
* expression
( lvalue )
The primary-expression operators
() [] . —>
have highest priority and group left-to-right. The unary operators
** & − ! ++ −− sizeof
˜
have priority below the primary operators but higher than any binary operator, and group right-to-left. Bina-
ry operators and the conditional operator all group left-to-right, and have priority decreasing as indicated:
binop:
* / %
+ −
>> <<
< > <= >=
== !=
&
-

C Reference Manual - 27

&&

? :
Assignment operators all have the same priority, and all group right-to-left.
asgnop:
= =+ =− =* =/ =% =>> =<< =& =ˆ =
The comma operator has the lowest priority, and groups left-to-right.
2. Declarations.
declaration:
decl-specifiers declarator-listopt ;
decl-specifiers:
type-specifier
sc-specifier
type-specifier sc-specifier
sc-specifier type-specifier
sc-specifier:
auto
static
extern
register
type-specifier:
int
char
float
double
struct { type-decl-list }
struct identifier { type-decl-list }
struct identifier
declarator-list:
declarator
declarator , declarator-list
declarator:
identifier
* declarator
declarator ( )
declarator [ constant-expressionopt ]
( declarator )
type-decl-list:
type-declaration
type-declaration type-decl-list
type-declaration:
type-specifier declarator-list ;

3. Statements.
statement:
expression ;
{ statement-list }
-

C Reference Manual - 28

if ( expression ) statement
if ( expression ) statement else statement
while ( expression ) statement
for ( expressionopt ; expressionopt ; expressionopt ) statement
switch ( expression ) statement
case constant-expression : statement
default : statement
break ;
continue ;
return ;
return ( expression ) ;
goto expression ;
identifier : statement
;
statement-list:
statement
statement statement-list

4. External definitions.
program:
external-definition
external-definition program
external-definition:
function-definition
data-definition
function-definition:
type-specifieropt function-declarator function-body
function-declarator:
declarator ( parameter-listopt )
parameter-list:
identifier
identifier , parameter-list
function-body:
type-decl-list function-statement
function-statement:
{ declaration-listopt statement-list }
data-definition:
externopt type-specifieropt init-declarator-listopt ;
init-declarator-list:
init-declarator
init-declarator , init-declarator-list
init-declarator:
declarator initializeropt
initializer:
constant
{ constant-expression-list }
-

C Reference Manual - 29

constant-expression-list:
constant-expression
constant-expression , constant-expression-list
constant-expression:
expression
5. Preprocessor
# define identifier token-string
# include "filename "
-

C Reference Manual - 30

APPENDIX 2
Implementation Peculiarities

This Appendix briefly summarizes the differences between the implementations of C on the PDP-11 under UNIX and
on the HIS 6070 under GCOS; it includes some known bugs in each implementation. Each entry is keyed by an indi-
cator as follows:

h hard to fix
g GCOS version should probably be changed
u UNIX version should probably be changed
d Inherent difference likely to remain

This list was prepared by M. E. Lesk, S. C. Johnson, E. N. Pinson, and the author.

A. Bugs or differences from C language specifications

hg A.1) GCOS does not do type conversions in ‘‘?:’’.

hg A.2) GCOS has a bug in int and real comparisons; the numbers are compared by subtraction, and
the difference must not overflow.
g A.3) When x is a float, the construction ‘‘test ? −x : x’’ is illegal on GCOS.
hg A.4) ‘‘p1−>p2 =+ 2’’ causes a compiler error, where p1 and p2 are pointers.
u A.5) On UNIX, the expression in a return statement is not converted to the type of the function, as
promised.
hug A.6) entry statement is not implemented at all.

B. Implementation differences

d B.1) Sizes of character constants differ; UNIX: 2, GCOS: 4.

d B.2) Table sizes in compilers differ.
d B.3) chars and ints have different sizes; chars are 8 bits on UNIX, 9 on GCOS; words are 16 bits
on UNIX and 36 on GCOS. There are corresponding differences in representations of floats
and doubles.
d B.4) Character arrays stored left to right in a word in GCOS, right to left in UNIX.
g B.5) Passing of floats and doubles differs; UNIX passes on stack, GCOS passes pointer (hidden to nor-
mal user).
g B.6) Structures and strings are aligned on a word boundary in UNIX, not aligned in GCOS.
g B.7) GCOS preprocessor supports #rename, #escape; UNIX has only #define, #include.
u B.8) Preprocessor is not invoked on UNIX unless first character of file is ‘‘#’’.
u B.9) The external definition ‘‘static int . . .’’ is legal on GCOS, but gets a diagnostic on UNIX. (On
GCOS it means an identifier global to the routines in the file but invisible to routines compiled
separately.)
g B.10) A compound statement on GCOS must contain one ‘‘;’’ but on UNIX may be empty.
g B.11) On GCOS case distinctions in identifiers and keywords are ignored; on UNIX case is significant
everywhere, with keywords in lower case.

C. Syntax Differences

g C.1) UNIX allows broader classes of initialization; on GCOS an initializer must be a constant, name,
or string. Similarly, GCOS is much stickier about wanting braces around initializers and in par-
ticular they must be present for array initialization.
g C.2) ‘‘int extern’’ illegal on GCOS; must have ‘‘extern int’’ (storage class before type).
g C.3) Externals on GCOS must have a type (not defaulted to int).
u C.4) GCOS allows initialization of internal static (same syntax as for external definitions).
g C.5) integer−>... is not allowed on GCOS.
g C.6) Some operators on pointers are illegal on GCOS (<, >).
-

C Reference Manual - 31

g C.7) register storage class means something on UNIX, but is not accepted on GCOS.
g C.8) Scope holes: ‘‘int x; f ( ) {int x;}’’ is illegal on UNIX but defines two variables on GCOS.
g C.9) When function names are used as arguments on UNIX, either ‘‘fname’’ or ‘‘&fname’’ may be
used to get a pointer to the function; on GCOS ‘‘&fname’’ generates a doubly-indirect pointer.
(Note that both are wrong since the ‘‘&’’ is supposed to be supplied for free.)

D. Operating System Dependencies

d D.1) GCOS allocates external scalars by SYMREF; UNIX allocates external scalars as labelled com-
mon; as a result there may be many uninitialized external definitions of the same variable on
UNIX but only one on GCOS.
d D.2) External names differ in allowable length and character set; on UNIX, 7 characters and both cas-
es; on GCOS 6 characters and only one case.

E. Semantic Differences

hg E.1) ‘‘int i, *p; p=i; i=p;’’ does nothing on UNIX, does something on GCOS (destroys right half of i) .
d E.2) ‘‘>>’’ means arithmetic shift on UNIX, logical on GCOS.
d E.3) When a char is converted to integer, the result is always positive on GCOS but can be negative
on UNIX.
d E.4) Arguments of subroutines are evaluated left-to-right on GCOS, right-to-left on UNIX.
 Programming in C — A Tutorial

Brian W. Kernighan

Bell Laboratories, Murray Hill, N. J.

1. Introduction
C is a computer language available on the GCOS and UNIX operating systems at Murray Hill and (in
preliminary form) on OS/360 at Holmdel. C lets you write your programs clearly and simply — it has de-
cent control flow facilities so your code can be read straight down the page, without labels or GOTO’s; it
lets you write code that is compact without being too cryptic; it encourages modularity and good program
organization; and it provides good data-structuring facilities.
This memorandum is a tutorial to make learning C as painless as possible. The first part concentrates
on the central features of C; the second part discusses those parts of the language which are useful (usually
for getting more efficient and smaller code) but which are not necessary for the new user. This is not a ref-
erence manual. Details and special cases will be skipped ruthlessly, and no attempt will be made to cover
every language feature. The order of presentation is hopefully pedagogical instead of logical. Users who
would like the full story should consult the C Reference Manual by D. M. Ritchie [1], which should be read
for details anyway. Runtime support is described in [2] and [3]; you will have to read one of these to learn
how to compile and run a C program.
We will assume that you are familiar with the mysteries of creating files, text editing, and the like in
the operating system you run on, and that you have programmed in some language before.

2. A Simple C Program

main( ) {
printf("hello, world");
}

A C program consists of one or more functions, which are similar to the functions and subroutines of
a Fortran program or the procedures of PL/I, and perhaps some external data definitions. main is such a
.
function, and in fact all C programs must have a main Execution of the program begins at the first state-
.
ment of main main will usually invoke other functions to perform its job, some coming from the same
program, and others from libraries.
One method of communicating data between functions is by arguments. The parentheses following
the function name surround the argument list; here main is a function of no arguments, indicated by ( ).
The {} enclose the statements of the function. Individual statements end with a semicolon but are other-
wise free-format.
printf is a library function which will format and print output on the terminal (unless some other
destination is specified). In this case it prints
C Tutorial -2-

hello, world
A function is invoked by naming it, followed by a list of arguments in parentheses. There is no CALL state-
ment as in Fortran or PL/I.

3. A Working C Program; Variables; Types and Type Declarations

Here’s a bigger program that adds three integers and prints their sum.
main( ) {
int a, b, c, sum;
a = 1; b = 2; c = 3;
sum = a + b + c;
printf("sum is %d", sum);
}

Arithmetic and the assignment statements are much the same as in Fortran (except for the semi-
colons) or PL/I. The format of C programs is quite free. We can put several statements on a line if we want,
or we can split a statement among several lines if it seems desirable. The split may be between any of the
operators or variables, but not in the middle of a name or operator. As a matter of style, spaces, tabs, and
newlines should be used freely to enhance readability.
C has four fundamental types of variables:
int integer (PDP-11: 16 bits; H6070: 36 bits; IBM360: 32 bits)
char one byte character (PDP-11, IBM360: 8 bits; H6070: 9 bits)
float single-precision floating point
double double-precision floating point
There are also arrays and structures of these basic types, pointers to them and functions that return them,
all of which we will meet shortly.
All variables in a C program must be declared, although this can sometimes be done implicitly by
context. Declarations must precede executable statements. The declaration
int a, b, c, sum;
declares a, b, c, and sum to be integers.
Variable names have one to eight characters, chosen from A-Z, a-z, 0-9, and , and start with a non-
digit. Stylistically, it’s much better to use only a single case and give functions and external variables
names that are unique in the first six characters. (Function and external variable names are used by various
assemblers, some of which are limited in the size and case of identifiers they can handle.) Furthermore,
keywords and library functions may only be recognized in one case.

4. Constants
We have already seen decimal integer constants in the previous example — 1, 2, and 3. Since C is
often used for system programming and bit-manipulation, octal numbers are an important part of the lan-
guage. In C, any number that begins with 0 (zero!) is an octal integer (and hence can’t have any 8’s or 9’s
in it). Thus 0777 is an octal constant, with decimal value 511.
A ‘‘character’’ is one byte (an inherently machine-dependent concept). Most often this is expressed
as a character constant, which is one character enclosed in single quotes. However, it may be any quantity
that fits in a byte, as in flags below:
char quest, newline, flags;
quest = ′?′;
newline = ′\n′;
flags = 077;
C Tutorial -3-

The sequence ‘\n’ is C notation for ‘‘newline character’’, which, when printed, skips the terminal to
the beginning of the next line. Notice that ‘\n’ represents only a single character. There are several other
‘‘escapes’’ like ‘\n’ for representing hard-to-get or invisible characters, such as ‘\t’ for tab, ‘\b’ for
backspace, ‘\0’ for end of file, and ‘\\’ for the backslash itself.
float and double constants are discussed in section 26.

5. Simple I/O — getchar, putchar, printf

main( ) {
char c;
c = getchar( );
putchar(c);
}

getchar and putchar are the basic I/O library functions in C. getchar fetches one character
from the standard input (usually the terminal) each time it is called, and returns that character as the value
of the function. When it reaches the end of whatever file it is reading, thereafter it returns the character rep-
resented by ‘\0’ (ascii NUL, which has value zero). We will see how to use this very shortly.
putchar puts one character out on the standard output (usually the terminal) each time it is called.
So the program above reads one character and writes it back out. By itself, this isn’t very interesting, but
observe that if we put a loop around this, and add a test for end of file, we have a complete program for
copying one file to another.
printf is a more complicated function for producing formatted output. We will talk about only the
simplest use of it. Basically, printf uses its first argument as formatting information, and any successive
arguments as variables to be output. Thus
printf ("hello, world\n");
is the simplest use — the string ‘‘hello, world\n’’ is printed out. No formatting information, no variables,
so the string is dumped out verbatim. The newline is necessary to put this out on a line by itself. (The con-
struction
"hello, world\n"
is really an array of chars . More about this shortly.)
More complicated, if sum is 6,
printf ("sum is %d\n", sum);
prints
sum is 6
Within the first argument of printf, the characters ‘‘%d’’ signify that the next argument in the argument
list is to be printed as a base 10 number.
Other useful formatting commands are ‘‘%c’’ to print out a single character, ‘‘%s’’ to print out an en-
tire string, and ‘‘%o’’ to print a number as octal instead of decimal (no leading zero). For example,
n = 511;
printf ("What is the value of %d in octal?", n);
printf (" %s! %d decimal is %o octal\n", "Right", n, n);
prints
What is the value of 511 in octal? Right! 511 decimal is 777
octal
Notice that there is no newline at the end of the first output line. Successive calls to printf (and/or
putchar, for that matter) simply put out characters. No newlines are printed unless you ask for them.
Similarly, on input, characters are read one at a time as you ask for them. Each line is generally terminated
C Tutorial -4-

by a newline (\n), but there is otherwise no concept of record.

6. If; relational operators; compound statements

The basic conditional-testing statement in C is the if statement:
c = getchar( );
if( c == ′?′ )
printf("why did you type a question mark?\n");
The simplest form of if is
if (expression) statement

The condition to be tested is any expression enclosed in parentheses. It is followed by a statement.

The expression is evaluated, and if its value is non-zero, the statement is executed. There’s an optional
else clause, to be described soon.
The character sequence ‘==’ is one of the relational operators in C; here is the complete set:
== equal to ( EQ . .
to Fortraners)
!= not equal to
> greater than
< less than
>= greater than or equal to
<= less than or equal to

The value of ‘‘expression relation expression’’ is 1 if the relation is true, and 0 if

false. Don’t forget that the equality test is ‘==’; a single ‘=’ causes an assignment, not a test, and invariably
leads to disaster.
Tests can be combined with the operators ‘&&’ (AND), ‘ ’ (OR), and ‘!’ (NOT). For example, we
can test whether a character is blank or tab or newline with
if( c==′ ′ c==′\t′ c==′\n′ ) ...
C guarantees that ‘&&’ and ‘ ’ are evaluated left to right — we shall soon see cases where this matters.
One of the nice things about C is that the statement part of an if can be made arbitrarily compli-
cated by enclosing a set of statements in {}. As a simple example, suppose we want to ensure that a is big-
ger than b, as part of a sort routine. The interchange of a and b takes three statements in C, grouped to-
gether by {}:
if (a < b) {
t = a;
a = b;
b = t;
}

As a general rule in C, anywhere you can use a simple statement, you can use any compound state-
ment, which is just a number of simple or compound ones enclosed in {}. There is no semicolon after the }
of a compound statement, but there is a semicolon after the last non-compound statement inside the {}.
The ability to replace single statements by complex ones at will is one feature that makes C much
more pleasant to use than Fortran. Logic (like the exchange in the previous example) which would require
several GOTO’s and labels in Fortran can and should be done in C without any, using compound state-
ments.

7. While Statement; Assignment within an Expression; Null Statement

The basic looping mechanism in C is the while statement. Here’s a program that copies its input to
its output a character at a time. Remember that ‘\0’ marks the end of file.
main( ) {
C Tutorial − 5 −

char c;
while( (c=getchar( )) != ′\0′ )
putchar(c);
}
The while statement is a loop, whose general form is
while (expression) statement
Its meaning is
(a) evaluate the expression
(b) if its value is true (i e , not zero) ..
do the statement, and go back to (a)
Because the expression is tested before the statement is executed, the statement part can be executed zero
times, which is often desirable. As in the if statement, the expression and the statement can both be arbi-
trarily complicated, although we haven’t seen that yet. Our example gets the character, assigns it to c, and
then tests if it’s a ‘\0’’. If it is not a ‘\0’, the statement part of the while is executed, printing the character.
The while then repeats. When the input character is finally a ‘\0’, the while terminates, and so does
main .
Notice that we used an assignment statement
c = getchar( )
within an expression. This is a handy notational shortcut which often produces clearer code. (In fact it is
often the only way to write the code cleanly. As an exercise, re-write the file-copy without using an assign-
ment inside an expression.) It works because an assignment statement has a value, just as any other expres-
sion does. Its value is the value of the right hand side. This also implies that we can use multiple assign-
ments like
x = y = z = 0;
Evaluation goes from right to left.
By the way, the extra parentheses in the assignment statement within the conditional were really nec-
essary: if we had said
c = getchar( ) != ′\0′
c would be set to 0 or 1 depending on whether the character fetched was an end of file or not. This is be-
cause in the absence of parentheses the assignment operator ‘=’ is evaluated after the relational operator
‘!=’. When in doubt, or even if not, parenthesize.
Since putchar(c) returns c as its function value, we could also copy the input to the output by
nesting the calls to getchar and putchar:
main( ) {
while( putchar(getchar( )) != ′\0′ ) ;
}
What statement is being repeated? None, or technically, the null statement, because all the work is really
done within the test part of the while . This version is slightly different from the previous one, because
the final ‘\0’ is copied to the output before we decide to stop.

8. Arithmetic
The arithmetic operators are the usual ‘+’, ‘−’, ‘*’, and ‘/’ (truncating integer division if the operands
are both int), and the remainder or mod operator ‘%’:
x = a%b;
sets x to the remainder after a is divided by b (i.e., a mod b) . The results are machine dependent unless
C Tutorial -6-

a and b are both positive.

In arithmetic, char variables can usually be treated like int variables. Arithmetic on characters is
quite legal, and often makes sense:
c = c + ′A′ − ′a′;
converts a single lower case ascii character stored in c to upper case, making use of the fact that corre-
sponding ascii letters are a fixed distance apart. The rule governing this arithmetic is that all chars are
converted to int before the arithmetic is done. Beware that conversion may involve sign-extension — if
the leftmost bit of a character is 1, the resulting integer might be negative. (This doesn’t happen with gen-
uine characters on any current machine.)
So to convert a file into lower case:
main( ) {
char c;
while( (c=getchar( )) != ′\0′ )
if( ′A′<=c && c<=′Z′ )
putchar(c+′a′−′A′);
else
putchar(c);
}
Characters have different sizes on different machines. Further, this code won’t work on an IBM machine,
because the letters in the ebcdic alphabet are not contiguous.

9. Else Clause; Conditional Expressions

We just used an else after an if . The most general form of if is
if (expression) statement1 else statement2
the else part is optional, but often useful. The canonical example sets x to the minimum of a and b:
if (a < b)
x = a;
else
x = b;
Observe that there’s a semicolon after x=a .
C provides an alternate form of conditional which is often more concise. It is called the ‘‘conditional
expression’’ because it is a conditional which actually has a value and can be used anywhere an expression
can. The value of
a<b ? a : b;
is a if a is less than b; it is b otherwise. In general, the form
expr1 ? expr2 : expr3
means ‘‘evaluate expr1 . If it is not zero, the value of the whole thing is expr2; otherwise the value is
.
expr3 ’’
To set x to the minimum of a and b, then:
x = (a<b ? a : b);
The parentheses aren’t necessary because ‘?:’ is evaluated before ‘=’, but safety first.
Going a step further, we could write the loop in the lower-case program as
while( (c=getchar( )) != ′\0′ )
putchar( (′A′<=c && c<=′Z′) ? c−′A′+′a′ : c );
C Tutorial -7-

If’s and else’s can be used to construct logic that branches one of several ways and then rejoins,
a common programming structure, in this way:
if(...)
{...}
else if(...)
{...}
else if(...)
{...}
else
{...}
The conditions are tested in order, and exactly one block is executed — either the first one whose if is sat-
.
isfied, or the one for the last else When this block is finished, the next statement executed is the one af-
. .
ter the last else If no action is to be taken for the ‘‘default’’ case, omit the last else
For example, to count letters, digits and others in a file, we could write
main( ) {
int let, dig, other, c;
let = dig = other = 0;
while( (c=getchar( )) != ′\0′ )
if( (′A′<=c && c<=′Z′) (′a′<=c && c<=′z′) ) ++let;
else if( ′0′<=c && c<=′9′ ) ++dig;
else ++other;
printf("%d letters, %d digits, %d others\n", let, dig, other);
}
The ‘++’ operator means ‘‘increment by 1’’; we will get to it in the next section.

10. Increment and Decrement Operators

In addition to the usual ‘−’, C also has two other interesting unary operators, ‘++’ (increment) and
‘−−’ (decrement). Suppose we want to count the lines in a file.
main( ) {
int c,n;
n = 0;
while( (c=getchar( )) != ′\0′ )
if( c == ′\n′ )
++n;
printf("%d lines\n", n);
}
++n is equivalent to n=n+1 but clearer, particularly when n is a complicated expression. ‘++’ and ‘−−’
can be applied only to int’s and char’s (and pointers which we haven’t got to yet).
The unusual feature of ‘++’ and ‘−−’ is that they can be used either before or after a variable. The
value of ++k is the value of k after it has been incremented. The value of k++ is k before it is increment-
ed. Suppose k is 5. Then
x = ++k;
increments k to 6 and then sets x to the resulting value, i.e., to 6. But
x = k++;
first sets x to to 5, and then increments k to 6. The incrementing effect of ++k and k++ is the same, but
their values are respectively 5 and 6. We shall soon see examples where both of these uses are important.
C Tutorial -8-

11. Arrays
In C, as in Fortran or PL/I, it is possible to make arrays whose elements are basic types. Thus we can
make an array of 10 integers with the declaration
int x[10];
The square brackets mean subscripting; parentheses are used only for function references. Array indexes
begin at zero, so the elements of x are
x[0], x[1], x[2], ..., x[9]
If an array has n elements, the largest subscript is n−1.
Multiple-dimension arrays are provided, though not much used above two dimensions. The declara-
tion and use look like
int name[10] [20];
n = name[i+j] [1] + name[k] [2];
Subscripts can be arbitrary integer expressions. Multi-dimension arrays are stored by row (opposite to For-
tran), so the rightmost subscript varies fastest; name has 10 rows and 20 columns.
Here is a program which reads a line, stores it in a buffer, and prints its length (excluding the newline
at the end).
main( ) {
int n, c;
char line[100];
n = 0;
while( (c=getchar( )) != ′\n′ ) {
if( n < 100 )
line[n] = c;
n++;
}
printf("length = %d\n", n);
}

As a more complicated problem, suppose we want to print the count for each line in the input, still
storing the first 100 characters of each line. Try it as an exercise before looking at the solution:
main( ) {
int n, c; char line[100];
n = 0;
while( (c=getchar( )) != ′\0′ )
if( c == ′\n′ ) {
printf("%d0, n);
n = 0;
}
else {
if( n < 100 ) line[n] = c;
n++;
}
}

12. Character Arrays; Strings

Text is usually kept as an array of characters, as we did with line[ ] in the example above. By
convention in C, the last character in a character array should be a ‘\0’ because most programs that manipu-
late character arrays expect it. For example, printf uses the ‘\0’ to detect the end of a character array
C Tutorial -9-

when printing it out with a ‘%s’.

We can copy a character array s into another t like this:
i = 0;
while( (t[i]=s[i]) != ′\0′ )
i++;

Most of the time we have to put in our own ‘\0’ at the end of a string; if we want to print the line with
printf, it’s necessary. This code prints the character count before the line:
main( ) {
int n;
char line[100];
n = 0;
while( (line[n++]=getchar( )) != ′\n′ );
line[n] = ′\0′;
printf("%d:\t%s", n, line);
}
Here we increment n in the subscript itself, but only after the previous value has been used. The character
is read, placed in line[n], and only then n is incremented.
There is one place and one place only where C puts in the ‘\0’ at the end of a character array for you,
and that is in the construction
"stuff between double quotes"
The compiler puts a ‘\0’ at the end automatically. Text enclosed in double quotes is called a string; its
properties are precisely those of an (initialized) array of characters.

13. For Statement

The for statement is a somewhat generalized while that lets us put the initialization and increment
parts of a loop into a single statement along with the test. The general form of the for is
for( initialization; expression; increment )
statement
The meaning is exactly
initialization;
while( expression ) {
statement
increment;
}
Thus, the following code does the same array copy as the example in the previous section:
for( i=0; (t[i]=s[i]) != ′\0′; i++ );
This slightly more ornate example adds up the elements of an array:
sum = 0;
for( i=0; i<n; i++)
sum = sum + array[i];

In the for statement, the initialization can be left out if you want, but the semicolon has to be there.
The increment is also optional. It is not followed by a semicolon. The second clause, the test, works the
same way as in the while: if the expression is true (not zero) do another loop, otherwise get on with the
next statement. As with the while, the for loop may be done zero times. If the expression is left out, it
is taken to be always true, so
for( ; ; ) ...
C Tutorial - 10 -

and
while( 1 ) ...
are both infinite loops.
You might ask why we use a for since it’s so much like a while . (You might also ask why we
use a while because...) The for is usually preferable because it keeps the code where it’s used and
sometimes eliminates the need for compound statements, as in this code that zeros a two-dimensional array:
for( i=0; i<n; i++ )
for( j=0; j<m; j++ )
array[i][j] = 0;

14. Functions; Comments

Suppose we want, as part of a larger program, to count the occurrences of the ascii characters in some
input text. Let us also map illegal characters (those with value>127 or <0) into one pile. Since this is pre-
sumably an isolated part of the program, good practice dictates making it a separate function. Here is one
way:
main( ) {
int hist[129]; /* 128 legal chars + 1 illegal group */
...
count(hist, 128); /* count the letters into hist */
printf( ...); /* comments look like this; use them */
... /* anywhere blanks, tabs or newlines could appear */
}
count(buf, size)
int size, buf[ ]; {
int i, c;
for( i=0; i<=size; i++ )
buf[i] = 0; /* set buf to zero */
while( (c=getchar( )) != ′\0′ ) { /* read til eof */
if( c > size c < 0 )
c = size; /* fix illegal input */
buf[c]++;
}
return;
}
We have already seen many examples of calling a function, so let us concentrate on how to define one.
Since count has two arguments, we need to declare them, as shown, giving their types, and in the case of
buf, the fact that it is an array. The declarations of arguments go between the argument list and the open-
.
ing ‘{’. There is no need to specify the size of the array buf, for it is defined outside of count
The return statement simply says to go back to the calling routine. In fact, we could have omitted
it, since a return is implied at the end of a function.
What if we wanted count to return a value, say the number of characters read? The return state-
ment allows for this too:
int i, c, nchar;
nchar = 0;
...
while( (c=getchar( )) != ′\0′ ) {
if( c > size c < 0 )
c = size;
buf[c]++;
nchar++;
C Tutorial − 11 −

}
return(nchar);
Any expression can appear within the parentheses. Here is a function to compute the minimum of two inte-
gers:
min(a, b)
int a, b; {
return( a < b ? a : b );
}

To copy a character array, we could write the function

strcopy(s1, s2) /* copies s1 to s2 */
char s1[ ], s2[ ]; {
int i;
for( i = 0; (s2[i] = s1[i]) != ′\0′; i++ );
}
As is often the case, all the work is done by the assignment statement embedded in the test part of the
for . Again, the declarations of the arguments s1 and s2 omit the sizes, because they don’t matter to
.
strcopy (In the section on pointers, we will see a more efficient way to do a string copy.)
There is a subtlety in function usage which can trap the unsuspecting Fortran programmer. Simple
variables (not arrays) are passed in C by ‘‘call by value’’, which means that the called function is given a
copy of its arguments, and doesn’t know their addresses. This makes it impossible to change the value of
one of the actual input arguments.
There are two ways out of this dilemma. One is to make special arrangements to pass to the function
the address of a variable instead of its value. The other is to make the variable a global or external variable,
which is known to each function by its name. We will discuss both possibilities in the next few sections.

15. Local and External Variables

If we say
f( ) {
int x;
...
}
g( ) {
int x;
...
}
.
each x is local to its own routine — the x in f is unrelated to the x in g (Local variables are also called
‘‘automatic’’.) Furthermore each local variable in a routine appears only when the function is called, and
disappears when the function is exited. Local variables have no memory from one call to the next and must
be explicitly initialized upon each entry. (There is a static storage class for making local variables with
memory; we won’t discuss it.)
As opposed to local variables, external variables are defined external to all functions, and are (poten-
tially) available to all functions. External storage always remains in existence. To make variables external
we have to define them external to all functions, and, wherever we want to use them, make a declaration.
main( ) {
extern int nchar, hist[ ];
...
count( );
...
}
C Tutorial − 12 −

count( ) {
extern int nchar, hist[ ];
int i, c;
...
}
int hist[129]; /* space for histogram */
int nchar; /* character count */
Roughly speaking, any function that wishes to access an external variable must contain an extern decla-
.
ration for it. The declaration is the same as others, except for the added keyword extern Furthermore,
there must somewhere be a definition of the external variables external to all functions.
External variables can be initialized; they are set to zero if not explicitly initialized. In its simplest
form, initialization is done by putting the value (which must be a constant) after the definition:
int nchar 0;
char flag ′f′;
etc .
This is discussed further in a later section.

This ends our discussion of what might be called the central core of C. You now have enough to
write quite substantial C programs, and it would probably be a good idea if you paused long enough to do
so. The rest of this tutorial will describe some more ornate constructions, useful but not essential.

16. Pointers
A pointer in C is the address of something. It is a rare case indeed when we care what the specific
address itself is, but pointers are a quite common way to get at the contents of something. The unary opera-
tor ‘&’ is used to produce the address of an object, if it has one. Thus
int a, b;
b = &a;
.
puts the address of a into b We can’t do much with it except print it or pass it to some other routine, be-
cause we haven’t given b the right kind of declaration. But if we declare that b is indeed a pointer to an in-
teger, we’re in good shape:
int a, *b, c;
b = &a;
c = *b;
b contains the address of a and ‘c = *b’ means to use the value in b as an address, i.e., as a pointer. The
effect is that we get back the contents of a, albeit rather indirectly. (It’s always the case that ‘*&x’ is the
same as x if x has an address.)
The most frequent use of pointers in C is for walking efficiently along arrays. In fact, in the imple-
mentation of an array, the array name represents the address of the zeroth element of the array, so you can’t
use it on the left side of an expression. (You can’t change the address of something by assigning to it.) If
we say
char *y;
char x[100];
y is of type pointer to character (although it doesn’t yet point anywhere). We can make y point to an ele-
ment of x by either of
y = &x[0];
y = x;
C Tutorial - 13 -

Since x is the address of x[0] this is legal and consistent.

Now ‘*y’ gives x[0] . More importantly,

*(y+1) gives x[1]

*(y+i) gives x[i]
and the sequence
y = &x[0];
y++;
leaves y pointing at x[1] .
Let’s use pointers in a function length that computes how long a character array is. Remember
that by convention all character arrays are terminated with a ‘\0’. (And if they aren’t, this program will
blow up inevitably.) The old way:
length(s)
char s[ ]; {
int n;
for( n=0; s[n] != ′\0′; )
n++;
return(n);
}
Rewriting with pointers gives
length(s)
char *s; {
int n;
for( n=0; *s != ′\0′; s++ )
n++;
return(n);
}
You can now see why we have to say what kind of thing s points to — if we’re to increment it with s++ we
have to increment it by the right amount.
The pointer version is more efficient (this is almost always true) but even more compact is
for( n=0; *s++ != ′\0′; n++ );
The ‘*s’ returns a character; the ‘++’ increments the pointer so we’ll get the next character next time
around. As you can see, as we make things more efficient, we also make them less clear. But ‘*s++’ is
an idiom so common that you have to know it.
Going a step further, here’s our function strcopy that copies a character array s to another t .
strcopy(s,t)
char *s, *t; {
while(*t++ = *s++);
}
We have omitted the test against ‘\0’, because ‘\0’ is identically zero; you will often see the code this way.
(You must have a space after the ‘=’: see section 25.)
For arguments to a function, and there only, the declarations
char s[ ];
char *s;
are equivalent — a pointer to a type, or an array of unspecified size of that type, are the same thing.
If this all seems mysterious, copy these forms until they become second nature. You don’t often need
anything more complicated.
C Tutorial - 14 -

17. Function Arguments

Look back at the function strcopy in the previous section. We passed it two string names as argu-
ments, then proceeded to clobber both of them by incrementation. So how come we don’t lose the original
strings in the function that called strcopy?
As we said before, C is a ‘‘call by value’’ language: when you make a function call like f(x), the
.
value of x is passed, not its address. So there’s no way to alter x from inside f If x is an array (char
x[10]) this isn’t a problem, because x is an address anyway, and you’re not trying to change it, just what
it addresses. This is why strcopy works as it does. And it’s convenient not to have to worry about mak-
ing temporary copies of the input arguments.
But what if x is a scalar and you do want to change it? In that case, you have to pass the address of x
to f, and then use it as a pointer. Thus for example, to interchange two integers, we must write
flip(x, y)
int *x, *y; {
int temp;
temp = *x;
*x = *y;
*y = temp;
}
and to call flip, we have to pass the addresses of the variables:
flip (&a, &b);

18. Multiple Levels of Pointers; Program Arguments

When a C program is called, the arguments on the command line are made available to the main pro-
gram as an argument count argc and an array of character strings argv containing the arguments. Ma-
nipulating these arguments is one of the most common uses of multiple levels of pointers (‘‘pointer to
pointer to ...’’). By convention, argc is greater than zero; the first argument (in argv[0]) is the com-
mand name itself.
Here is a program that simply echoes its arguments.
main(argc, argv)
int argc;
char **argv; {
int i;
for( i=1; i < argc; i++ )
printf("%s ", argv[i]);
putchar(′\n′);
}
Step by step: main is called with two arguments, the argument count and the array of arguments. argv is
a pointer to an array, whose individual elements are pointers to arrays of characters. The zeroth argument is
the name of the command itself, so we start to print with the first argument, until we’ve printed them all.
Each argv[i] is a character array, so we use a ‘%s’ in the printf .
You will sometimes see the declaration of argv written as
char *argv[ ];
which is equivalent. But we can’t use char argv[ ][ ], because both dimensions are variable and there
would be no way to figure out how big the array is.
.
Here’s a bigger example using argc and argv A common convention in C programs is that if the
first argument is ‘−’, it indicates a flag of some sort. For example, suppose we want a program to be
callable as
prog −abc arg1 arg2 ...
C Tutorial - 15 -

where the ‘−’ argument is optional; if it is present, it may be followed by any combination of a, b, and c.
main(argc, argv)
int argc;
char **argv; {
...
aflag = bflag = cflag = 0;
if( argc > 1 && argv[1][0] == ′−′ ) {
for( i=1; (c=argv[1][i]) != ′\0′; i++ )
if( c==′a′ )
aflag++;
else if( c==′b′ )
bflag++;
else if( c==′c′ )
cflag++;
else
printf("%c?\n", c);
−−argc;
++argv;
}
...
There are several things worth noticing about this code. First, there is a real need for the left-to-right
evaluation that && provides; we don’t want to look at argv[1] unless we know it’s there. Second, the
statements
−−argc;
++argv;
let us march along the argument list by one position, so we can skip over the flag argument as if it had never
existed — the rest of the program is independent of whether or not there was a flag argument. This only
works because argv is a pointer which can be incremented.

19. The Switch Statement; Break; Continue

The switch statement can be used to replace the multi-way test we used in the last example. When
the tests are like this:
if( c == ′a′ ) ...
else if( c == ′b′ ) ...
else if( c == ′c′ ) ...
else ...
testing a value against a series of constants, the switch statement is often clearer and usually gives better
code. Use it like this:
switch( c ) {

case ′a′:
aflag++;
break;
case ′b′:
bflag++;
break;
case ′c′:
cflag++;
break;
default:
printf("%c?\n", c);
C Tutorial − 16 −

break;
}
The case statements label the various actions we want; default gets done if none of the other cases are
satisfied. (A default is optional; if it isn’t there, and none of the cases match, you just fall out the bot-
tom.)
The break statement in this example is new. It is there because the cases are just labels, and after
you do one of them, you fall through to the next unless you take some explicit action to escape. This is a
mixed blessing. On the positive side, you can have multiple cases on a single statement; we might want to
allow both upper and lower case letters in our flag field, so we could say
case ′a′: case ′A′: ...
case ′b′: case ′B′: ...
etc .
But what if we just want to get out after doing case ‘a’ ? We could get out of a case of the switch
with a label and a goto, but this is really ugly. The break statement lets us exit without either goto or
label.
switch( c ) {

case ′a′:
aflag++;
break;
case ′b′:
bflag++;
break;
...
}
/* the break statements get us here directly */
The break statement also works in for and while statements — it causes an immediate exit from the
loop.
The continue statement works only inside for’s and while’s; it causes the next iteration of
the loop to be started. This means it goes to the increment part of the for and the test part of the while .
We could have used a continue in our example to get on with the next iteration of the for, but it seems
clearer to use break instead.

20. Structures
The main use of structures is to lump together collections of disparate variable types, so they can
conveniently be treated as a unit. For example, if we were writing a compiler or assembler, we might need
for each identifier information like its name (a character array), its source line number (an integer), some
type information (a character, perhaps), and probably a usage count (another integer).
char id[10];
int line;
char type;
int usage;

We can make a structure out of this quite easily. We first tell C what the structure will look like, that
is, what kinds of things it contains; after that we can actually reserve storage for it, either in the same state-
ment or separately. The simplest thing is to define it and allocate storage all at once:
struct {
char id[10];
int line;
char type;
int usage;
C Tutorial − 17 −

} sym;

This defines sym to be a structure with the specified shape; id, line, type and usage are mem-
bers of the structure. The way we refer to any particular member of the structure is
structure-name . member
as in
.
sym type = 077;
.
if( sym usage == 0 ) ...
.
while( sym id[j++] ) ...
etc.
Although the names of structure members never stand alone, they still have to be unique — there can’t be
another id or usage in some other structure.
So far we haven’t gained much. The advantages of structures start to come when we have arrays of
structures, or when we want to pass complicated data layouts between functions. Suppose we wanted to
make a symbol table for up to 100 identifiers. We could extend our definitions like
char id[100][10];
int line[100];
char type[100];
int usage[100];
but a structure lets us rearrange this spread-out information so all the data about a single identifer is collect-
ed into one lump:
struct {
char id[10];
int line;
char type;
int usage;
} sym[100];
This makes sym an array of structures; each array element has the specified shape. Now we can refer to
members as
.
sym[i] usage++; /* increment usage of i-th identifier */
.
for( j=0; sym[i] id[j++] != ′\0′; ) ...
etc.
Thus to print a list of all identifiers that haven’t been used, together with their line number,
for( i=0; i<nsym; i++ )
.
if( sym[i] usage == 0 )
printf("%d\t%s\n", sym[i] line, sym[i] id); . .
Suppose we now want to write a function lookup(name) which will tell us if name already exists
in sym, by giving its index, or that it doesn’t, by returning a −1. We can’t pass a structure to a function di-
rectly — we have to either define it externally, or pass a pointer to it. Let’s try the first way first.
int nsym 0; /* current length of symbol table */
struct {
char id[10];
int line;
char type;
int usage;
} sym[100]; /* symbol table */
main( ) {
...
C Tutorial − 18 −

if( (index = lookup(newname)) >= 0 )

sym[index] usage++;. /* already there ... */
else
install(newname, newline, newtype);
...
}
lookup(s)
char *s; {
int i;
extern struct {
char id[10];
int line;
char type;
int usage;
} sym[ ];
for( i=0; i<nsym; i++ )
if( compar(s, sym[i] id) > 0 ) .
return(i);
return(−1);
}

compar(s1,s2) /* return 1 if s1==s2, 0 otherwise */

char *s1, *s2; {
while( *s1++ == *s2 )
if( *s2++ == ′\0′ )
return(1);
return(0);
}
The declaration of the structure in lookup isn’t needed if the external definition precedes its use in the
same source file, as we shall see in a moment.
Now what if we want to use pointers?
struct symtag {
char id[10];
int line;
char type;
int usage;
} sym[100], *psym;

psym = &sym[0]; /* or p = sym; */

This makes psym a pointer to our kind of structure (the symbol table), then initializes it to point to the first
element of sym .
Notice that we added something after the word struct: a ‘‘tag’’ called symtag . This puts a
name on our structure definition so we can refer to it later without repeating the definition. It’s not neces-
sary but useful. In fact we could have said
struct symtag {
... structure definition
};
which wouldn’t have assigned any storage at all, and then said
struct symtag sym[100];
struct symtag *psym;
C Tutorial - 19 -

which would define the array and the pointer. This could be condensed further, to
struct symtag sym[100], *psym;
The way we actually refer to an member of a structure by a pointer is like this:
ptr −> structure-member
The symbol ‘−>’ means we’re pointing at a member of a structure; ‘−>’ is only used in that context. ptr is
a pointer to the (base of) a structure that contains the structure member. The expression ptr−>struc-
ture-member refers to the indicated member of the pointed-to structure. Thus we have constructions
like:
psym−>type = 1;
psym−>id[0] = ′a′;
and so on.
For more complicated pointer expressions, it’s wise to use parentheses to make it clear who goes with
what. For example,
struct { int x, *y; } *p;
p−>x++ increments x
++p−>x so does this!
(++p)−>x increments p before getting x
* p−>y++ uses y as a pointer, then increments it
* (p−>y)++ so does this
* (p++)−>y uses y as a pointer, then increments p
The way to remember these is that −>, . (dot), ( ) and [ ] bind very tightly. An expression involving
.
one of these is treated as a unit. p−>x, a[i], y x and f(b) are names exactly as abc is.
If p is a pointer to a structure, any arithmetic on p takes into account the acutal size of the structure.
For instance, p++ increments p by the correct amount to get the next element of the array of structures.
But don’t assume that the size of a structure is the sum of the sizes of its members — because of alignments
of different sized objects, there may be ‘‘holes’’ in a structure.
Enough theory. Here is the lookup example, this time with pointers.
struct symtag {
char id[10];
int line;
char type;
int usage;
} sym[100];
main( ) {
struct symtag *lookup( );
struct symtag *psym;
...
if( (psym = lookup(newname)) ) /* non−zero pointer */
psym −> usage++; /* means already there */
else
install(newname, newline, newtype);
...
}
struct symtag *lookup(s)
char *s; {
struct symtag *p;
for( p=sym; p < &sym[nsym]; p++ )
if( compar(s, p−>id) > 0)
return(p);
C Tutorial − 20 −

return(0);
}
The function compar doesn’t change: ‘p−>id’ refers to a string.
In main we test the pointer returned by lookup against zero, relying on the fact that a pointer is by
definition never zero when it really points at something. The other pointer manipulations are trivial.
The only complexity is the set of lines like
struct symtag *lookup( );
This brings us to an area that we will treat only hurriedly — the question of function types. So far, all of
our functions have returned integers (or characters, which are much the same). What do we do when the
function returns something else, like a pointer to a structure? The rule is that any function that doesn’t re-
turn an int has to say explicitly what it does return. The type information goes before the function name
(which can make the name hard to see). Examples:
char f(a)
int a; {
...
}

int *g( ) { ... }

struct symtag *lookup(s) char *s; { ... }

The function f returns a character, g returns a pointer to an integer, and lookup returns a pointer to a
structure that looks like symtag . And if we’re going to use one of these functions, we have to make a
declaration where we use it, as we did in main above.
Notice th parallelism between the declarations
struct symtag *lookup( );
struct symtag *psym;
In effect, this says that lookup( ) and psym are both used the same way — as a pointer to a strcture —
even though one is a variable and the other is a function.

21. Initialization of Variables

An external variable may be initialized at compile time by following its name with an initializing val-
ue when it is defined. The initializing value has to be something whose value is known at compile time,
like a constant.
int x 0; /* "0" could be any constant */
int a ′a′;
char flag 0177;
int *p &y[1]; /* p now points to y[1] */
An external array can be initialized by following its name with a list of initializations enclosed in braces:
int x[4] {0,1,2,3}; /* makes x[i] = i */
int y[ ] {0,1,2,3}; /* makes y big enough for 4 values */
char * msg "syntax error\n"; /* braces unnecessary here */
char * keyword[ ]{
"if",
"else",
"for",
"while",
"break",
"continue",
0
C Tutorial − 21 −

};
This last one is very useful — it makes keyword an array of pointers to character strings, with a zero at
the end so we can identify the last element easily. A simple lookup routine could scan this until it either
finds a match or encounters a zero keyword pointer:
lookup(str) /* search for str in keyword[ ] */
char *str; {
int i,j,r;
for( i=0; keyword[i] != 0; i++) {
for( j=0; (r=keyword[i][j]) == str[j] && r != ′\0′; j++ );
if( r == str[j] )
return(i);
}
return(−1);
}

Sorry — neither local variables nor structures can be initialized.

22. Scope Rules: Who Knows About What

A complete C program need not be compiled all at once; the source text of the program may be kept
in several files, and previously compiled routines may be loaded from libraries. How do we arrange that
data gets passed from one routine to another? We have already seen how to use function arguments and
values, so let us talk about external data. Warning: the words declaration and definition are used precisely
in this section; don’t treat them as the same thing.
A major shortcut exists for making extern declarations. If the definition of a variable appears be-
fore its use in some function, no extern declaration is needed within the function. Thus, if a file contains
f1( ) { ... }
int foo;
f2( ) { ... foo = 1; ... }
f3( ) { ... if ( foo ) ... }
no declaration of foo is needed in either f2 or or f3, because the external definition of foo appears be-
fore them. But if f1 wants to use foo, it has to contain the declaration
f1( ) {
extern int foo;
...
}

This is true also of any function that exists on another file — if it wants foo it has to use an extern
declaration for it. (If somewhere there is an extern declaration for something, there must also eventually
be an external definition of it, or you’ll get an ‘‘undefined symbol’’ message.)
There are some hidden pitfalls in external declarations and definitions if you use multiple source files.
To avoid them, first, define and initialize each external variable only once in the entire set of files:
int foo 0;
You can get away with multiple external definitions on UNIX, but not on GCOS, so don’t ask for trouble.
Multiple initializations are illegal everywhere. Second, at the beginning of any file that contains functions
needing a variable whose definition is in some other file, put in an extern declaration, outside of any
function:
extern int foo;
f1( ) { ... }
etc .
C Tutorial − 22 −

The #include compiler control line, to be discussed shortly, lets you make a single copy of the ex-
ternal declarations for a program and then stick them into each of the source files making up the program.

23. #define, #include

C provides a very limited macro facility. You can say
#define name something
and thereafter anywhere ‘‘name’’ appears as a token, ‘‘something’’ will be substituted. This is particularly
useful in parametering the sizes of arrays:
#define ARRAYSIZE 100
int arr[ARRAYSIZE];
...
while( i++ < ARRAYSIZE ) ...
(now we can alter the entire program by changing only the define) or in setting up mysterious constants:
#define SET 01
#define INTERRUPT 02 /* interrupt bit */
#define ENABLED 04
...
if( x & (SET | INTERRUPT | ENABLED) ) ...
Now we have meaningful words instead of mysterious constants. (The mysterious operators ‘&’ (AND)
and ‘ ’ (OR) will be covered in the next section.) It’s an excellent practice to write programs without any
literal constants except in #define statements.
There are several warnings about #define . First, there’s no semicolon at the end of a #de-
fine; all the text from the name to the end of the line (except for comments) is taken to be the ‘‘some-
thing’’. When it’s put into the text, blanks are placed around it. Good style typically makes the name in the
#define upper case — this makes parameters more visible. Definitions affect things only after they oc-
cur, and only within the file in which they occur. Defines can’t be nested. Last, if there is a #define in a
file, then the first character of the file must be a ‘#’, to signal the preprocessor that definitions exist.

The other control word known to C is #include . To include one file in your source at compila-
tion time, say
#include "filename"
This is useful for putting a lot of heavily used data definitions and #define statements at the beginning of
a file to be compiled. As with #define, the first line of a file containing a #include has to begin with
a ‘#’. And #include can’t be nested — an included file can’t contain another #include .
24. Bit Operators
C has several operators for logical bit-operations. For example,
x = x & 0177;
forms the bit-wise AND of x and 0177, effectively retaining only the last seven bits of x . Other operators
are
inclusive OR
ˆ (circumflex) exclusive OR
˜ (tilde) 1′s complement
! logical NOT
<< left shift (as in x<<2)
>> right shift (arithmetic on PDP-11; logical on H6070, IBM360)
C Tutorial - 23 -

25. Assignment Operators

An unusual feature of C is that the normal binary operators like ‘+’, ‘−’, etc. can be combined with
the assignment operator ‘=’ to form new assignment operators. For example,
x =− 10;
uses the assignment operator ‘=−’ to decrement x by 10, and
x =& 0177
forms the AND of x and 0177. This convention is a useful notational shortcut, particularly if x is a compli-
cated expression. The classic example is summing an array:
for( sum=i=0; i<n; i++ )
sum =+ array[i];
But the spaces around the operator are critical! For instance,
x = −10;
sets x to −10, while
x =− 10;
subtracts 10 from x . When no space is present,
x=−10;
also decreases x by 10. This is quite contrary to the experience of most programmers. In particular, watch
out for things like
c=*s++;
y=&x[0];
both of which are almost certainly not what you wanted. Newer versions of various compilers are courte-
ous enough to warn you about the ambiguity.
Because all other operators in an expression are evaluated before the assignment operator, the order
of evaluation should be watched carefully:
x = x<<y | z;
.
means ‘‘shift x left y places, then OR with z, and store in x ’’ But
x =<< y | z;
means ‘‘shift x left by y|z places’’, which is rather different.

26. Floating Point

We’ve skipped over floating point so far, and the treatment here will be hasty. C has single and dou-
ble precision numbers (where the precision depends on the machine at hand). For example,
double sum;
float avg, y[10];
sum = 0 0; .
for( i=0; i<n; i++ )
sum =+ y[i];
avg = sum/n;
forms the sum and average of the array y .
All floating arithmetic is done in double precision. Mixed mode arithmetic is legal; if an arithmetic
operator in an expression has both operands int or char, the arithmetic done is integer, but if one
operand is int or char and the other is float or double, both operands are converted to double .
Thus if i and j are int and x is float,
(x+i)/j converts i and j to float
x + i/j does i/j integer, then converts
C Tutorial − 24 −

Type conversion may be made by assignment; for instance,

int m, n;
float x, y;
m = x;
y = n;
converts x to integer (truncating toward zero), and n to floating point.
Floating constants are just like those in Fortran or PL/I, except that the exponent letter is ‘e’ instead
of ‘E’. Thus:
.
pi = 3 14159;
.
large = 1 23456789e10;

.
printf will format floating point numbers: ‘‘%w df’’ in the format string will print the corre-
sponding variable in a field w digits wide, with d decimal places. An e instead of an f will produce expo-
nential notation.

27. Horrors! goto’s and labels

C has a goto statement and labels, so you can branch about the way you used to. But most of the
time goto’s aren’t needed. (How many have we used up to this point?) The code can almost always be
more clearly expressed by for/while, if/else, and compound statements.
One use of goto’s with some legitimacy is in a program which contains a long loop, where a
while(1) would be too extended. Then you might write
mainloop:
...
goto mainloop;
Another use is to implement a break out of more than one level of for or while . goto’s can only
branch to labels within the same function.

28. Acknowledgements
I am indebted to a veritable host of readers who made valuable criticisms on several drafts of this tu-
torial. They ranged in experience from complete beginners through several implementors of C compilers to
the C language designer himself. Needless to say, this is a wide enough spectrum of opinion that no one is
satisfied (including me); comments and suggestions are still welcome, so that some future version might be
improved.

References
C is an extension of B, which was designed by D. M. Ritchie and K. L. Thompson [4]. The C lan-
guage design and UNIX implementation are the work of D. M. Ritchie. The GCOS version was begun by A.
Snyder and B. A. Barres, and completed by S. C. Johnson and M. E. Lesk. The IBM version is primarily
due to T. G. Peterson, with the assistance of M. E. Lesk.
[1] D. M. Ritchie, C Reference Manual. Bell Labs, Jan. 1974.
[2] M. E. Lesk & B. A. Barres, The GCOS C Library. Bell Labs, Jan. 1974.
C Tutorial - 25 -

[3] D. M. Ritchie & K. Thompson, UNIX Programmer’s Manual. 5th Edition, Bell Labs,
1974.
[4] S. C. Johnson & B. W. Kernighan, The Programming Language B. Computer Sci-
ence Technical Report 8, Bell Labs, 1972.
 UNIX Assembler Reference Manual

Dennis M. Ritchie

Bell Laboratories
Murray Hill, New Jersey

0. Introduction
This document describes the usage and input syntax of the UNIX PDP-11 assembler as. The details of the PDP-11 are
not described; consult the DEC documents ‘‘PDP-11/20 Handbook’’ and ‘‘PDP-11/45 Handbook.’’
The input syntax of the UNIX assembler is generally similar to that of the DEC assembler PAL-11R, although its inter-
nal workings and output format are unrelated. It may be useful to read the publication DEC-11-ASDB-D, which de-
scribes PAL-11R, although naturally one must use care in assuming that its rules apply to as.
As is a rather ordinary two-pass assembler without macro capabilities. It produces an output file which contains re-
location information and a complete symbol table; thus the output is acceptable to the UNIX link-editor ld, which
may be used to combine the outputs of several assembler runs and to obtain object programs from libraries. The
output format has been designed so that if a program contains no unresolved references to external symbols, it is ex-
ecutable without further processing.

1. Usage
as is used as follows:

as [ — ] file1 . . .

If the optional ‘‘—’’ argument is given, all undefined symbols in the current assembly will be made undefined-exter-
nal. See the .globl directive below.
The other arguments name files which are concatenated and assembled. Thus programs may be written in several
pieces and assembled together.
The output of the assembler is placed on the file a.out in the current directory. If there were no unresolved external
references, and no errors detected, a.out is made executable; otherwise, if it is produced at all, it is made non-ex-
ecutable.

2. Lexical conventions
Assembler tokens include identifiers (alternatively, ‘‘symbols’’ or ‘‘names’’), temporary symbols, constants, and op-
erators.

2.1 Identifiers
An identifier consists of a sequence of alphanumeric characters (including period ‘‘ . ’’, underscore ‘‘ ’’, and tilde
‘‘˜’’ as alphanumeric) of which the first may not be numeric. Only the first eight characters are significant. When a
name begins with a tilde, the tilde is discarded and that occurrence of the identifier generates a unique entry in the
symbol table which can match no other occurrence of the identifier. This feature is used by the C compiler to place
names of local variables in the output symbol table without having to worry about making them unique.

2.2 Temporary symbols

-

Assembler Manual - 2

A temporary symbol consists of a digit followed by ‘‘f ’’ or ‘‘b’’. Temporary symbols are discussed fully in §5.1.

2.3 Constants
An octal constant consists of a sequence of digits; ‘‘8’’ and ‘‘9’’ are taken to have octal value 10 and 11. The con-
stant is truncated to 16 bits and interpreted in two’s complement notation.
A decimal constant consists of a sequence of digits terminated by a decimal point ‘‘.’’. The magnitude of the con-
stant should be representable in 15 bits; i.e., be less than 32,768.
A single-character constant consists of a single quote ‘‘ ′ ’’ followed by an ASCII character not a new-line. Certain
dual-character escape sequences are acceptable in place of the ASCII character to represent new-line and other non-
graphics (see String statements, §5.5). The constant’s value has the code for the given character in the least signifi-
cant byte of the word and is null-padded on the left.
A double-character constant consists of a double quote ‘‘ " ’’ followed by a pair of ASCII characters not including
new-line. Certain dual-character escape sequences are acceptable in place of either of the ASCII characters to repre-
sent new-line and other non-graphics (see String statements, §5.5). The constant’s value has the code for the first
given character in the least significant byte and that for the second character in the most significant byte.

2.4 Operators
There are several single- and double-character operators; see §6.

2.5 Blanks
Blank and tab characters may be interspersed freely between tokens, but may not be used within tokens (except char-
acter constants). A blank or tab is required to separate adjacent identifiers or constants not otherwise separated.

2.6 Comments
The character ‘‘ / ’’ introduces a comment, which extends through the end of the line on which it appears. Com-
ments are ignored by the assembler.

3. Segments
Assembled code and data fall into three segments: the text segment, the data segment, and the bss segment. The text
segment is the one in which the assembler begins, and it is the one into which instructions are typically placed. The
UNIX system will, if desired, enforce the purity of the text segment of programs by trapping write operations into it.
Object programs produced by the assembler must be processed by the link-editor ld (using its ‘‘—n’’ flag) if the text
segment is to be write-protected. A single copy of the text segment is shared among all processes executing such a
program.
The data segment is available for placing data or instructions which will be modified during execution. Anything
which may go in the text segment may be put into the data segment. In programs with write-protected, sharable text
segments, data segment contains the initialized but variable parts of a program. If the text segment is not pure, the
data segment begins immediately after the text segment; if the text segment is pure, the data segment begins at the
lowest 8K byte boundary after the text segment.
The bss segment may not contain any explicitly initialized code or data. The length of the bss segment (like that of
text or data) is determined by the high-water mark of the location counter within it. The bss segment is actually an
extension of the data segment and begins immediately after it. At the start of execution of a program, the bss seg-
ment is set to 0. Typically the bss segment is set up by statements exemplified by

lab: . = .+10

The advantage in using the bss segment for storage that starts off empty is that the initialization information need not
be stored in the output file. See also Location counter and Assignment statements below.

4. The location counter

One special symbol, ‘‘ . ’’, is the location counter. Its value at any time is the offset within the appropriate segment
of the start of the statement in which it appears. The location counter may be assigned to, with the restriction that
the current segment may not change; furthermore, the value of ‘‘ . ’’ may not decrease. If the effect of the assign-
-

Assembler Manual - 3

ment is to increase the value of ‘‘ . ’’, the required number of null bytes are generated (but see Segments above).

5. Statements
A source program is composed of a sequence of statements. Statements are separated either by new-lines or by
semicolons. There are five kinds of statements: null statements, expression statements, assignment statements, string
statements, and keyword statements.
Any kind of statement may be preceded by one or more labels.

5.1 Labels
There are two kinds of label: name labels and numeric labels. A name label consists of a name followed by a colon
( : ). The effect of a name label is to assign the current value and type of the location counter ‘‘ . ’’ to the name. An
error is indicated in pass 1 if the name is already defined; an error is indicated in pass 2 if the ‘‘ . ’’ value assigned
changes the definition of the label.
A numeric label consists of a digit 0 to 9 followed by a colon ( : ). Such a label serves to define temporary symbols
of the form ‘‘n b’’ and ‘‘n f ’’, where n is the digit of the label. As in the case of name labels, a numeric label assigns
the current value and type of ‘‘ . ’’ to the temporary symbol. However, several numeric labels with the same digit
may be used within the same assembly. References of the form ‘‘n f ’’ refer to the first numeric label ‘‘n :’’ f orward
from the reference; ‘‘nb’’ symbols refer to the first ‘‘n :’’ label backward from the reference. This sort of temporary
label was introduced by Knuth [The Art of Computer Programming, Vol I: Fundamental Algorithms ]. Such labels
tend to conserve both the symbol table space of the assembler and the inventive powers of the programmer.

5.2 Null statements

A null statement is an empty statement (which may, however, have labels). A null statement is ignored by the as-
sembler. Common examples of null statements are empty lines or lines containing only a label.

5.3 Expression statements

An expression statement consists of an arithmetic expression not beginning with a keyword. The assembler com-
putes its (16-bit) value and places it in the output stream, together with the appropriate relocation bits.

5.4 Assignment statements

An assignment statement consists of an identifier, an equals sign ( = ), and an expression. The value and type of the
expression are assigned to the identifier. It is not required that the type or value be the same in pass 2 as in pass 1,
nor is it an error to redefine any symbol by assignment.
Any external attribute of the expression is lost across an assignment. This means that it is not possible to declare a
global symbol by assigning to it, and that it is impossible to define a symbol to be offset from a non-locally defined
global symbol.
As mentioned, it is permissible to assign to the location counter ‘‘ . ’’. It is required, however, that the type of the ex-
pression assigned be of the same type as ‘‘ . ’’, and it is forbidden to decrease the value of ‘‘ . ’’. In practice, the most
common assignment to ‘‘ . ’’ has the form ‘‘. = . + n’’ for some number n; this has the effect of generating n null
bytes.

5.5 String statements

A string statement generates a sequence of bytes containing ASCII characters. A string statement consists of a left
string quote ‘‘<’’ followed by a sequence of ASCII characters not including newline, followed by a right string quote
‘‘>’’. Any of the ASCII characters may be replaced by a two-character escape sequence to represent certain non-
graphic characters, as follows:
-

Assembler Manual - 4

\n NL (012)
\t HT (011)
\e EOT (004)
\0 NUL (000)
\r CR (015)
\a ACK (006)
\p PFX (033)
\\ \
\> >

The last two are included so that the escape character and the right string quote may be represented. The same es-
cape sequences may also be used within single- and double-character constants (see §2.3 above).

5.6 Keyword statements

Keyword statements are numerically the most common type, since most machine instructions are of this sort. A
keyword statement begins with one of the many predefined keywords of the assembler; the syntax of the remainder
depends on the keyword. All the keywords are listed below with the syntax they require.

6. Expressions
An expression is a sequence of symbols representing a value. Its constituents are identifiers, constants, temporary
symbols, operators, and brackets. Each expression has a type.
All operators in expressions are fundamentally binary in nature; if an operand is missing on the left, a 0 of absolute
type is assumed. Arithmetic is two’s complement and has 16 bits of precision. All operators have equal precedence,
and expressions are evaluated strictly left to right except for the effect of brackets.

6.1 Expression operators

The operators are:

(blank) when there is no operator between operands, the effect is exactly the same as if a ‘‘+’’ had ap-
peared.
+ addition
— subtraction
* multiplication
\/ division (note that plain ‘‘ / ’’ starts a comment)
& bitwise and
 bitwise or
>> logical right shift
<< logical left shift
% modulo
! a ! b is a or ( not b ); i.e., the or of the first operand and the one’s complement of the second;
most common use is as a unary.
ˆ result has the value of first operand and the type of the second; most often used to define new
machine instructions with syntax identical to existing instructions.

Expressions may be grouped by use of square brackets ‘‘ [ ] ’’. (Round parentheses are reserved for address modes.)

6.2 Types
The assembler deals with a number of types of expressions. Most types are attached to keywords and used to select
the routine which treats that keyword. The types likely to be met explicitly are:
-

Assembler Manual - 5

undefined
Upon first encounter, each symbol is undefined. It may become undefined if it is assigned an unde-
fined expression. It is an error to attempt to assemble an undefined expression in pass 2; in pass 1, it
is not (except that certain keywords require operands which are not undefined).
undefined external
A symbol which is declared .globl but not defined in the current assembly is an undefined external. If
such a symbol is declared, the link editor ld must be used to load the assembler’s output with another
routine that defines the undefined reference.
absolute
An absolute symbol is one defined ultimately from a constant. Its value is unaffected by any possible
future applications of the link-editor to the output file.
text
The value of a text symbol is measured with respect to the beginning of the text segment of the pro-
gram. If the assembler output is link-edited, its text symbols may change in value since the program
need not be the first in the link editor’s output. Most text symbols are defined by appearing as labels.
At the start of an assembly, the value of ‘‘ . ’’ is text 0.
data
The value of a data symbol is measured with respect to the origin of the data segment of a program.
Like text symbols, the value of a data symbol may change during a subsequent link-editor run since
previously loaded programs may have data segments. After the first .data statement, the value of
‘‘ . ’’ is data 0.
bss
The value of a bss symbol is measured from the beginning of the bss segment of a program. Like text
and data symbols, the value of a bss symbol may change during a subsequent link-editor run, since
previously loaded programs may have bss segments. After the first .bss statement, the value of ‘‘ . ’’
is bss 0.
external absolute, text, data, or bss
symbols declared .globl but defined within an assembly as absolute, text, data, or bss symbols may be
used exactly as if they were not declared .globl; however, their value and type are available to the link
editor so that the program may be loaded with others that reference these symbols.
register
The symbols
r0 . . . r5
fr0 . . . fr5
sp
pc
are predefined as register symbols. Either they or symbols defined from them must be used to refer to
the six general-purpose, six floating-point, and the 2 special-purpose machine registers. The behavior
of the floating register names is identical to that of the corresponding general register names; the for-
mer are provided as a mnemonic aid.
other types
Each keyword known to the assembler has a type which is used to select the routine which processes
the associated keyword statement. The behavior of such symbols when not used as keywords is the
same as if they were absolute.

6.3 Type propagation in expressions

When operands are combined by expression operators, the result has a type which depends on the types of the
operands and on the operator. The rules involved are complex to state but were intended to be sensible and pre-
dictable. For purposes of expression evaluation the important types are
undefined
absolute
text
-

Assembler Manual - 6

data
bss
undefined external
other
The combination rules are then: If one of the operands is undefined, the result is undefined. If both operands are ab-
solute, the result is absolute. If an absolute is combined with one of the ‘‘other types’’ mentioned above, or with a
register expression, the result has the register or other type. As a consequence, one can refer to r3 as ‘‘r0+3’’. If two
operands of ‘‘other type’’ are combined, the result has the numerically larger type (not that this fact is very useful,
since the values are not made public). An ‘‘other type’’ combined with an explicitly discussed type other than abso-
lute acts like an absolute.
Further rules applying to particular operators are:

+ If one operand is text-, data-, or bss-segment relocatable, or is an undefined external, the result has
the postulated type and the other operand must be absolute.
— If the first operand is a relocatable text-, data-, or bss-segment symbol, the second operand may be
absolute (in which case the result has the type of the first operand); or the second operand may
have the same type as the first (in which case the result is absolute). If the first operand is external
undefined, the second must be absolute. All other combinations are illegal.
ˆ This operator follows no other rule than that the result has the value of the first operand and the
type of the second.
others It is illegal to apply these operators to any but absolute symbols.

7. Pseudo-operations
The keywords listed below introduce statements which generate data in unusual forms or influence the later opera-
tions of the assembler. The metanotation
[ stuff ] . . .
means that 0 or more instances of the given stuff may appear. Also, boldface tokens are literals, italic words are sub-
stitutable.

7.1 .byte expression [ , expression ] . . .

The expressions in the comma-separated list are truncated to 8 bits and assembled in successive bytes. The expres-
sions must be absolute. This statement and the string statement above are the only ones which assemble data one
byte at at time.

7.2 .even
If the location counter ‘‘ . ’’ is odd, it is advanced by one so the next statement will be assembled at a word boundary.

7.3 .if expression

The expression must be absolute and defined in pass 1. If its value is nonzero, the .if is ignored; if zero, the state-
ments between the .if and the matching .endif (below) are ignored. .if may be nested. The effect of .if cannot ex-
tend beyond the end of the input file in which it appears. (The statements are not totally ignored, in the following
sense: .ifs and .endifs are scanned for, and moreover all names are entered in the symbol table. Thus names occur-
ring only inside an .if will show up as undefined if the symbol table is listed.)

7.4 .endif
This statement marks the end of a conditionally-assembled section of code. See .if above.

7.5 .globl name [ , name ] . . .

This statement makes the names external. If they are otherwise defined (by assignment or appearance as a label)
they act within the assembly exactly as if the .globl statement were not given; however, the link editor ld may be
used to combine this routine with other routines that refer these symbols.
-

Assembler Manual - 7

Conversely, if the given symbols are not defined within the current assembly, the link editor can combine the output
of this assembly with that of others which define the symbols.
As discussed in §1, it is possible to force the assembler to make all otherwise undefined symbols external.

7.6 .text
7.7 .data
7.8 .bss
These three pseudo-operations cause the assembler to begin assembling into the text, data, or bss segment respec-
tively. Assembly starts in the text segment. It is forbidden to assemble any code or data into the bss segment, but
symbols may be defined and ‘‘ . ’’ moved about by assignment.

7.9 .comm name , expression

Provided the name is not defined elsewhere, this statement is equivalent to
.globl name
name = expression ˆ name
That is, the type of name is ‘‘undefined external’’, and its value is expression. In fact the name behaves in the cur-
rent assembly just like an undefined external. However, the link-editor ld has been special-cased so that all external
symbols which are not otherwise defined, and which have a non-zero value, are defined to lie in the bss segment, and
enough space is left after the symbol to hold expression bytes. All symbols which become defined in this way are
located before all the explicitly defined bss-segment locations.

8. Machine instructions
Because of the rather complicated instruction and addressing structure of the PDP-11, the syntax of machine instruc-
tion statements is varied. Although the following sections give the syntax in detail, the 11/20 and 11/45 handbooks
should be consulted on the semantics.

8.1 Sources and Destinations

The syntax of general source and destination addresses is the same. Each must have one of the following forms,
where reg is a register symbol, and expr is any sort of expression:

syntax words mode

reg 0 0+reg
( reg ) + 0 2+reg
— ( reg ) 0 4+reg
expr ( reg ) 1 6+reg
( reg ) 0 1+reg
* reg 0 1+reg
* ( reg ) + 0 3+reg
* — ( reg ) 0 5+reg
* ( reg ) 1 7+reg
* expr ( reg ) 1 7+reg
expr 1 67
$ expr 1 27
* expr 1 77
* $ expr 1 37

The words column gives the number of address words generated; the mode column gives the octal address-mode
number. The syntax of the address forms is identical to that in DEC assemblers, except that ‘‘*’’ has been substituted
for ‘‘@’’ and ‘‘$’’ for ‘‘#’’; the UNIX typing conventions make ‘‘@’’ and ‘‘#’’ rather inconvenient.
Notice that mode ‘‘*reg’’ is identical to ‘‘(reg)’’; that ‘‘*(reg)’’ generates an index word (namely, 0); and that ad-
dresses consisting of an unadorned expression are assembled as pc-relative references independent of the type of the
expression. To force a non-relative reference, the form ‘‘*$expr’’ can be used, but notice that further indirection is
-

Assembler Manual - 8

impossible.

8.3 Simple machine instructions

The following instructions are defined as absolute symbols:
clc
clv
clz
cln
sec
sev
sez
sen
They therefore require no special syntax. The PDP-11 hardware allows more than one of the ‘‘clear’’ class, or alter-
natively more than one of the ‘‘set’’ class to be or-ed together; this may be expressed as follows:
clc  clv

8.4 Branch
The following instructions take an expression as operand. The expression must lie in the same segment as the refer-
ence, cannot be undefined-external, and its value cannot differ from the current location of ‘‘ . ’’ by more than 254
bytes:
br blos
bne bvc
beq bvs
bge bhis
blt bec (= bcc)
bgt bcc
ble blo
bpl bcs
bmi bes (= bcs)
bhi
bes (‘‘branch on error set’’) and bec (‘‘branch on error clear’’) are intended to test the error bit returned by system
calls (which is the c-bit).

8.5 Extended branch instructions

The following symbols are followed by an expression representing an address in the same segment as ‘‘ . ’’. If the
target address is close enough, a branch-type instruction is generated; if the address is too far away, a jmp will be
used.
jbr jlos
jne jvc
jeq jvs
jge jhis
jlt jec
jgt jcc
jle jlo
jpl jcs
jmi jes
jhi
jbr turns into a plain jmp if its target is too remote; the others (whose names are contructed by replacing the ‘‘b’’ in
the branch instruction’s name by ‘‘j’’ ) turn into the converse branch over a jmp to the target address.

8.6 Single operand instructions

The following symbols are names of single-operand machine instructions. The form of address expected is dis-
cussed in §8.1 above.
-

Assembler Manual - 9

clr sbcb
clrb ror
com rorb
comb rol
inc rolb
incb asr
dec asrb
decb asl
neg aslb
negb jmp
adc swab
adcb tst
sbc tstb

8.7 Double operand instructions

The following instructions take a general source and destination (§8.1), separated by a comma, as operands.
mov
movb
cmp
cmpb
bit
bitb
bic
bicb
bis
bisb
add
sub

8.8 Miscellaneous instructions

The following instructions have more specialized syntax. Here reg is a register name, src and dst a general source or
destination (§8.1), and expr is an expression:
jsr reg,dst
rts reg
sys expr
ash src , reg (or, als)
ashc src , reg (or, alsc)
mul src , reg (or, mpy)
div src , reg (or, dvd)
xor reg , dst
sxt dst
mark expr
sob reg , expr
sys is another name for the trap instruction. It is used to code system calls. Its operand is required to be expressible
in 6 bits. The alternative forms for ash, ashc, mul, and div are provided to avoid conflict with EAE register names
should they be needed.
The expression in mark must be expressible in six bits, and the expression in sob must be in the same segment as
‘‘ . ’’, must not be external-undefined, must be less than ‘‘ . ’’, and must be within 510 bytes of ‘‘ . ’’.

8.9 Floating-point unit instructions

The following floating-point operations are defined, with syntax as indicated:
cfcc
setf
setd
-

Assembler Manual - 10

seti
setl
clrf fdst
negf fdst
absf fdst
tstf fsrc
movf fsrc, freg (= ldf )
movf freg, fdst (= stf )
movif src, freg (= ldcif )
movfi freg, dst (= stcfi )
movof fsrc, freg (= ldcdf )
movfo freg, fdst (= stcfd )
movie src, freg (= ldexp)
movei freg, dst (= stexp)
addf fsrc, freg
subf fsrc, freg
mulf fsrc, freg
divf fsrc, freg
cmpf fsrc, freg
modf fsrc, freg
ldfps src
stfps dst
stst dst
fsrc, fdst, and freg mean floating-point source, destination, and register respectively. Their syntax is identical to that
for their non-floating counterparts, but note that only floating registers 0—3 can be a freg.
The names of several of the operations have been changed to bring out an analogy with certain fixed-point instruc-
tions. The only strange case is movf, which turns into either stf or ldf depending respectively on whether its first
operand is or is not a register. Warning: ldf sets the floating condition codes, stf does not.

9. Other symbols
9.1 . .
The symbol ‘‘ . . ’’ is the relocation counter. Just before each assembled word is placed in the output stream, the cur-
rent value of this symbol is added to the word if the word refers to a text, data or bss segment location. If the output
word is a pc-relative address word which refers to an absolute location, the value of ‘‘ . . ’’ is subtracted.
Thus the value of ‘‘ . . ’’ can be taken to mean the starting core location of the program. In UNIX systems with reloca-
tion hardware, the initial value of ‘‘ . . ’’ is 0.
The value of ‘‘ . . ’’ may be changed by assignment. Such a course of action is sometimes necessary, but the conse-
quences should be carefully thought out. It is particularly ticklish to change ‘‘ . . ’’ midway in an assembly or to do
so in a program which will be treated by the loader, which has its own notions of ‘‘ . . ’’.

9.2 System calls

The following absolute symbols may be used to code calls to the UNIX system (see the sys instruction above).
break nice
chdir open
chmod read
chown seek
close setuid
creat signal
exec stat
exit stime
fork stty
fstat tell
getuid time
gtty umount
-

Assembler Manual - 11

link unlink
makdir wait
mdate write
mount
Warning: the wait system call is not the same as the wait instruction, which is not defined in the assembler.

10. Diagnostics
When an input file cannot be read, its name followed by a question mark is typed and assembly ceases. When syn-
tactic or semantic errors occur, a single-character diagnostic is typed out together with the line number and the file
name in which it occurred. Errors in pass 1 cause cancellation of pass 2. The possible errors are:

) parentheses error
] parentheses error
> string not terminated properly
* indirection ( * ) used illegally
. illegal assignment to ‘‘ . ’’
A error in address
B branch address is odd or too remote
E error in expression
F error in local (‘‘f ’’ or ‘‘b’’) type symbol
G garbage (unknown) character
I end of file inside an .if
M multiply defined symbol as label
O word quantity assembled at odd address
P phase error— ‘‘ . ’’ different in pass 1 and 2
R relocation error
U undefined symbol
X syntax error
 A Tutorial Introduction to the UNIX Text Editor

B. W. Kernighan

Bell Laboratories, Murray Hill, N. J.

Introduction Creating Text − the Append command ‘‘a’’

Ed is a ‘‘text editor’’, that is, an interactive pro- As our first problem, suppose we want to create
gram for creating and modifying ‘‘text’’, using direc- some text starting from scratch. Perhaps we are typing
tions provided by a user at a terminal. The text is often the very first draft of a paper; clearly it will have to start
a document like this one, or a program or perhaps data somewhere, and undergo modifications later. This sec-
for a program. tion will show how to get some text in, just to get
This introduction is meant to simplify learning started. Later we’ll talk about how to change it.
ed. The recommended way to learn ed is to read this When ed is first started, it is rather like working
document, simultaneously using ed to follow the exam- with a blank piece of paper − there is no text or infor-
ples, then to read the description in section I of the mation present. This must be supplied by the person
UNIX manual, all the while experimenting with ed. using ed; it is usually done by typing in the text, or by
(Solicitation of advice from experienced users is also reading it into ed from a file. We will start by typing in
useful.) some text, and return shortly to how to read files.
Do the exercises! They cover material not com- First a bit of terminology. In ed jargon, the text
pletely discussed in the actual text. An appendix sum- being worked on is said to be ‘‘kept in a buffer.’’ Think
marizes the commands. of the buffer as a work space, if you like, or simply as
the information that you are going to be editing. In
Disclaimer effect the buffer is like the piece of paper, on which we
This is an introduction and a tutorial. For this will write things, then change some of them, and finally
reason, no attempt is made to cover more than a part of file the whole thing away for another day.
the facilities that ed offers (although this fraction The user tells ed what to do to his text by typing
includes the most useful and frequently used parts). instructions called ‘‘commands.’’ Most commands
Also, there is not enough space to explain basic UNIX consist of a single letter, which must be typed in lower
procedures. We will assume that you know how to log case. Each command is typed on a separate line.
on to UNIX, and that you have at least a vague under- (Sometimes the command is preceded by information
standing of what a file is. about what line or lines of text are to be affected − we
You must also know what character to type as will discuss these shortly.) Ed makes no response to
the end-of-line on your particular terminal. This is a most commands − there is no prompting or typing of
‘‘newline’’ on Model 37 Teletypes, and ‘‘return’’ on messages like ‘‘ready’’. (This silence is preferred by
most others. Throughout, we will refer to this charac- experienced users, but sometimes a hangup for begin-
ter, whatever it is, as ‘‘newline’’. ners.)
The first command is append, written as the let-
Getting Started ter
We’ll assume that you have logged in to UNIX a
and it has just said ‘‘%’’. The easiest way to get ed is all by itself. It means ‘‘append (or add) text lines to the
to type buffer, as I type them in.’’ Appending is rather like
ed (followed by a newline) writing fresh material on a piece of paper.
You are now ready to go − ed is waiting for you to tell So to enter lines of text into the buffer, we just
it what to do. type an ‘‘a’’ followed by a newline, followed by the
lines of text we want, like this:
a
 –2–

Now is the time Leaving ed − the Quit command ‘‘q’’

for all good men To terminate a session with ed, save the text
to come to the aid of their party. you’re working on by writing it onto a file using the
. ‘‘w’’ command, and then type the command
The only way to stop appending is to type a line q
that contains only a period. The ‘‘.’’ is used to tell ed which stands for quit. The system will respond with
that we have finished appending. (Even experienced ‘‘%’’. At this point your buffer vanishes, with all its
users forget that terminating ‘‘.’’ sometimes. If ed text, which is why you want to write it out before quit-
seems to be ignoring you, type an extra line with just ting.
‘‘.’’ on it. You may then find you’ve added some
garbage lines to your text, which you’ll have to take out Exercise 1:
later.)
Enter ed and create some text using
After the append command has been done, the
buffer will contain the three lines a
. . . text . . .
Now is the time .
for all good men
to come to the aid of their party. Write it out using ‘‘w’’. Then leave ed with the ‘‘q’’
command, and print the file, to see that everything
The ‘‘a’’ and ‘‘.’’ aren’t there, because they are not text. worked. (To print a file, say
To add more text to what we already have, just pr filename
issue another ‘‘a’’ command, and continue typing.
or
Error Messages − ‘‘?’’ cat filename
If at any time you make an error in the com- in response to ‘‘%’’. Try both.)
mands you type to ed, it will tell you by typing
? Reading text from a file − the Edit command ‘‘e’’
This is about as cryptic as it can be, but with practice, A common way to get text into the buffer is to
you can usually figure out how you goofed. read it from a file in the file system. This is what you
do to edit text that you saved with the ‘‘w’’ command in
Writing text out as a file − the Write command ‘‘w’’ a previous session. The edit command ‘‘e’’ fetches the
entire contents of a file into the buffer. So if we had
It’s likely that we’ll want to save our text for
saved the three lines ‘‘Now is the time’’, etc., with a
later use. To write out the contents of the buffer onto a
‘‘w’’ command in an earlier session, the ed command
file, we use the write command
e junk
w
would fetch the entire contents of the file ‘‘junk’’ into
followed by the filename we want to write on. This
the buffer, and respond
will copy the buffer’s contents onto the specified file
(destroying any previous information on the file). To 68
save the text on a file named ‘‘junk’’, for example, type which is the number of characters in ‘‘junk’’. If any-
w junk thing was already in the buffer, it is deleted first.
Leave a space between ‘‘w’’ and the file name. Ed will If we use the ‘‘e’’ command to read a file into
respond by printing the number of characters it wrote the buffer, then we need not use a file name after a sub-
out. In our case, ed would respond with sequent ‘‘w’’ command; ed remembers the last file
name used in an ‘‘e’’ command, and ‘‘w’’ will write on
68 this file. Thus a common way to operate is
(Remember that blanks and the newline character at the ed
end of each line are included in the character count.) e file
Writing a file just makes a copy of the text − the [editing session]
buffer’s contents are not disturbed, so we can go on w
adding lines to it. This is an important point. Ed at all q
times works on a copy of a file, not the file itself. No
change in the contents of a file takes place until you You can find out at any time what file name ed is
give a ‘‘w’’ command. (Writing out the text onto a file remembering by typing the file command ‘‘f ’’. In our
from time to time as it is being created is a good idea, case, if we typed
since if the system crashes or if you make some horri-
f
ble mistake, you will lose all the text in the buffer but
any text that was written onto a file is relatively safe.) ed would reply
 -3-

junk for all good men

Suppose we want to print all the lines in the

Reading text from a file − the Read command ‘‘r’’ buffer. We could use ‘‘1,3p’’ as above if we knew there
Sometimes we want to read a file into the buffer were exactly 3 lines in the buffer. But in general, we
without destroying anything that is already there. This don’t know how many there are, so what do we use for
is done by the read command ‘‘r’’. The command the ending line number? Ed provides a shorthand sym-
r junk bol for ‘‘line number of last line in buffer’’ − the dollar
sign ‘‘$’’. Use it this way:
will read the file ‘‘junk’’ into the buffer; it adds it to the
end of whatever is already in the buffer. So if we do a 1,$p
read after an edit: This will print all the lines in the buffer (line 1 to last
e junk line.) If you want to stop the printing before it is fin-
r junk ished, push the DEL or Delete key; ed will type

the buffer will contain two copies of the text (six lines). ?

Now is the time and wait for the next command.

for all good men To print the last line of the buffer, we could use
to come to the aid of their party. $,$p
Now is the time
for all good men but ed lets us abbreviate this to
to come to the aid of their party. $p
Like the ‘‘w’’ and ‘‘e’’ commands, ‘‘r’’ prints the num- We can print any single line by typing the line number
ber of characters read in, after the reading operation is followed by a ‘‘p’’. Thus
complete.
1p
Generally speaking, ‘‘r’’ is much less used than
produces the response
‘‘e’’.
Now is the time
Exercise 2: which is the first line of the buffer.
Experiment with the ‘‘e’’ command − try read- In fact, ed lets us abbreviate even further: we can
ing and printing various files. You may get an error print any single line by typing just the line number − no
‘‘?’’, typically because you spelled the file name wrong. need to type the letter ‘‘p’’. So if we say
Try alternately reading and appending to see that they
$
work similarly. Verify that
ed will print the last line of the buffer for us.
ed filename
We can also use ‘‘$’’ in combinations like
is exactly equivalent to
$−1,$p
ed
e filename which prints the last two lines of the buffer. This helps
when we want to see how far we got in typing.
What does
f filename Exercise 3:
do? As before, create some text using the append
command and experiment with the ‘‘p’’ command. You
Printing the contents of the buffer − the Print com- will find, for example, that you can’t print line 0 or a
mand ‘‘p’’ line beyond the end of the buffer, and that attempts to
To print or list the contents of the buffer (or print a buffer in reverse order by saying
parts of it) on the terminal, we use the print command 3,1p
p don’t work.
The way this is done is as follows. We specify the lines
where we want printing to begin and where we want it The current line − ‘‘Dot’’ or ‘‘.’’
to end, separated by a comma, and followed by the let- Suppose our buffer still contains the six lines as
ter ‘‘p’’. Thus to print the first two lines of the buffer, above, that we have just typed
for example, (that is, lines 1 through 2) we say
1,3p
1,2p (starting line=1, ending line=2 p)
and ed has printed the three lines for us. Try typing just
Ed will respond with
p (no line numbers).
Now is the time
 –4–

This will print typing ‘‘ˆ’’ − it’s equivalent to ‘‘.−1p’’.

to come to the aid of their party.
Deleting lines: the ‘‘d’’ command
which is the third line of the buffer. In fact it is the last
(most recent) line that we have done anything with. Suppose we want to get rid of the three extra
(We just printed it!) We can repeat this ‘‘p’’ command lines in the buffer. This is done by the delete command
without line numbers, and it will continue to print line d
3. Except that ‘‘d’’ deletes lines instead of printing them,
The reason is that ed maintains a record of the its action is similar to that of ‘‘p’’. The lines to be
last line that we did anything to (in this case, line 3, deleted are specified for ‘‘d’’ exactly as they are for
which we just printed) so that it can be used instead of ‘‘p’’:
an explicit line number. This most recent line is starting line, ending line d
referred to by the shorthand symbol
Thus the command
. (pronounced ‘‘dot’’).
4,$d
Dot is a line number in the same way that ‘‘$’’ is; it
means exactly ‘‘the current line’’, or loosely, ‘‘the line deletes lines 4 through the end. There are now three
we most recently did something to.’’ We can use it in lines left, as we can check by using
several ways − one possibility is to say 1,$p
.,$p And notice that ‘‘$’’ now is line 3! Dot is set to the
This will print all the lines from (including) the current next line after the last line deleted, unless the last line
line to the end of the buffer. In our case these are lines deleted is the last line in the buffer. In that case, dot is
3 through 6. set to ‘‘$’’.
Some commands change the value of dot, while Exercise 4:
others do not. The print command sets dot to the num-
ber of the last line printed; by our last command, we Experiment with ‘‘a’’, ‘‘e’’, ‘‘r’’, ‘‘w’’, ‘‘p’’, and
would have ‘‘.’’ = ‘‘$’’ = 6. ‘‘d’’ until you are sure that you know what they do, and
until you understand how dot, ‘‘$’’, and line numbers
Dot is most useful when used in combinations are used.
like this one:
If you are adventurous, try using line numbers
.+1 (or equivalently, .+1p) with ‘‘a’’, ‘‘r’’, and ‘‘w’’ as well. You will find that
This means ‘‘print the next line’’ and gives us a handy ‘‘a’’ will append lines after the line number that you
way to step slowly through a buffer. We can also say specify (rather than after dot); that ‘‘r’’ reads a file in
.−1 (or .−1p ) after the line number you specify (not necessarily at the
end of the buffer); and that ‘‘w’’ will write out exactly
which means ‘‘print the line before the current line.’’ the lines you specify, not necessarily the whole buffer.
This enables us to go backwards if we wish. Another These variations are sometimes handy. For instance
useful one is something like you can insert a file at the beginning of a buffer by say-
.−3,.−1p ing
which prints the previous three lines. 0r filename
Don’t forget that all of these change the value of and you can enter lines at the beginning of the buffer by
dot. You can find out what dot is at any time by typing saying
.= 0a
. . . text . . .
Ed will respond by printing the value of dot.
.
Let’s summarize some things about the ‘‘p’’
command and dot. Essentially ‘‘p’’ can be preceded by Notice that ‘‘.w’’ is very different from
0, 1, or 2 line numbers. If there is no line number .
given, it prints the ‘‘current line’’, the line that dot w
refers to. If there is one line number given (with or
without the letter ‘‘p’’), it prints that line (and dot is set Modifying text: the Substitute command ‘‘s’’
there); and if there are two line numbers, it prints all the
lines in that range (and sets dot to the last line printed.) We are now ready to try one of the most impor-
If two line numbers are specified the first can’t be big- tant of all commands − the substitute command
ger than the second (see Exercise 2.) s
Typing a single newline will cause printing of This is the command that is used to change individual
the next line − it’s equivalent to ‘‘.+1p’’. Try it. Try words or letters within a line or group of lines. It is
 -5-

what we use, for example, for correcting spelling mis- Nowxx is the time
takes and typing errors. we can say
Suppose that by a typing error, line 1 says s/xx//p
Now is th time to get
− the ‘‘e’’ has been left off ‘‘the’’. We can use ‘‘s’’ to Now is the time
fix this up as follows:
Notice that ‘‘//’’ here means ‘‘no characters’’, not a
1s/th/the/ blank. There is a difference! (See below for another
This says: ‘‘in line 1, substitute for the characters ‘th’ meaning of ‘‘//’’.)
the characters ‘the’.’’ To verify that it works (ed will
not print the result automatically) we say Exercise 5:
p Experiment with the substitute command. See
and get what happens if you substitute for some word on a line
with several occurrences of that word. For example, do
Now is the time this:
which is what we wanted. Notice that dot must have a
been set to the line where the substitution took place, the other side of the coin
since the ‘‘p’’ command printed that line. Dot is .
always set this way with the ‘‘s’’ command. s/the/on the/p
The general way to use the substitute command You will get
is
on the other side of the coin
starting–line, ending–line s/change this/to this/
A substitute command changes only the first occur-
Whatever string of characters is between the first pair rence of the first string. You can change all occur-
of slashes is replaced by whatever is between the sec- rences by adding a ‘‘g’’ (for ‘‘global’’) to the ‘‘s’’ com-
ond pair, in all the lines between starting line and end- mand, like this:
ing line. Only the first occurrence on each line is
changed, however. If you want to change every occur- s/ . . . / . . . /gp
rence, see Exercise 5. The rules for line numbers are Try other characters instead of slashes to delimit the
the same as those for ‘‘p’’, except that dot is set to the two sets of characters in the ‘‘s’’ command − anything
last line changed. (But there is a trap for the unwary: if should work except blanks or tabs.
no substitution took place, dot is not changed. This (If you get funny results using any of the charac-
causes an error ‘‘?’’ as a warning.) ters
Thus we can say ˆ . $ [ * \
1,$s/speling/spelling/ read the section on ‘‘Special Characters’’.)
and correct the first spelling mistake on each line in the
text. (This is useful for people who are consistent mis- Context searching − ‘‘/ . . . /’’
spellers!) With the substitute command mastered, we can
If no line numbers are given, the ‘‘s’’ command move on to another highly important idea of ed − con-
assumes we mean ‘‘make the substitution on line dot’’, text searching.
so it changes things only on the current line. This leads Suppose we have our original three line text in
to the very common sequence the buffer:
s/something/something else/p Now is the time
which makes some correction on the current line, and for all good men
then prints it, to make sure it worked out right. If it to come to the aid of their party.
didn’t, we can try again. (Notice that we put a print Suppose we want to find the line that contains ‘‘their’’
command on the same line as the substitute. With few so we can change it to ‘‘the’’. Now with only three
exceptions, ‘‘p’’ can follow any command; no other lines in the buffer, it’s pretty easy to keep track of what
multi-command lines are legal.) line the word ‘‘their’’ is on. But if the buffer contained
It’s also legal to say several hundred lines, and we’d been making changes,
s/ . . . // deleting and rearranging lines, and so on, we would no
longer really know what this line number would be.
which means ‘‘change the first string of characters to Context searching is simply a method of specifying the
nothing’’, i.e., remove them. This is useful for deleting desired line, regardless of what its number is, by speci-
extra words in a line or removing extra letters from fying some context on it.
words. For instance, if we had
 -6-

The way we say ‘‘search for a line that contains or

this particular string of characters’’ is to type /party/−1s/good/bad/
/string of characters we want to find/ The choice is dictated only by convenience. We could
For example, the ed line print all three lines by, for instance
/their/ /Now/,/party/p
is a context search which is sufficient to find the desired or
line − it will locate the next occurrence of the charac- /Now/,/Now/+2p
ters between slashes (‘‘their’’). It also sets dot to that
line and prints the line for verification: or by any number of similar combinations. The first
one of these might be better if we don’t know how
to come to the aid of their party. many lines are involved. (Of course, if there were only
‘‘Next occurrence’’ means that ed starts looking for the three lines in the buffer, we’d use
string at line ‘‘.+1’’, searches to the end of the buffer, 1,$p
then continues at line 1 and searches to line dot. (That
is, the search ‘‘wraps around’’ from ‘‘$’’ to 1.) It scans but not if there were several hundred.)
all the lines in the buffer until it either finds the desired The basic rule is: a context search expression is
line or gets back to dot again. If the given string of the same as a line number, so it can be used wherever a
characters can’t be found in any line, ed types the error line number is needed.
message
? Exercise 6:

Otherwise it prints the line it found. Experiment with context searching. Try a body
of text with several occurrences of the same string of
We can do both the search for the desired line characters, and scan through it using the same context
and a substitution all at once, like this: search.
/their/s/their/the/p Try using context searches as line numbers for
which will yield the substitute, print and delete commands. (They can
also be used with ‘‘r’’, ‘‘w’’, and ‘‘a’’.)
to come to the aid of the party.
Try context searching using ‘‘?text?’’ instead of
There were three parts to that last command: context
‘‘/text/’’. This scans lines in the buffer in reverse order
search for the desired line, make the substitution, print
rather than normal. This is sometimes useful if you go
the line.
too far while looking for some string of characters − it’s
The expression ‘‘/their/’’ is a context search an easy way to back up.
expression. In their simplest form, all context search
(If you get funny results with any of the charac-
expressions are like this − a string of characters sur-
ters
rounded by slashes. Context searches are interchange-
able with line numbers, so they can be used by them- ˆ . $ [ * \
selves to find and print a desired line, or as line num- read the section on ‘‘Special Characters’’.)
bers for some other command, like ‘‘s’’. We used them
Ed provides a shorthand for repeating a context
both ways in the examples above.
search for the same string. For example, the ed line
Suppose the buffer contains the three familiar number
lines
/string/
Now is the time
for all good men will find the next occurrence of ‘‘string’’. It often hap-
to come to the aid of their party. pens that this is not the desired line, so the search must
be repeated. This can be done by typing merely
Then the ed line numbers
//
/Now/+1
/good/ This shorthand stands for ‘‘the most recently used con-
/party/−1 text search expression.’’ It can also be used as the first
string of the substitute command, as in
are all context search expressions, and they all refer to
the same line (line 2). To make a change in line 2, we /string1/s//string2/
could say which will find the next occurrence of ‘‘string1’’ and
/Now/+1s/good/bad/ replace it by ‘‘string2’’. This can save a lot of typing.
Similarly
or
??
/good/s/good/bad/
means ‘‘scan backwards for the same expression.’’
 -7-

Change and Insert − ‘‘c’’ and ‘‘i’’ Experiment with ‘‘a’’ and ‘‘i’’, to see that they
This section discusses the change command are similar, but not the same. You will observe that

c line–number a
. . . text . . .
which is used to change or replace a group of one or .
more lines, and the insert command
appends after the given line, while
i
line–number i
which is used for inserting a group of one or more . . . text . . .
lines. .
‘‘Change’’, written as inserts before it. Observe that if no line number is
c given, ‘‘i’’ inserts before line dot, while ‘‘a’’ appends
is used to replace a number of lines with different lines, after line dot.
which are typed in at the terminal. For example, to
change lines ‘‘.+1’’ through ‘‘$’’ to something else, Moving text around: the ‘‘m’’ command
type The move command ‘‘m’’ is used for cutting and
pasting − it lets you move a group of lines from one
.+1,$c
place to another in the buffer. Suppose we want to put
. . . type the lines of text you want here . . .
the first three lines of the buffer at the end instead. We
.
could do it by saying:
The lines you type between the ‘‘c’’ command and the
1,3w temp
‘‘.’’ will take the place of the original lines between
$r temp
start line and end line. This is most useful in replacing
1,3d
a line or several lines which have errors in them.
If only one line is specified in the ‘‘c’’ com- (Do you see why?) but we can do it a lot easier with
mand, then just that line is replaced. (You can type in the ‘‘m’’ command:
as many replacement lines as you like.) Notice the use 1,3m$
of ‘‘.’’ to end the input − this works just like the ‘‘.’’ in The general case is
the append command and must appear by itself on a
new line. If no line number is given, line dot is start line, end line m after this line
replaced. The value of dot is set to the last line you Notice that there is a third line to be specified − the
typed in. place where the moved stuff gets put. Of course the
‘‘Insert’’ is similar to append − for instance lines to be moved can be specified by context searches;
if we had
/string/i
. . . type the lines to be inserted here . . . First paragraph
. ...
end of first paragraph.
will insert the given text before the next line that con- Second paragraph
tains ‘‘string’’. The text between ‘‘i’’ and ‘‘.’’ is ...
inserted before the specified line. If no line number is end of second paragraph.
specified dot is used. Dot is set to the last line inserted.
we could reverse the two paragraphs like this:
Exercise 7: /Second/,/second/m/First/−1
‘‘Change’’ is rather like a combination of delete Notice the ‘‘−1’’ − the moved text goes after the line
followed by insert. Experiment to verify that mentioned. Dot gets set to the last line moved.
start, end d
i The global commands ‘‘g’’ and ‘‘v’’
. . . text . . . The global command ‘‘g’’ is used to execute one
. or more ed commands on all those lines in the buffer
is almost the same as that match some specified string. For example
start, end c g/peling/p
. . . text . . . prints all lines that contain ‘‘peling’’. More usefully,
.
g/peling/s//pelling/gp
These are not precisely the same if line ‘‘$’’ gets
makes the substitution everywhere on the line, then
deleted. Check this out. What is dot?
prints each corrected line. Compare this to
1,$s/peling/pelling/gp
 –8–

which only prints the last line substituted. Another will only find an occurrence of ‘‘string’’ that is at the
subtle difference is that the ‘‘g’’ command does not end of some line. This implies, of course, that
give a ‘‘?’’ if ‘‘peling’’ is not found where the ‘‘s’’ com- /ˆstring$/
mand will.
will find only a line that contains just ‘‘string’’, and
There may be several commands (including ‘‘a’’,
‘‘c’’ ‘‘i’’ ‘‘r’’, ‘‘w’’, but not ‘‘g’’); in that case, every /ˆ.$/
line except the last must end with a backslash ‘‘\’’: finds a line containing exactly one character.
g/xxx/.–1s/abc/def/\ The character ‘‘.’’, as we mentioned above,
.+2s/ghi/jkl/\ matches anything;
.–2,.p
/x.y/
makes changes in the lines before and after each line
matches any of
that contains ‘‘xxx’’, then prints all three lines.
x+y
The ‘‘v’’ command is the same as ‘‘g’’, except
x–y
that the commands are executed on every line that does
xy
not match the string following ‘‘v’’:
x.y
v/ /d
This is useful in conjunction with ‘‘*’’, which is a repe-
deletes every line that does not contain a blank. tition character; ‘‘a*’’ is a shorthand for ‘‘any number
of a’s,’’ so ‘‘.*’’ matches any number of anythings.
Special Characters This is used like this:
You may have noticed that things just don’t s/.*/stuff/
work right when you used some characters like ‘‘.’’,
which changes an entire line, or
‘‘*’’, ‘‘$’’, and others in context searches and the sub-
stitute command. The reason is rather complex, s/.*,//
although the cure is simple. Basically, ed treats these which deletes all characters in the line up to and includ-
characters as special, with special meanings. For ing the last comma. (Since ‘‘.*’’ finds the longest pos-
instance, in a context search or the first string of the sible match, this goes up to the last comma.)
substitute command only,
‘‘[’’ is used with ‘‘]’’ to form ‘‘character
/x.y/ classes’’; for example,
means ‘‘a line with an x, any character, and a y,’’ not /[1234567890]/
just ‘‘a line with an x, a period, and a y.’’ A complete
matches any single digit _ any one of the characters
list of the special characters that can cause trouble is
inside the braces will cause a match.
the following:
Finally, the ‘‘&’’ is another shorthand character -
ˆ . $ [ * \
it is used only on the right-hand part of a substitute
Warning: The backslash character \ is special to ed. command where it means ‘‘whatever was matched on
For safety’s sake, avoid it where possible. If you have the left-hand side’’. It is used to save typing. Suppose
to use one of the special characters in a substitute com- the current line contained
mand, you can turn off its magic meaning temporarily
Now is the time
by preceding it with the backslash. Thus
and we wanted to put parentheses around it. We could
s/\\\.\*/backslash dot star/
just retype the line, but this is tedious. Or we could say
will change ‘‘\.*’’ into ‘‘backslash dot star’’.
s/ˆ/(/
Here is a hurried synopsis of the other special s/$/)/
characters. First, the circumflex ‘‘ ˆ ’’ signifies the
using our knowledge of ‘‘ˆ’’ and ‘‘$’’. But the easiest
beginning of a line. Thus
way uses the ‘‘&’’:
/ˆstring/
s/.*/(&)/
finds ‘‘string’’ only if it is at the beginning of a line: it
This says ‘‘match the whole line, and replace it by itself
will find
surrounded by parens.’’ The ‘‘&’’ can be used several
string times in a line; consider using
but not s/.*/&? &!!/
the string... to produce
The dollar-sign ‘‘$’’ is just the opposite of the circum- Now is the time? Now is the time!!
flex; it means the end of a line:
/string$/
 -9-

We don’t have to match the whole line, of m (move) Move lines specified to after the line named
course: if the buffer contains after m. Dot is set to the last line moved.
the end of the world p (print) Print specified lines. If none specified, print
we could type line dot. A single line number is equivalent to ‘‘line-
number p’’. A single newline prints ‘‘.+1’’, the next
/world/s//& is at hand/ line.
to produce q (quit) Exit from ed. Wipes out all text in buffer!!
the end of the world is at hand r (read) Read a file into buffer (at end unless specified
Observe this expression carefully, for it illustrates how elsewhere.) Dot set to last line read.
to take advantage of ed to save typing. The string s (substitute) s/string1/string2/ will substitute the char-
‘‘/world/’’ found the desired line; the shorthand ‘‘//’’ acters of ‘string2’ for ‘string1’ in specified lines. If no
found the same word in the line; and the ‘‘&’’ saved us line is specified, make substitution in line dot. Dot is
from typing it again. set to last line in which a substitution took place, which
The ‘‘&’’ is a special character only within the means that if no substitution took place, dot is not
replacement text of a substitute command, and has no changed. s changes only the first occurrence of string1
special meaning elsewhere. We can turn off the special on a line; to change all of them, type a ‘‘g’’ after the
meaning of ‘‘&’’ by preceding it with a ‘‘\’’: final slash.
s/ampersand/\&/ v (exclude) v/---/commands executes ‘‘commands’’ on
those lines that do not contain ‘‘---’’.
will convert the word ‘‘ampersand’’ into the literal sym-
bol ‘‘&’’ in the current line. w (write) Write out buffer onto a file. Dot is not
changed.
.= (dot value) Print value of dot. (‘‘=’’ by itself prints
the value of ‘‘$’’.)
Summary of Commands and Line Numbers
! (temporary escape)
The general form of ed commands is the com- Execute this line as a UNIX command.
mand name, perhaps preceded by one or two line num-
bers, and, in the case of e, r and w, followed by a file /-----/ Context search. Search for next line which con-
name. Only one command is allowed per line, but a p tains this string of characters. Print it. Dot is set to line
command may follow any other command (except for where string found. Search starts at ‘‘.+1’’, wraps
e, r, w and q). around from ‘‘$’’ to 1, and continues to dot, if neces-
sary.
a (append) Add lines to the buffer (at line dot, unless a ?-----? Context search in reverse direction. Start
different line is specified). Appending continues until search at ‘‘.−1’’, scan to 1, wrap around to ‘‘$’’.
‘‘.’’ is typed on a new line. Dot is set to the last line
appended.
c (change) Change the specified lines to the new text
which follows. The new lines are terminated by a ‘‘.’’.
If no lines are specified, replace line dot. Dot is set to
last line changed.
d (delete) Delete the lines specified. If none are speci-
fied, delete line dot. Dot is set to the first undeleted
line, unless ‘‘$’’ is deleted, in which case dot is set to
‘‘$’’.
e (edit) Edit new file. Any previous contents of the
buffer are thrown away, so issue a w beforehand if you
want to save them.
f (file) Print remembered filename. If a name follows f
the remembered name will be set to it.
g (global) g/---/commands will execute the commands
on those lines that contain ‘‘---’’, which can be any con-
text search expression.
i (insert) Insert lines before specified line (or dot) until
a ‘‘.’’ is typed on a new line. Dot is set to last line
inserted.
 UNIX for Beginners

Brian W. Kernighan
Bell Laboratories, Murray Hill, N. J.

In many ways, UNIX is the state of the art Logging In

in computer operating systems. From the user’s Most of the details about logging in are in
point of view, it is easy to learn and use, and pre- the manual section called ‘‘How to Get Started’’
sents few of the usual impediments to getting the (pages iv-v in the 5th Edition). Here are a cou-
job done. ple of extra warnings.
It is hard, however, for the beginner to You must have a UNIX login name, which
know where to start, and how to make the best you can get from whoever administers your sys-
use of the facilities available. The purpose of tem. You also need to know the phone number.
this introduction is to point out high spots for UNIX is capable of dealing with a variety of ter-
new users, so they can get used to the main ideas minals: Terminet 300’s; Execuport, TI and simi-
of UNIX and start making good use of it quickly. lar portables; video terminals; GSI’s; and even
This paper is not an attempt to re-write the venerable Teletype in its various forms. But
the UNIX Programmer’s Manual; often the dis- note: UNIX will not handle IBM 2741 terminals
cussion of something is simply ‘‘read section x and their derivatives (e.g., some Anderson-
in the manual.’’ (This implies that you will need Jacobsons, Novar). Furthermore, UNIX is
a copy of the UNIX Programmer’s Manual.) strongly oriented towards devices with lower
Rather it suggests in what order to read the man- case. If your terminal produces only upper case
ual, and it collects together things that are stated (e.g., model 33 Teletype), life will be so difficult
only indirectly in the manual. that you should look for another terminal.
There are five sections: Be sure to set the switches appropriately
1. Getting Started: How to log in to a UNIX, on your device: speed (if it’s variable) to 30
how to type, what to do about mistakes in characters per second, lower case, full duplex,
typing, how to log out. Some of this is even parity, and any others that local wisdom
dependent on which UNIX you log into advises. Establish a connection using whatever
(phone numbers, for example) and what magic is needed for your terminal. UNIX should
terminal you use, so this section must type ‘‘login:’’ at you. If it types garbage, you
necessarily be supplemented by local may be at the wrong speed; push the ‘break’ or
information. ‘interrupt’ key once. If that fails to produce a
login message, consult a guru.
2. Day-to-day Use: Things you need every
day to use UNIX effectively: generally When you get a ‘‘login:’’ message, type
useful commands; the file system. your login name in lower case. Follow it by a
RETURN if the terminal has one. If a password is
3. Document Preparation: Preparing required, you will be asked for it, and (if possi-
manuscripts is one of the most common ble) printing will be turned off while you type it,
uses for UNIX. This section contains again followed by a RETURN. (On M37 Tele-
advice, but not extensive instructions on types always use NEWLINE or LINEFEED in place
any of the formatting programs. of RETURN).
4. Writing Programs: UNIX is an excellent The culmination of your login efforts is a
vehicle for developing programs. This percent sign ‘‘%’’. The percent sign means that
section talks about some of the tools, but UNIX is ready to accept commands from the ter-
again is not a tutorial in any of the pro- minal. (You may also get a message of the day
gramming languages that UNIX provides. just before the percent sign or a notification that
5. A UNIX Reading List. An annotated bibli- you have mail.)
ography of documents worth reading by
new users. Typing Commands
Once you’ve seen the percent sign, you
can type commands, which are requests that
I. GETTING STARTED UNIX do something. Try typing

date
 –2–

followed by RETURN. You should get back and start over (on the same line!).
something like What if you must enter a sharp or at-sign
Sun Sep 22 10:52:29 EDT 1974 as part of the text? If you precede either ‘‘#’’ or
‘‘@’’ by a backslash ‘‘\’’, it loses its erase mean-
Don’t forget the RETURN after the command, or ing. This implies that to erase a backslash, you
nothing will happen. If you think you’re being have to type two sharps or two at-signs. The
ignored, type a RETURN; something should hap- backslash is used extensively in UNIX to indicate
pen. We won’t show the carriage returns, but that the following character is in some way spe-
they have to be there. cial.
Another command you might try is who,
which tells you everyone who is currently Readahead
logged in: UNIX has full readahead, which means

who that you can type as fast as you want, whenever

you want, even when some command is typing
gives something like at you. If you type during output, your input
pjp ttyf Sep 22 09:40 characters will appear intermixed with the out-
bwk ttyg Sep 22 09:48 put characters, but they will be stored away by
UNIX and interpreted in the correct order. So you
mel ttyh Sep 22 09:58
can type two commands one after another with-
The time is when the user logged in. out waiting for the first to finish or even begin.
If you make a mistake typing the com-
mand name, UNIX will tell you. For example, if Stopping a Program
you type You can stop most programs by typing
the character ‘‘DEL’’ (perhaps called ‘‘delete’’
whom
or ‘‘rubout’’ on your terminal). There are excep-
you will be told tions, like the text editor, where DEL stops what-
ever the program is doing but leaves you in that
whom: not found
program. You can also just hang up the phone.
The ‘‘interrupt’’ or ‘‘break’’ key found on most
Strange Terminal Behavior terminals has no effect.
Sometimes you can get into a state where
Logging Out
your terminal acts strangely. For example, each
letter may be typed twice, or the RETURN may The easiest way to log out is to hang up
not cause a line feed. You can often fix this by the phone. You can also type
logging out and logging back in. Or you can
login name–of–new–user
read the description of the command stty in sec-
tion I of the manual. This will also tell you how and let someone else use the terminal you were
to get intelligent treatment of tab characters on. It is not sufficient just to turn off the termi-
(which are much used in UNIX) if your terminal nal. UNIX has no time-out mechanism, so you’ll
doesn’t have tabs. If it does have computer-set- be there forever unless you hang up.
table tabs, the command tabs will set the stops
correctly for you. Mail
When you log in, you may sometimes get
Mistakes in Typing the message
If you make a typing mistake, and see it
You have mail.
before the carriage return has been typed, there
are two ways to recover. The sharp-character UNIX provides a postal system so you can send
‘‘#’’ erases the last character typed; in fact suc- and receive letters from other users of the sys-
cessive uses of ‘‘#’’ erase characters back to the tem. To read your mail, issue the command
beginning of the line (but not beyond). So if you
type badly, you can correct as you go: mail

dd#atte##e Your mail will be printed, and then you will be

asked
is the same as ‘‘date’’.
Save?
The at-sign ‘‘@’’ erases all of the charac-
ters typed so far on the current input line, so if If you do want to save the mail, type y, for
the line is irretrievably fouled up, type an ‘‘@’’ ‘‘yes’’; any other response means ‘‘no’’.
 -3-

How do you send mail to someone else? (o+o), for ‘‘over and out’’.
Suppose it is to go to ‘‘joe’’ (assuming ‘‘joe’’ is To terminate the conversation, each side
someone’s login name). The easiest way is this: must type a ‘‘control–d’’ character alone
on a line. (‘‘Delete’’ also works.) When
mail joe
the other person types his ‘‘control–d’’,
now type in the text of the letter
you will get the message ‘‘EOT’’ on your
on as many lines as you like ...
terminal.
after the last line of the letter
type the character ‘‘control–d’’, If you write to someone who isn’t logged
that is, hold down ‘‘control’’ and type in, or who doesn’t want to be disturbed, you’ll
a letter ‘‘d’’. be told. If the target is logged in but doesn’t
And that’s it. The ‘‘control-d’’ sequence, usu- answer after a decent interval, simply type ‘‘con-
ally called ‘‘EOT’’, is used throughout UNIX to trol-d’’.
mark the end of input from a terminal, so you
might as well get used to it. On-line Manual
There are other ways to send mail — you The UNIX Programmer’s Manual is typi-
can send a previously prepared letter, and you cally kept on-line. If you get stuck on some-
can mail to a number of people all at once. For thing, and can’t find an expert to assist you, you
more details see mail (I). can print on your terminal some manual section
that might help. It’s also useful for getting the
The notation mail (I) means the com-
most up-to-date information on a command. To
mand mail in section (I) of the UNIX Program-
print a manual section, type ‘‘man section-
mer’s Manual.
name’’. Thus to read up on the who command,
type
Writing to other users
At some point in your UNIX career, out of man who
the blue will come a message like If the section in question isn’t in part I of the
Message from joe... manual, you have to give the section number as
well, as in
accompanied by a startling beep. It means that
Joe wants to talk to you, but unless you take man 6 chess
explicit action you won’t be able to talk back. Of course you’re out of luck if you can’t remem-
To respond, type the command ber the section name.
write joe
II. DAY-TO-DAY USE
This establishes a two-way communication path.
Now whatever Joe types on his terminal will Creating Files — The Editor
appear on yours and vice versa. The path is
If we have to type a paper or a letter or a
slow, rather like talking to the moon. (If you are
program, how do we get the information stored
in the middle of something, you have to get to a
in the machine? Most of these tasks are done
state where you can type a command. Normally,
with the UNIX ‘‘text editor’’ ed. Since ed is thor-
whatever program you are running has to termi-
oughly documented in ed (I) and explained in A
nate or be terminated. If you’re editing, you can
Tutorial Introduction to the UNIX Text Editor,
escape temporarily from the editor — read the
we won’t spend any time here describing how to
manual.)
use it. All we want it for right now is to make
A protocol is needed to keep what you some files. (A file is just a collection of infor-
type from getting garbled up with what Joe mation stored in the machine, a simplistic but
types. Typically it’s like this: adequate definition.)
Joe types ‘‘write smith’’ and waits. To create a file with some text in it, do the
Smith types ‘‘write joe’’ and waits. following:
Joe now types his message (as many lines
ed (invokes the text editor)
as he likes). When he’s ready for a reply,
a (command to ‘‘ed’’, to add text)
he signals it by typing (o), which stands
now type in
for ‘‘over’’.
whatever text you want ...
Now Smith types a reply, also terminated
. (signals the end of adding text)
by (o).
This cycle repeats until someone gets At this point we could do various editing opera-
tired; he then signals his intent to quit with tions on the text we typed in, such as correcting
 -4-

spelling mistakes, rearranging paragraphs and Printing Files

the like. Finally, we write the information we Now that you’ve got a file of text, how do
have typed into a file with the editor command you print it so people can look at it? There are a
‘‘w’’: host of programs that do that, probably more
w junk than are needed.
One simple thing is to use the editor,
ed will respond with the number of characters it
since printing is often done just before making
wrote into the file called ‘‘junk’’.
changes anyway. You can say
Suppose we now add a few more lines
with ‘‘a’’, terminate them with ‘‘.’’, and write the ed junk
whole thing out as ‘‘temp’’, using 1,$p

w temp ed will reply with the count of the characters in

‘‘junk’’ and then print all the lines in the file.
We should now have two files, a smaller one After you learn how to use the editor, you can be
called ‘‘junk’’ and a bigger one (bigger by the selective about the parts you print.
extra lines) called ‘‘temp’’. Type a ‘‘q’’ to quit
There are times when it’s not feasible to
the editor.
use the editor for printing. For example, there is
a limit on how big a file ed can handle (about
What files are out there?
65,000 characters or 4000 lines). Secondly, it
The ls (for ‘‘list’’) command lists the will only print one file at a time, and sometimes
names (not contents) of any of the files that UNIX you want to print several, one after another. So
knows about. If we type here are a couple of alternatives.
ls First is cat, the simplest of all the printing
programs. cat simply copies all the files in a list
the response will be
onto the terminal. So you can say
junk
cat junk
temp
or, to print two files,
which are indeed our two files. They are sorted
into alphabetical order automatically, but other cat junk temp
variations are possible. For example, if we add
The two files are simply concatenated (hence the
the optional argument ‘‘-t’’,
name ‘‘cat’’) onto the terminal.
ls –t pr produces formatted printouts of files.
lists them in the order in which they were last As with cat, pr prints all the files in a list. The
changed, most recent first. The ‘‘-l’’ option difference is that it produces headings with date,
gives a ‘‘long’’ listing: time, page number and file name at the top of
each page, and extra lines to skip over the fold in
ls –l the paper. Thus,
will produce something like pr junk temp
–rw–rw–rw– 1 bwk 41 Sep 22 12:56 junk will list ‘‘junk’’ neatly, then skip to the top of a
–rw–rw–rw– 1 bwk 78 Sep 22 12:57 temp new page and list ‘‘temp’’ neatly.
The date and time are of the last change to the pr will also produce multi-column out-
file. The 41 and 78 are the number of characters put:
(you got the same thing from ed). ‘‘bwk’’ is the
pr –3 junk
owner of the file — the person who created it.
The ‘‘-rw-rw-rw-’’ tells who has permission to prints ‘‘junk’’ in 3-column format. You can use
read and write the file, in this case everyone. any reasonable number in place of ‘‘3’’ and pr
Options can be combined: ‘‘ls -lt’’ would will do its best.
give the same thing, but sorted into time order. It should be noted that pr is not a format-
You can also name the files you’re interested in, ting program in the sense of shuffling lines
and ls will list the information about them only. around and justifying margins. The true format-
More details can be found in ls (I). ters are roff, nroff, and troff, which we will get
It is generally true of UNIX programs that to in the section on document preparation.
‘‘flag’’ arguments like ‘‘-t’’ precede filename There are also programs that print files on
arguments. a high-speed printer. Look in your manual
 -5-

under opr and lpr. Which to use depends on the On to some more positive suggestions.
hardware configuration of your machine. Suppose you’re typing a large document like a
book. Logically this divides into many small
Shuffling Files About pieces, like chapters and perhaps sections. Phys-
Now that you have some files in the file ically it must be divided too, for ed will not han-
system and some experience in printing them, dle big files. Thus you should type the docu-
you can try bigger things. For example, you can ment as a number of files. You might have a
move a file from one place to another (which separate file for each chapter, called
amounts to giving a file a new name), like this: chap1
mv junk precious chap2
etc...
This means that what used to be ‘‘junk’’ is now
‘‘precious’’. If you do an ls command now, you Or, if each chapter were broken into several
will get files, you might have

precious chap1.1
temp chap1.2
chap1.3
Beware that if you move a file to another one ...
that already exists, the already existing contents chap2.1
are lost forever. chap2.2
If you want to make a copy of a file (that ...
is, to have two versions of something), you can You can now tell at a glance where a particular
use the cp command: file fits into the whole.
cp precious temp1 There are advantages to a systematic
naming convention which are not obvious to the
makes a duplicate copy of ‘‘precious’’ in
novice UNIX user. What if you wanted to print
‘‘temp1’’.
the whole book? You could say
Finally, when you get tired of creating
and moving files, there is a command to remove pr chap1.1 chap1.2 chap1.3 ......
files from the file system, called rm. but you would get tired pretty fast, and would
rm temp temp1 probably even make mistakes. Fortunately, there
is a shortcut. You can say
will remove all of the files named. You will get
a warning message if one of the named files pr chap*
wasn’t there. The ‘‘*’’ means ‘‘anything at all’’, so this trans-
lates into ‘‘print all files whose names begin
Filename, What’s in a with ‘chap’ ’’, listed in alphabetical order. This
So far we have used filenames without shorthand notation is not a property of the pr
ever saying what’s a legal name, so it’s time for command, by the way. It is system-wide, a ser-
a couple of rules. First, filenames are limited to vice of the program that interprets commands
14 characters, which is enough to be descriptive. (the ‘‘shell’’ sh (I)). Using that fact, you can see
Second, although you can use almost any char- how to list the files of the book:
acter in a filename, common sense says you
ls chap*
should stick to ones that are visible, and that you
should probably avoid characters that might be produces
used with other meanings. We already saw, for
example, that in the ls command, ‘‘ls -t’’ meant chap1.1
to list in time order. So if you had a file whose chap1.2
name was ‘‘-t’’, you would have a tough time chap1.3
listing it by name. There are a number of other ...
characters which have special meaning either to The ‘‘*’’ is not limited to the last position in a
UNIX as a whole or to numerous commands. To filename — it can be anywhere. Thus
avoid pitfalls, you would probably do well to use
only letters, numbers and the period. (Don’t use rm *junk*
the period as the first character of a filename, for removes all files that contain ‘‘junk’’ as any part
reasons too complicated to go into.) of their name. As a special case, ‘‘*’’ by itself
matches every filename, so
 -6-

pr * ‘‘/’’ is the name of the root of the tree (a conven-

tion used by UNIX). You will get a response
prints all the files (alphabetical order), and
something like this:
rm *
bin
removes all files. (You had better be sure that’s dev
what you wanted to say!) etc
lib
The ‘‘*’’ is not the only pattern-matching
tmp
feature available. Suppose you want to print
usr
only chapters 1 through 4 and 9 of the book.
Then you can say This is a collection of the basic directories of
files that UNIX knows about. On most systems,
pr chap[12349]*
‘‘usr’’ is a directory that contains all the normal
The ‘‘[...]’’ means to match any of the charac- users of the system, like you. Now try
ters inside the brackets. You can also do this
ls /usr
with
This should list a long series of names, among
pr chap[1–49]*
which is your own login name. Finally, try
‘‘[a-z]’’ matches any character in the range a
ls /usr/your–name
through z. There is also a ‘‘?’’ character, which
matches any single character, so You should get what you get from a plain
pr ? ls
will print all files which have single-character Now try
names.
cat /usr/your–name/junk
Of these niceties, ‘‘*’’ is probably the
most useful, and you should get used to it. The (if ‘‘junk’’ is still around). The name
others are frills, but worth knowing.
/usr/your–name/junk
If you should ever have to turn off the
special meaning of ‘‘*’’, ‘‘?’’, etc., enclose the is called the ‘‘pathname’’ of the file that you nor-
entire argument in quotes (single or double), as mally think of as ‘‘junk’’. ‘‘Pathname’’ has an
in obvious meaning: it represents the full name of
the path you have to follow through the tree of
ls "?" directories to get to a particular file. It is a uni-
versal rule in UNIX that anywhere you can use an
ordinary filename, you can use a pathname.
What’s in a Filename, Continued
Here is a picture which may make this
When you first made that file called
clearer:
‘‘junk’’, how did UNIX know that there wasn’t
another ‘‘junk’’ somewhere else, especially since (root)
the person in the next office is also reading this /|\
/ | \
tutorial? The reason is that generally each user / | \
of UNIX has his own ‘‘directory’’, which contains bin etc usr dev tmp
only the files that belong to him. When you cre- /|\ /|\ /|\ /|\ /|\
ate a new file, unless you take special action, the / | \
new file is made in your own directory, and is / | \
adam eve mary
unrelated to any other file of the same name that / / \ \
might exist in someone else’s directory. / \ junk
The set of all files that UNIX knows about junk temp
are organized into a (usually big) tree, with your
Notice that Mary’s ‘‘junk’’ is unrelated to
files located several branches up into the tree. It
Eve’s.
is possible for you to ‘‘walk’’ around this tree,
and to find any file in the system, by starting at This isn’t too exciting if all the files of
the root of the tree and walking along the right interest are in your own directory, but if you
set of branches. work with someone else or on several projects
concurrently, it becomes handy indeed. For
To begin, type
example, your friends can print your book by
ls / saying
 -7-

pr /usr/your–name/chap* mkdir book

Similarly, you can find out what files your then go to it with
neighbor has by saying
chdir book
ls /usr/neighbor–name
then start typing chapters. The book is now
or make your own copy of one of his files by found in (presumably)
cp /usr/your–neighbor/his–file yourfile /usr/your–name/book

(If your neighbor doesn’t want you pok- To delete a directory, see rmdir (I).
ing around in his files, or vice versa, privacy can You can go up one level in the tree of
be arranged. Each file and directory can have files by saying
read-write-execute permissions for the owner, a
chdir ..
group, and everyone else, to control access. See
ls (I) and chmod (I) for details. As a matter of ‘‘..’’ is the name of the parent of whatever direc-
observed fact, most users most of the time find tory you are currently in. For completeness, ‘‘.’’
openness of more benefit than privacy.) is an alternate name for the directory you are in.
As a final experiment with pathnames, try
Using Files instead of the Terminal
ls /bin /usr/bin
Most of the commands we have seen so
Do some of the names look familiar? When you far produce output on the terminal; some, like
run a program, by typing its name after a ‘‘%’’, the editor, also take their input from the termi-
the system simply looks for a file of that name. nal. It is universal in UNIX that the terminal can
It looks first in your directory (where it typically be replaced by a file for either or both of input
doesn’t find it), then in ‘‘/bin’’ and finally in and output. As one example, you could say
‘‘/usr/bin’’. There is nothing magic about com-
ls
mands like cat or ls, except that they have been
collected into two places to be easy to find and to get a list of files. But you can also say
administer.
ls >filelist

What if you work regularly with someone to get a list of your files in the file ‘‘filelist’’.
else on common information in his directory? (‘‘filelist’’ will be created if it doesn’t already
You could just log in as your friend each time exist, or overwritten if it does.) The symbol ‘‘>’’
you want to, but you can also say ‘‘I want to is used throughout UNIX to mean ‘‘put the output
work on his files instead of my own’’. This is on the following file, rather than on the termi-
done by changing the directory that you are cur- nal’’. Nothing is produced on the terminal. As
rently in: another example, you could concatenate several
files into one by capturing the output of cat in a
chdir /usr/your–friend file:
Now when you use a filename in something like cat f1 f2 f3 >temp
cat or pr, it refers to the file in ‘‘your-friend’s’’
directory. Changing directories doesn’t affect Similarly, the symbol ‘‘<’’ means to take
any permissions associated with a file — if you the input for a program from the following file,
couldn’t access a file from your own directory, instead of from the terminal. Thus, you could
changing to another directory won’t alter that make up a script of commonly used editing com-
fact. mands and put them into a file called ‘‘script’’.
If you forget what directory you’re in, Then you can run the script on a file by saying
type ed file <script
pwd
(‘‘print working directory’’) to find out. Pipes
It is often convenient to arrange one’s One of the novel contributions of UNIX is
files so that all the files related to one thing are the idea of a pipe. A pipe is simply a way to
in a directory separate from other projects. For connect the output of one program to the input
example, when you write your book, you might of another program, so the two run as a sequence
want to keep all the text in a directory called of processes — a pipe-line.
book. So make one with
 -8-

For example, the output from interfering with what you’re

doing on the terminal, it would be better to have
pr f g h
said
will print the files ‘‘f ’’, ‘‘g’’ and ‘‘h’’, beginning
ed file <script >lines &
each on a new page. Suppose you want them
run together instead. You could say which would save the output lines in a file called
‘‘lines’’.
cat f g h >temp
pr temp When you initiate a command with ‘‘&’’,
rm temp UNIX replies with a number called the process
number, which identifies the command in case
but this is more work than necessary. Clearly you later want to stop it. If you do, you can say
what we want is to take the output of cat and
connect it to the input of pr. So let us use a kill process–number
pipe: You might also read ps (I).
cat f g h | pr You can say
The vertical bar means to take the output from (command–1; command–2; command–3) &
cat, which would normally have gone to the ter-
minal, and put it into pr, which formats it neatly. to start these commands in the background, or
you can start a background pipeline with
Any program that reads from the terminal
can read from a pipe instead; any program that command–1 | command–2 &
writes on the terminal can drive a pipe. You can
have as many elements in a pipeline as you wish. Just as you can tell the editor or some
similar program to take its input from a file
Many UNIX programs are written so that
instead of from the terminal, you can tell the
they will take their input from one or more files
shell to read a file to get commands. (Why not?
if file arguments are given; if no arguments are
The shell after all is just a program, albeit a
given they will read from the terminal, and thus
clever one.) For instance, suppose you want to
can be used in pipelines.
set tabs on your terminal, and find out the date
and who’s on the system every time you log in.
The Shell
Then you can put the three necessary commands
We have already mentioned once or twice ( tabs; date; who) into a file, let’s call it ‘‘xxx’’,
the mysterious ‘‘shell,’’ which is in fact sh (I). and then run it with either
The shell is the program that interprets what you
type as commands and arguments. It also looks sh xxx
after translating ‘‘*’’, etc., into lists of filenames. or
The shell has other capabilities too. For sh <xxx
example, you can start two programs with one
command line by separating the commands with This says to run the shell with the file ‘‘xxx’’ as
a semicolon; the shell recognizes the semicolon input. The effect is as if you had typed the con-
and breaks the line into two commands. Thus tents of ‘‘xxx’’ on the terminal. (If this is to be a
regular thing, you can eliminate the need to type
date; who ‘‘sh’’; see chmod (I) and sh (I).)
does both commands before returning with a The shell has quite a few other capabili-
‘‘%’’. ties as well, some of which we’ll get to in the
You can also have more than one program section on programming.
running simultaneously if you wish. For exam-
ple, if you are doing something time-consuming,
like the editor script of an earlier section, and III. DOCUMENT PREPARATION
you don’t want to wait around for the results
before starting something else, you can say UNIX is extensively used for document

ed file <script & preparation. There are three major formatting

programs, that is, programs which produce a text
The ampersand at the end of a command line with justified right margins, automatic page
says ‘‘start this command running, then take fur- numbering and titling, automatic hyphenation,
ther commands from the terminal immediately.’’ and the like. The simplest of these formatters is
Thus the script will begin, but you can do some- roff, which in fact is simple enough that if you
thing else at the same time. Of course, to keep type almost any text into a file and ‘‘roff’’ it, you
 -9-

will get plausibly formatted output. You can do If you simply want to ensure that subse-
better with a little knowledge, but basically it’s quent text appears on a fresh output line, you
easy to learn and use. We’ll get back to roff can use the command ‘‘.br’’ (for ‘‘break’’)
shortly. instead of ‘‘.sp’’.
nroff is similar to roff but does much less Most of the other commonly-used roff
for you automatically. It will do a great deal commands are equally simple. For example you
more, once you know how to use it. can center one or more lines with the ‘‘.ce’’
Both roff and nroff are designed to pro- command.
duce output on terminals, line-printers, and the .ce
like. The third formatter, troff (pronounced Title of Paper
‘‘tee-roff’’), instead drives a Graphic Systems .sp 2
phototypesetter, which produces very high qual-
ity output on photographic paper. This paper causes the title to be centered, then followed by
was printed on the phototypesetter by troff. two blank lines. As with ‘‘.sp’’, ‘‘.ce’’ can be
followed by a number; in that case, that many
Because nroff and troff are relatively
input lines are centered.
hard to learn to use effectively, several ‘‘pack-
ages’’ of canned formatting requests are avail- ‘‘.ul’’ underlines lines, and can also be
able which let you do things like paragraphs, followed by a number:
running titles, multi-column output, and so on, .ce 2
with little effort. Regrettably, details vary from .ul 2
system to system. An Earth–shaking Paper
.sp
ROFF John Q. Scientist
The basic idea of roff (and of nroff and
will center and underline the two text lines.
troff, for that matter) is that the text to be for-
Notice that the ‘‘.sp’’ between them is not part
matted contains within it ‘‘formatting com-
of the line count.
mands’’ that indicate in detail how the formatted
text is to look. For example, there might be You can get multiple-line spacing instead
commands that specify how long lines are, of the default single-spacing with the ‘‘.ls’’ com-
whether to use single or double spacing, and mand:
what running titles to use on each page. In gen- .ls 2
eral, you don’t have to spell out all of the possi-
ble formatting details. Most of them have causes double spacing.
‘‘default values’’, which you will get if you say If you’re typing things like tables, you
nothing at all. For example, unless you take spe- will not want the automatic filling-up and justifi-
cial precautions, you’ll get single-spaced output, cation of output lines that is done by default.
65-character lines, justified right margins, and You can turn this off with the command ‘‘.nf ’’
58 text lines per page when you roff a file. This (no-fill), and then back on again with ‘‘.fi’’ (fill).
is the reason that roff is so simple — most of the Thus
decisions have already been made for you.
this section is filled by default.
Some things do have to be done, however.
.nf
If you want a document broken into paragraphs,
here lines will appear just
you have to tell roff where to add the extra blank
as you typed them —
lines. This is done with the ‘‘.sp’’ command:
no extra spaces, no moving of words.
this is the end of one paragraph. .fi
.sp Now go back to filling up output lines.
This begins the next paragraph ...
You can change the line-length with
In roff (and in nroff and troff), formatting com- ‘‘.ll’’, and the left margin (the indent) by ‘‘.in’’.
mands consist of a period followed by two let- These are often used together to make offset
ters, and they must appear at the beginning of a blocks of text:
line, all by themselves. The ‘‘.sp’’ command
tells roff to finish printing any of the previous .ll −10
line that might be still unprinted, then print a .in +10
blank line before continuing. You can have this text will be moved 10
more space if you wish; ‘‘.sp 2’’ asks for 2 spaces to the right and the
spaces, and so on. lines will also be shortened 10
characters from the right. The
 – 10 –

‘‘+’’ and ‘‘−’’ mean to change too early. For example, if you decide that each
the previous value by that paragraph is to have a space and an indent of 10
much. Now revert: characters, you might type, before each,
.ll +10
.sp
.in −10
.ti +10
Notice that ‘‘.ll +10’’ adds ten characters to the
But what happens when later you decide that it
line length, while ‘‘.ll 10’’ makes the line ten
would have been better to have no space and an
characters long.
indent of only 5 characters? It’s tedious indeed
The ‘‘.ti’’ command indents (in either to go back and patch this up.
direction) just like ‘‘.in’’, except for only one
Fortunately, all of the formatters let you
line. Thus to make a new paragraph with a
delay decisions until the actual moment of run-
10-character indent, you would say
ning. The secret is to define a new operation
.sp (called a macro), for each formatting operation
.ti +10 you want to do, like making a new paragraph.
New paragraph ... You can say, in all three formatters,

You can put running titles on both top and .de PP

bottom of each page, like this: .sp
.ti +10
.he "left top"center top"right top" ..
.fo "left bottom"center bottom"right bottom"
This defines ‘‘.PP’’ as a new roff (or nroff or
The header or footer is divided into three parts, troff) operation, whose meaning is exactly
which are marked off by any character you like.
(We used a double quote.) If there’s nothing .sp
between the markers, that part of the title will be .ti +10
blank. If you use a percent sign anywhere in (The ‘‘..’’ marks the end of the definition.)
‘‘.he’’ or ‘‘.fo’’, the current page number will be Whenever ‘‘.PP’’ is encountered in the text, it is
inserted. So to get centered page numbers with as if you had typed the two lines of the definition
dashes around them, at the top, use in place of it.
.he ""− % −"" The beauty of this scheme is that now, if
you change your mind about what a paragraph
You can skip to the top of a new page at any
should look like, you can change the formatted
time with the ‘‘.bp’’ command; if ‘‘.bp’’ is fol-
output merely by changing the definition of
lowed by a number, that will be the new page
‘‘.PP’’ and re-running the formatter.
number.
As a rule of thumb, for all but the most
The foregoing is probably enough about
trivial jobs, you should type a document in terms
roff for you to go off and format most everyday
of a set of macros like ‘‘.PP’’, and then define
documents. Read roff (I) for more details.
them appropriately. As long as you have entered
the text in some systematic way, it can always be
Hints for Preparing Documents
cleaned up and re-formatted by a judicious com-
Most documents go through several ver- bination of editing and macro definitions. The
sions (always more than you expected) before packages of formatting commands that we men-
they are finally finished. Accordingly, you tioned earlier are simply collections of macros
should do whatever possible to make the job of designed for particular formatting tasks.
changing them easy.
One of the main differences between roff
First, when you do the purely mechanical and the other formatters is that macros in roff
operations of typing, type so subsequent editing can only be lines of text and formatting com-
will be easy. Start each sentence on a new line. mands. In nroff and troff, macros may have
Make lines short, and break lines at natural arguments, so they can have different effects
places, such as after commas and semicolons, depending on how they are called (in exactly the
rather than randomly. Since most people change same way that the ‘‘.sp’’ command has an argu-
documents by rewriting phrases and adding, ment, the number of spaces you want).
deleting and rearranging sentences, these pre-
cautions simplify any editing you have to do Miscellany
later.
In addition to the basic formatters, UNIX
The second aspect of making change easy provides a host of supporting programs. eqn
is not to commit yourself to formatting details
 - 11 -

and neqn let you integrate mathematics into the $p

text of a document, in a language that closely e chap1.2
resembles the way you would speak it aloud. 1p
spell and typo detect possible spelling mistakes $p
in a document. grep looks for lines containing a etc.
particular text pattern (rather like the editor’s
But instead you can do the job once and for all.
context search does, but on a whole series of
Type
files). For example,
ls chap* >temp
grep "ing$" chap*
to get the list of filenames into a file. Then edit
will find all lines ending in the letters ‘‘ing’’ in
this file to make the necessary series of editing
the series of files ‘‘chap*’’. (It is almost always
commands (using the global commands of ed),
a good practice to put quotes around the pattern
and write it into ‘‘script’’. Now the command
you’re searching for, in case it contains charac-
ters that have a special meaning for the shell.) ed <script
wc counts the words and (optionally) will produce the same output as the laborious
lines in a set of files. tr translates characters hand typing.
into other characters; for example it will convert
upper to lower case and vice versa. This trans- The pipe mechanism lets you fabricate
lates upper into lower: quite complicated operations out of spare parts
already built. For example, the first draft of the
tr "[A–Z]" "[a–z]" spell program was (roughly)

diff prints a list of the differences cat ... (collect the files)
between two files, so you can compare two ver- | tr ... (put each word on a new line,
sions of something automatically (which cer- delete punctuation, etc.)
tainly beats proofreading by hand). sort sorts | sort (into dictionary order)
files in a variety of ways; cref makes cross-refer- | uniq (strip out duplicates)
ences; ptx makes a permuted index (keyword-in- | comm (list words found in text but
context listing). not in dictionary)
Most of these programs are either inde-
pendently documented (like eqn and neqn), or Programming the Shell
are sufficiently simple that the description in the An option often overlooked by newcom-
UNIX Programmer’s Manual is adequate explana- ers is that the shell is itself a programming lan-
tion. guage, and since UNIX already has a host of
building-block programs, you can sometimes
IV. PROGRAMMING avoid writing a special purpose program merely
UNIX is a marvelously pleasant and pro- by piecing together some of the building blocks
ductive system for writing programs; productiv- with shell command files.
ity seems to be an order of magnitude higher As an unlikely example, suppose you
than on other interactive systems. want to count the number of users on the
There will be no attempt made to teach machine every hour. You could type
any of the programming languages available on
date
UNIX, but a few words of advice are in order.
who | wc –l
First, UNIX is written in C, as is most of the
applications code. If you are undertaking any- every hour, and write down the numbers, but that
thing substantial, C is the only reasonable is rather primitive. The next step is probably to
choice. More on that in a moment. But remem- say
ber that there are quite a few programs already
written, some of which have substantial power. (date; who | wc –l) >>users

The editor can be made to do things that which uses ‘‘>>’’ to append to the end of the file
would normally require special programs on ‘‘users’’. (We haven’t mentioned ‘‘>>’’ before
other systems. For example, to list the first and — it’s another service of the shell.) Now all you
last lines of each of a set of files, say a book, you have to do is to put a loop around this, and
could laboriously type ensure that it’s done every hour. Thus, place the
following commands into a file, say ‘‘count’’:
ed
e chap1.1
1p
 – 12 –

: loop Programming in C
(date; who | wc –l) >>users As we said, C is the language of choice:
sleep 3600 everything in UNIX is tuned to it. It is also a
goto loop remarkably easy language to use once you get
The command : is followed by a space and a started. Sections II and III of the manual
label, which you can then goto. Notice that it’s describe the system interfaces, that is, how you
quite legal to branch backwards. Now if you do I/O and similar functions.
issue the command You can write quite significant C pro-
grams with the level of I/O and system interface
sh count &
described in Programming in C: A Tutorial, if
the users will be counted every hour, and you you use existing programs and pipes to help.
can go on with other things. (You will have to For example, rather than learning how to open
use kill to stop counting.) and close files you can (at least temporarily)
If you would like ‘‘every hour’’ to be a write a program that reads from its standard
parameter, you can arrange for that too: input, and use cat to concatentate several files
into it. This may not be adequate for the long
: loop run, but for the early stages it’s just right.
(date; who | wc – l) >>users
There are a number of supporting pro-
sleep $1
grams that go with C. The C debugger, cdb, is
goto loop
marginally useful for digging through the dead
‘‘$1’’ means the first argument when this proce- bodies of C programs. db, the assembly lan-
dure is invoked. If you say guage debugger, is actually more useful most of
the time, but you have to know more about the
sh count 60 machine and system to use it well. The most
it will count every minute. A shell program can effective debugging tool is still careful thought,
have up to nine arguments, ‘‘$1’’ through ‘‘$9’’. coupled with judiciously placed print statements.
The other aspect of programming is con- You can instrument C programs and thus
ditional testing. The if command can test condi- find out where they spend their time and what
tions and execute commands accordingly. As a parts are worth optimising. Compile the rou-
simple example, suppose you want to add to tines with the ‘‘-p’’ option; after the test run use
your login sequence something to print your prof to print an execution profile. The com-
mail if you have some. Thus, knowing that mail mand time will give you the gross run-time
is stored in a file called ‘mailbox’, you could say statistics of a program, but it’s not super accurate
or reproducible.
if –r mailbox mail
C programs that don’t depend too much
This says ‘‘if the file ‘mailbox’ is readable, on special features of UNIX can be moved to the
execute the mail command.’’ Honeywell 6070 and IBM 370 systems with mod-
As another example, you could arrange est effort. Read The GCOS C Library by M. E.
that the ‘‘count’’ procedure count every hour by Lesk and B. A. Barres for details.
default, but allow an optional argument to spec-
ify a different time. Simply replace the ‘‘sleep Miscellany
$1’’ line by If you have to use Fortran, you might
consider ratfor, which gives you the decent con-
if $1x = x sleep 3600
trol structures and free-form input that character-
if $1x != x sleep $1
ize C, yet lets you write code that is still portable
The construction to other environments. Bear in mind that UNIX
Fortran tends to produce large and relatively
if $1x = x
slow-running programs. Furthermore, support-
tests whether ‘‘$1’’, the first argument, was pre- ing software like db, prof, etc., are all virtually
sent or absent. useless with Fortran programs.
More complicated conditions can be If you want to use assembly language (all
tested: you can find out the status of an executed heavens forfend!), try the implementation lan-
command, and you can combine conditions with guage LIL, which gives you many of the advan-
‘and’, ‘or’, ‘not’ and parentheses — see if (I). tages of a high-level language, like decent con-
You should also read shift (I) which describes trol flow structures, but still lets you get close to
how to manipulate arguments to shell command the machine if you really want to.
files.
 - 13 -

If your application requires you to trans- Library. Bell Laboratories internal memoran-
late a language into a set of actions or another dum.
language, you are in effect building a compiler, K. Thompson and D. M. Ritchie, Setting Up
though probably a small one. In that case, you UNIX.
should be using the yacc compiler-compiler, M. D. McIlroy, UNIX Summary.
which helps you develop a compiler quickly.
D. M. Ritchie, The UNIX I/O System.
V. UNIX READING LIST A. D. Hall, The M6 Macro Processor, Comput-
ing Science Tech. Rep. 2.
General:
J. F. Ossanna, NROFF User’s Manual — Second
UNIX Programmer’s Manual (Ken Thompson, Edition, Bell Laboratories internal memoran-
Dennis Ritchie, and a cast of thousands). Lists dum.
commands, system routines and interfaces, file D. M. Ritchie and K. Thompson, Regenerating
formats, and some of the maintenance proce- System Software.
dures. You can’t live without this, although you
B. W. Kernighan, Ratfor—A Rational Fortran,
will probably only read section I.
Bell Laboratories internal memorandum.
The UNIX Time-sharing System (Ken Thompson, M. D. McIlroy, Synthetic English Speech by
Dennis Ritchie). CACM, July 1974. An Rule, Computing Science Tech. Rep. 14.
overview of the system, for people interested in M. D. McIlroy, Bell Laboratories internal mem-
operating systems. Worth reading by anyone orandum.
who programs. Contains a remarkable number
J. F. Ossanna, TROFF Users’ Manual, Bell Lab-
of one-sentence observations on how to do
oratories internal memorandum.
things right.
B. W. Kernighan, TROFF Made Trivial, Bell
Document Preparation: Laboratories internal memorandum.
A Tutorial Introduction to the UNIX Text Editor. R. H. Morris and L. L. Cherry, Computer Detec-
(Brian Kernighan). Bell Laboratories internal tion of Typographical Errors, Computing Sci-
memorandum. Weak on the more esoteric uses ence Tech. Rep. 18.
of the editor, but still probably the easiest way to S. C. Johnson, YACC (Yet Another Compiler-
learn ed. Compiler), Bell Laboratories internal memoran-
Typing Documents on UNIX. (Mike Lesk). Bell dum.
Laboratories internal memorandum. A macro P. J. Plauger, Programming in LIL: A Tutorial,
package to isolate the novice from the vagaries Bell Laboratories internal memorandum.
of the formatting programs. If this specific
package isn’t available on your system, some-
thing similar probably is. This one works with
both nroff and troff.
Index
Programming:
Programming in C: A Tutorial (Brian
Kernighan). Bell Laboratories internal memo- & (asynchronous process) 8
randum. The easiest way to start learning C, but ; (multiple processes) 8
it’s no help at all with the interface to the system * (pattern match) 5
beyond the simplest IO. Should be read in con- [ ] (pattern match) 6
junction with ? (pattern match) 6
C Reference Manual (Dennis Ritchie). Bell <> (redirect I/O) 7
Laboratories internal memorandum. An excel- >> (file append) 12
lent reference, but a bit heavy going for the backslash (\) 2
beginner, especially one who has never used a cat (concatenate files) 4
language like C. cdb (C debugger) 12
chdir (change directory) 7
Others:
chmod (change protection) 7
D. M. Ritchie, UNIX Assembler Reference command arguments 4
Manual. command files 8
B. W. Kernighan and L. L. Cherry, A System for cp (copy files) 5
Typesetting Mathematics, Computing Science cref (cross reference) 11
Tech. Rep. 17. date 2
M. E. Lesk and B. A. Barres, The GCOS C db (assembly debugger) 13
delete (DEL) 2
 - 14 -

diff (file comparison) 11 troff (typesetting) 9

directories 7 typo (find spelling mistakes) 11
document formatting 9 wc (word count) 11
ed (editor) 3 who (who is looged in) 2
editor programming 11 write (to a user) 3
EOT (end of file) 3 yacc (compiler-compiler) 13
eqn (mathematics) 11
erase character (#) 2
file system structure 6
filenames 5
file protection 7
goto 12
grep (pattern matching) 11
if (condition test) 12
index 14
kill a program 8
kill a character (@) 2
lil (high-level assembler) 13
login 1
logout 2
ls (list file names) 4
macro for formatting 10
mail 2
multi-columns printing (pr) 5
mv (move files) 5
nroff 9
on-line manual 3
opr (offline print) 5
pathname 6
pattern match in filenames 5
pipes ( | ) 8
pr (print files) 4
prof (run-time monitor) 13
protection 7
ptx (permuted index) 11
pwd (working directory) 7
quotes 6
ratfor (decent Fortran) 13
readahead 2
reading list 13
redirect I/O (<>) 7
RETURN key 1
rm (remove files) 5
rmdir (remove directory) 7
roff (text formatting) 9
root (of file system) 6
shell (command interpreter) 8
shell arguments ($) 12
shell programming 12
shift (shell arguments) 12
sleep 12
sort 11
spell (find spelling mistakes)
stopping a program 2
stty (set terminal options) 2
tabs (set tab stops) 2
terminal types 1
time (time programs) 13
tr (translate characters) 11
 RATFOR — A Rational Fortran

B. W. Kernighan

Bell Laboratories
Murray Hill, N. J. 07974

Fortran programs are hard to read, write and debug. To make program development easier,
RATFOR provides a set of decent control structures:

statement grouping
completely general if - else statements
while, for and do for looping
break and next for controlling loop exits

and some ‘‘syntactic sugar’’:

free form input (e.g., multiple statements/line)

unobtrusive comment convention
translation of >, >=, etc., into .GT., .GE., etc.
‘‘define’’ statement for symbolic parameters
‘‘include’’ statement for including source files

RATFOR not only makes programming in Fortran more enjoyable, but also allows structured
programming, in the sense of coding without GOTO’s. RATFOR programs tend to be markedly easier to
write, read, and make correct than their equivalents in standard Fortran.

RATFOR is a preprocessor, translating the input into standard Fortran constructions. RATFOR
programs may readily be written so the generated Fortran is portable; program transferability is easy to
achieve. RATFOR is written in itself, so it is also portable.

The grammar of RATFOR is as follows:

prog : stat
| prog stat
stat : if( condition ) stat
| if( condition ) stat else stat
| while( condition ) stat
| for( initialization; condition; increment ) stat
| do do-part stat
| break
| next
| digits stat
| { prog }
| anything unrecognizable

In the grammar above, condition can be any legal Fortran condition like "‘A .EQ. B’," i.e., anything that
could appear in a legal Fortran logical IF statement. stat is, of course, any Fortran or RATFOR statement,
or any collection of these enclosed in braces.
 -2-

The while statement performs a loop while some specified condition is true. The test is per-
formed at the beginning of the loop, so it is possible to do a while zero times, which can’t be done with a
Fortran DO.

The for statement is a somewhat generalized while statement that allows an initialization and
an incrementing step as well as a termination condition on a loop. initialization is any single Fortran state-
ment, which gets done once before the loop begins. increment is any single Fortran statement, which gets
done at the end of each pass through the loop, before the test.

for and while are useful for backward loops, chaining along lists, loops that must be done zero
times, and similar things which are hard to express with a DO statement, and obscure to write out directly.

The do statement is like a Fortran DO, except that no label or CONTINUE is needed. The do-
part that follows the do keyword has to be something that can legally go into a Fortran DO statement.

A break causes an immediate exit from a for, while or do to the following statement. The next
statement causes an immediate transfer to the increment part of a for or do, and the test part of a while.
Both break and next refer to the innermost enclosing for, while or do.

Statements can be placed anywhere on a line; long statements are continued automatically.
Multiple statements may appear on one line, if they are separated by semicolons. No semi-colon is needed
at the end of a line, if RATFOR can guess whether the statement ends there. Lines ending with any charac-
ters obviously a continuation, like plus or comma, are assumed to be continued on the next line. Any state-
ment that begins with an all-numeric field is assumed to be a Fortran label, and placed in columns 1-5. PP
A ‘#’ character in a line marks the beginning of a comment; the comment is terminated by the end of a line.

Text enclosed in matching single or double quotes is converted to nH... by RATFOR, but is
otherwise unaltered. Characters like ‘>’, ‘>=’, and ‘&’ are translated into their longer Fortran equivalents
‘.GT.’, ‘.GE’, and ‘.AND’, except within quotes.

Any string of alphanumeric characters can be defined as a name; thereafter, whenever that
name occurs in the input (delimited by non-alphanumerics) it is replaced by the rest of the definition line
(comments are stripped off).

An entire source file may be included by saying ‘‘include filename’’ at the appropriate place.
 YACC − Yet Another Compiler-Compiler

Stephen C. Johnson

ABSTRACT

Computer program input generally has some structure; in fact, every computer pro-
gram which does input can be thought of as defining an ‘‘input language’’ which it
accepts. The input languages may be as complex as a programming language, or as sim-
ple as a sequence of numbers. Unfortunately, standard input facilities are restricted, diffi-
cult to use and change, and do not completely check their inputs for validity.
Yacc provides a general tool for controlling the input to a computer program. The
Yacc user describes the structures of his input, together with code which is to be invoked
when each such structure is recognized. Yacc turns such a specification into a subroutine
which may be invoked to handle the input process; frequently, it is convenient and appro-
priate to have most of the flow of control in the user´s application handled by this subrou-
tine.
The input subroutine produced by Yacc calls a user supplied routine to return the
next basic input item. Thus, the user can specify his input in terms of individual input
characters, or, if he wishes, in terms of higher level constructs such as names and num-
bers. The user supplied routine may also handle idiomatic features such as comment and
continuation conventions, which typically defy easy specification.
Yacc is written in C[7], and runs under UNIX. The subroutine which is output may
be in C or in Ratfor[4], at the user´s choice; Ratfor permits translation of the output sub-
routine into portable Fortran[5]. The class of specifications accepted is a very general
one, called LALR(1) grammars with disambiguating rules. The theory behind Yacc has
been described elsewhere[1,2,3].
Yacc was originally designed to help produce the ‘‘front end’’ of compilers; in
addition to this use, it has been successfully used in many application programs, includ-
ing a phototypesetter language, a document retrieval system, a Fortran debugging system,
and the Ratfor compiler.

January 6, 1998
 YACC − Yet Another Compiler-Compiler

Stephen C. Johnson

Section 0: Introduction
Yacc provides a general tool for imposing structure on the input to a computer program. The Yacc
user prepares a specification of the input process; this includes rules which describe the input structure,
code which is to be invoked when these structures are recognized, and a low-level routine to do the basic
input. Yacc then produces a subroutine to do the input procedure; this subroutine, called a parser, calls the
user-supplied low-level input routine (called the lexical analyzer) to pick up the basic items (called tokens)
from the input stream. These tokens are organized according to the input structure rules, called grammar
rules; when one of these rules has been recognized, then the user code supplied for this rule, called an
action, is invoked; actions have the ability to return values and make use of the values of other actions.
The heart of the input specification is a collection of grammar rules. Each rule describes an allow-
able structure and gives it a name. For example, one grammar rule might be
date : month name day ´,´ year ;
Here, date, month name, day, and year represent structures of interest in the input process; presumably,
month name, day, and year are defined elsewhere. The comma ‘‘,’’ is quoted by single quotes; this implies
that the comma is to appear literally in the input. The colon and semicolon merely serve as punctuation in
the rule, and have no significance in controlling the input. Thus, with proper definitions, the input
July 4, 1776
might be matched by the above rule.
As we mentioned above, an important part of the input process is carried out by the lexical analyzer.
This user routine reads the true input stream, recognizing those structures which are more conveniently or
more efficiently recognized directly, and communicates these recognized tokens to the parser. For histori-
cal reasons, the name of a structure recognized by the lexical analyzer is called a terminal symbol name,
while the name of a structure recognized by the parser is called a nonterminal symbol name. To avoid the
obvious confusion of terminology, we shall usually refer to terminal symbol names as token names.
There is considerable leeway in deciding whether to recognize structures by the lexical analyzer or by
a grammar rule. Thus, in the above example it would be possible to have other rules of the form
month name : ´J´ ´a´ ´n´ ;
month name : ´F´ ´e´ ´b´ ;

...

month name : ´D´ ´e´ ´c´ ;

Here, the lexical analyzer would only need to recognize individual letters, and month name would be a
nonterminal symbol. Rules of this sort tend to be a bit wasteful of time and space, and may even restrict the
power of the input process (although they are easy to write). For a more efficient input process, the lexical
analyzer itself might recognize the month names, and return an indication that a month name was seen; in
this case, month name would be a token.
Literal characters, such as ‘‘,’’, must also be passed through the lexical analyzer, and are considered
tokens.
As an example of the flexibility of the grammar rule approach, we might add to the above specifica-
tions the rule
 -2-

date : month ´/´ day ´/´ year ;

and thus optionally allow the form
7/4/1776
as a synonym for
July 4, 1776
In most cases, this new rule could be ‘‘slipped in’’ to a working system with minimal effort, and a very
small chance of disrupting existing input.
Frequently, the input being read does not conform to the specifications due to errors in the input. The
parsers produced by Yacc have the very desirable property that they will detect these input errors at the ear-
liest place at which this can be done with a left-to-right scan; thus, not only is the chance of reading and
computing with bad input data substantially reduced, but the bad data can usually be quickly found. Error
handling facilities, entered as part of the input specifications, frequently permit the reentry of bad data, or
the continuation of the input process after skipping over the bad data.
In some cases, Yacc fails to produce a parser when given a set of specifications. For example, the
specifications may be self contradictory, or they may require a more powerful recognition mechanism than
that available to Yacc. The former cases probably represent true design errors; the latter cases can often be
corrected by making the lexical analyzer more powerful, or by rewriting some of the grammar rules. The
class of specifications which Yacc can handle compares very favorably with other systems of this type;
moreover, the constructions which are difficult for Yacc to handle are also frequently difficult for human
beings to handle. Some users have reported that the discipline of formulating valid Yacc specifications for
their input revealed errors of conception or design early in the program development.
The next several sections describe the basic process of preparing a Yacc specification; Section 1
describes the preparation of grammar rules, Section 2 the preparation of the user supplied actions associ-
ated with these rules, and Section 3 the preparation of lexical analyzers. In Section 4, we discuss the diag-
nostics produced when Yacc is unable to produce a parser from the given specifications. This section also
describes a simple, frequently useful mechanism for handling operator precedences. Section 5 discusses
error detection and recovery. Sections 6C and 6R discuss the operating environment and special features of
the subroutines which Yacc produces in C and Ratfor, respectively. Section 7 gives some hints which may
lead to better designed, more efficient, and clearer specifications. Finally, Section 8 has a brief summary.
Appendix A has a brief example, and Appendix B tells how to run Yacc on the UNIX operating system.
Appendix C has a brief description of mechanisms and syntax which are no longer actively supported, but
which are provided for historical continuity with older versions of Yacc.

Section 1: Basic Specifications

As we noted above, names refer to either tokens or nonterminal symbols. Yacc requires those names
which will be used as token names to be declared as such. In addition, for reasons which will be discussed
in Section 3, it is usually desirable to include the lexical analyzer as part of the specification file; it may be
useful to include other programs as well. Thus, every specification file consists of three sections: the decla-
rations, (grammar) rules, and programs. The sections are separated by double percent ‘‘%%’’ marks. (The
per-cent ‘‘%’’ is generally used in Yacc specifications as an escape character.)
In other words, a full specification file looks like
declarations
%%
rules
%%
programs

The declaration section may be empty. Moreover, if the programs section is omitted, the second %%
mark may be omitted also; thus, the smallest legal Yacc specification is
 -3-

%%
rules

Blanks, tabs, and newlines are ignored except that they may not appear in names or multi-character
reserved symbols. Comments may appear wherever a name or operator is legal; they are enclosed in /* . . .
*/, as in C and PL/I.
The rules section is made up of one or more grammar rules. A grammar rule has the form:
A : BODY ;
A represents a nonterminal name, and BODY represents a sequence of zero or more names and literals.
Notice that the colon and the semicolon are Yacc punctuation.
Names may be of arbitrary length, and may be made up of letters, dot ‘‘.’’, underscore ‘‘ ’’, and non-
initial digits. Notice that Yacc considers that upper and lower case letters are distinct. The names used in
the body of a grammar rule may represent tokens or nonterminal symbols.
A literal consists of a character enclosed in single quotes ‘‘´’’. As in C, the backslash ‘‘\’’ is an
escape character within literals, and all the C escapes are recognized. Thus
´\n´ represents newline
´\r´ represents return
´\´´ represents single quote ‘‘´’’
´\\´ represents backslash ‘‘\’’
´\t´ represents tab
´\b´ represents backspace
´\xxx´ represents ‘‘xxx’’ in octal
For a number of technical reasons, the nul character (´\0´ or 000) should never be used in grammar rules.
If there are several grammar rules with the same left hand side, the vertical bar ‘‘|’’ can be used to
avoid rewriting the left hand side. In addition, the semicolon at the end of a rule can be dropped before a
vertical bar. Thus the grammar rules
A:BCD ;
A:EF ;
A:G ;
can be given to Yacc as
A: BCD|
EF|
G;
It is not necessary that all grammar rules with the same left side appear together in the grammar rules sec-
tion, although it makes the input much more readable, and easy to change.
If a nonterminal symbol matches the empty string, this can be indicated in the obvious way:
empty : ;

As we mentioned above, names which represent tokens must be declared as such. The simplest way
of doing this is to write
%token name1 name2 . . .
in the declarations section. (See Sections 3 and 4 for much more discussion). Every name not defined in
the declarations section is assumed to represent a nonterminal symbol. If, by the end of the rules section,
some nonterminal symbol has not appeared on the left of any rule, then an error message is produced and
Yacc halts.
The left hand side of the first grammar rule in the grammar rules section has special importance; it is
taken to be the controlling nonterminal symbol for the entire input process; in technical language it is called
 -4-

the start symbol. In effect, the parser is designed to recognize the start symbol; thus, this symbol generally
represents the largest, most general structure described by the grammar rules.
The end of the input is signaled by a special token, called the endmarker. If the tokens up to, but not
including, the endmarker form a structure which matches the start symbol, the parser subroutine returns to
its caller when the endmarker is seen; we say that it accepts the input. If the endmarker is seen in any other
context, it is an error.
It is the job of the user supplied lexical analyzer to return the endmarker when appropriate; see sec-
tion 3, below. Frequently, the endmarker token represents some reasonably obvious I/O status, such as
‘‘end-of-file’’ or ‘‘end-of-record’’.

Section 2: Actions
To each grammar rule, the user may associate an action to be performed each time the rule is recog-
nized in the input process. This action may return a value, and may obtain the values returned by previous
actions in the grammar rule. In addition, the lexical analyzer can return values for tokens, if desired.
When invoking Yacc, the user specifies a programming language; currently, Ratfor and C are sup-
ported. An action is an arbitrary statement in this language, and as such can do input and output, call sub-
programs, and alter external vectors and variables (recall that a ‘‘statement’’ in both C and Ratfor can be
compound and do many distinct tasks). An action is specified by an equal sign ‘‘=’’ at the end of a gram-
mar rule, followed by one or more statements, enclosed in curly braces ‘‘{’’ and ‘‘}’’. For example,
A: ´(´ B ´)´ = { hello( 1, "abc" ); }
and
XXX: YYY ZZZ =
{
printf("a message\n");
flag = 25;
}
are grammar rules with actions in C. A grammar rule with an action need not end with a semicolon; in fact,
it is an error to have a semicolon before the equal sign.
To facilitate easy communication between the actions and the parser, the action statements are altered
slightly. The symbol ‘‘dollar sign’’ ‘‘$’’ is used as a signal to Yacc in this context.
To return a value, the action normally sets the pseudo-variable ‘‘$$’’ to some integer value. For
example, an action which does nothing but return the value 1 is
= { $$ = 1; }

To obtain the values returned by previous actions and the lexical analyzer, the action may use the
(integer) pseudo-variables $1, $2, . . ., which refer to the values returned by the components of the right
side of a rule, reading from left to right. Thus, if the rule is
A: B C D ;
for example, then $2 has the value returned by C, and $3 the value returned by D.
As a more concrete example, we might have the rule
expression: ´(´ expression ´)´ ;
We wish the value returned by this rule to be the value of the expression in parentheses. Then we write
expression: ´(´ expression ´)´ = { $$ = $2 ; }

As a default, the value of a rule is the value of the first element in it ($1). This is true even if there is
no explicit action given for the rule. Thus, grammar rules of the form
 -5-

A: B ;
frequently need not have an explict action.
Notice that, although the values of actions are integers, these integers may in fact contain pointers (in
C) or indices into an array (in Ratfor); in this way, actions can return and reference more complex data
structures.
Sometimes, we wish to get control before a rule is fully parsed, as well as at the end of the rule.
There is no explicit mechanism in Yacc to allow this; the same effect can be obtained, however, by intro-
ducing a new symbol which matches the empty string, and inserting an action for this symbol. For exam-
ple, we might have a rule describing an ‘‘if ’’ statement:
statement: IF ´(´ expr ´)´ THEN statement
Suppose that we wish to get control after seeing the right parenthesis in order to output some code. We
might accomplish this by the rules:
statement: IF ´(´ expr ´)´ actn THEN statement
= { call action1 }

actn: /* matches the empty string */

= { call action2 }

Thus, the new nonterminal symbol actn matches no input, but serves only to call action2 after the
right parenthesis is seen.
Frequently, it is more natural in such cases to break the rule into parts where the action is needed.
Thus, the above example might also have been written
statement: ifpart THEN statement
= { call action1 }

ifpart: IF ´(´ expr ´)´

= { call action2 }

In many applications, output is not done directly by the actions; rather, a data structure, such as a
parse tree, is constructed in memory, and transformations are applied to it before output is generated. Parse
trees are particularly easy to construct, given routines which build and maintain the tree structure desired.
For example, suppose we have a C function ‘‘node’’, written so that the call
node( L, n1, n2 )
creates a node with label L, and descendants n1 and n2, and returns a pointer to the newly created node.
Then we can cause a parse tree to be built by supplying actions such as:
expr: expr ´+´ expr
= { $$ = node( ´+´, $1, $3 ); }
in our specification.
The user may define other variables to be used by the actions. Declarations and definitions can
appear in two places in the Yacc specification: in the declarations section, and at the head of the rules sec-
tions, before the first grammar rule. In each case, the declarations and definitions are enclosed in the marks
‘‘%{’’ and ‘‘%}’’. Declarations and definitions placed in the declarations section have global scope, and
are thus known to the action statements and the lexical analyzer. Declarations and definitions placed at the
head of the rules section have scope local to the action statements. Thus, in the above example, we might
have included
%{ int variable 0; %}
in the declarations section, or, perhaps,
 -6-

%{ static int variable; %}

at the head of the rules section. If we were writing Ratfor actions, we might want to include some COM-
MON statements at the beginning of the rules section, to allow for easy communication between the actions
and other routines. For both C and Ratfor, Yacc has used only external names beginning in ‘‘yy’’; the user
should avoid such names.

Section 3: Lexical Analysis

The user must supply a lexical analyzer which reads the input stream and communicates tokens (with
values, if desired) to the parser. The lexical analyzer is an integer valued function called yylex, in both C
and Ratfor. The function returns an integer which represents the type of the token. The value to be associ-
ated in the parser with that token is assigned to the integer variable yylval. Thus, a lexical analyzer written
in C should begin
yylex ( ) {
extern int yylval;
...
while a lexical analyzer written in Ratfor should begin
integer function yylex(yylval)
integer yylval
...

Clearly, the parser and the lexical analyzer must agree on the type numbers in order for communica-
tion between them to take place. These numbers may be chosen by Yacc, or chosen by the user. In either
case, the ‘‘define’’ mechanisms of C and Ratfor are used to allow the lexical analyzer to return these num-
bers symbolically. For example, suppose that the token name DIGIT has been defined in the declarations
section of the specification. The relevant portion of the lexical analyzer (in C) might look like:
yylex( ) {
extern int yylval;
int c;
...
c = getchar( );
...
if( c >= ´0´ && c <= ´9´ ) {
yylval = c−´0´;
return(DIGIT);
}
...

The relevant portion of the Ratfor lexical analyzer might look like:
 -7-

integer function yylex(yylval)

integer yylval, digits(10), c
...
data digits(1) / "0" /;
data digits(2) / "1" /;
...
data digits(10) / "9" /;
...
# set c to the next input character
...
do i = 1, 10 {
if(c .EQ. digits(i)) {
yylval = i−1
yylex = DIGIT
return
}
}
...

In both cases, the intent is to return a token type of DIGIT, and a value equal to the numerical value
of the digit. Provided that the lexical analyzer code is placed in the programs section of the specification,
the identifier DIGIT will be redefined to be equal to the type number associated with the token name
DIGIT.
This mechanism leads to clear and easily modified lexical analyzers; the only pitfall is that it makes it
important to avoid using any names in the grammar which are reserved or significant in the chosen lan-
guage; thus, in both C and Ratfor, the use of token names of ‘‘if ’’ or ‘‘yylex’’ will almost certainly cause
severe difficulties when the lexical analyzer is compiled. The token name ‘‘error’’ is reserved for error han-
dling, and should not be used naively (see Section 5).
As mentioned above, the type numbers may be chosen by Yacc or by the user. In the default situa-
tion, the numbers are chosen by Yacc. The default type number for a literal character is the numerical value
of the character, considered as a 1 byte integer. Other token names are assigned type numbers starting at
257. It is a difficult, machine dependent operation to determine the numerical value of an input character in
Ratfor (or Fortran). Thus, the Ratfor user of Yacc will probably wish to set his own type numbers, or not
use any literals in his specification.
To assign a type number to a token (including literals), the first appearance of the token name or lit-
eral in the declarations section can be immediately followed by a nonnegative integer. This integer is taken
to be the type number of the name or literal. Names and literals not defined by this mechanism retain their
default definition. It is important that all type numbers be distinct.
There is one exception to this situation. For sticky historical reasons, the endmarker must have type
number 0. Note that this is not unattractive in C, since the nul character is returned upon end of file; in Rat-
for, it makes no sense. This type number cannot be redefined by the user; thus, all lexical analyzers should
be prepared to return 0 as a type number upon reaching the end of their input.

Section 4: Ambiguity, Conflicts, and Precedence

A set of grammar rules is ambiguous if there is some input string which can be structured in two or
more different ways. For example, the grammar rule
expr : expr ´−´ expr ;
is a natural way of expressing the fact that one way of forming an arithmetic expression is to put two other
expressions together with a minus sign between them. Unfortunately, this grammar rule does not com-
pletely specify the way that all complex inputs should be structured. For example, if we have input of the
form
 -8-

expr − expr − expr

the rule would permit us to treat this input either as
( expr − expr ) − expr
or as
expr − ( expr − expr )
(We speak of the first as left association of operators, and the second as right association).
Yacc detects such ambiguities when it is attempting to build the parser. It is instructive to consider
the problem that confronts the parser when it is given an input such as
expr − expr − expr
When the parser has read the second expr, the input which it has seen:
expr − expr
matches the right side of the grammar rule above. One valid thing for the parser to do is to reduce the input
it has seen by applying this rule; after applying the rule, it would have reduced the input it had already seen
to expr (the left side of the rule). It could then read the final part of the input:
− expr
and again reduce by the rule. We see that the effect of this is to take the left associative interpretation.
Alternatively, when the parser has seen
expr − expr
it could defer the immediate application of the rule, and continue reading (the technical term is shifting) the
input until it had seen
expr − expr − expr
It could then apply the grammar rule to the rightmost three symbols, reducing them to expr and leaving
expr − expr
Now it can reduce by the rule again; the effect is to take the right associative interpretation. Thus, having
read
expr − expr
the parser can do two legal things, a shift or a reduction, and has no way of deciding between them. We
refer to this as a shift/reduce conflict. It may also happen that the parser has a choice of two legal reduc-
tions; this is called a reduce/reduce conflict.
When there are shift/reduce or reduce/reduce conflicts, Yacc still produces a parser. It does this by
selecting one of the valid steps wherever it has a choice. A rule which describes which choice to make in a
given situation is called a disambiguating rule.
Yacc has two disambiguating rules which are invoked by default, in the absence of any user directives
to the contrary:
1. In a shift/reduce conflict, the default is to do the shift.
2. In a reduce/reduce conflict, the default is to reduce by the earlier grammar rule (in the input
sequence).
Rule 1 implies that reductions are deferred whenever there is a choice, in favor of shifts. Rule 2 gives
the user rather crude control over the behavior of the parser in this situation, but the proper use of
reduce/reduce conflicts is still a black art, and is properly considered an advanced topic.
Conflicts may arise because of mistakes in input or logic, or because the grammar rules, while con-
sistent, require a more complex parser than Yacc can construct. In these cases, the application of
 -9-

disambiguating rules is inappropriate, and leads to a parser which is in error. For this reason, Yacc always
reports the number of shift/reduce and reduce/reduce conflicts which were resolved by Rule 1 and Rule 2.
In general, whenever it is possible to apply disambiguating rules to produce a correct parser, it is also
possible to rewrite the grammar rules so that the same inputs are read, but there are no conflicts. For this
reason, most previous systems like Yacc have considered conflicts to be fatal errors. Our experience has
suggested that this rewriting is somewhat unnatural to do, and produces slower parsers; thus, Yacc will pro-
duce parsers even in the presence of conflicts.
As an example of the power of disambiguating rules, consider a fragment from a programming lan-
guage involving an ‘‘if-then-else’’ construction:
stat : IF ´(´ cond ´)´ stat |
IF ´(´ cond ´)´ stat ELSE stat ;
Here, we consider IF and ELSE to be tokens, cond to be a nonterminal symbol describing conditional (logi-
cal) expressions, and stat to be a nonterminal symbol describing statements. In the following, we shall refer
to these two rules as the simple-if rule and the if-else rule, respectively.
These two rules form an ambiguous construction, since input of the form
IF ( C1 ) IF ( C2 ) S1 ELSE S2
can be structured according to these rules in two ways:
IF ( C1 ) {
IF ( C2 ) S1
}
ELSE S2
or
IF ( C1 ) {
IF ( C2 ) S1
ELSE S2
}
The second interpretation is the one given in most programming languages which have this construct. Each
ELSE is associated with the last preceding ‘‘un-ELSE’d’’ IF. In this example, consider the situation where
the parser has seen
IF ( C1 ) IF ( C2 ) S1
and is looking at the ELSE. It can immediately reduce by the simple-if rule to get
IF ( C1 ) stat
and then read the remaining input,
ELSE S2
and reduce
IF ( C1 ) stat ELSE S2
by the if-else rule. This leads to the first of the above groupings of the input.
On the other hand, we may shift the ELSE and read S2, and then reduce the right hand portion of
IF ( C1 ) IF ( C2 ) S1 ELSE S2
by the if-else rule to get
IF ( C1 ) stat
which can be reduced by the simple-if rule. This leads to the second of the above groupings of the input,
which is usually desired.
 - 10 -

Once again the parser can do two valid things − we have a shift/reduce conflict. The application of
disambiguating rule 1 tells the parser to shift in this case, which leads to the desired grouping.
Notice that this shift/reduce conflict arises only when there is a particular current input symbol,
ELSE, and particular inputs already seen, such as
IF ( C1 ) IF ( C2 ) S1
In general, there may be many conflicts, and each one will be associated with an input symbol and a set of
previously read inputs. The previously read inputs are characterized by the state of the parser, which is
assigned a nonnegative integer. The number of states in the parser is typically two to five times the number
of grammar rules.
When Yacc is invoked with the verbose (−v) option (see Appendix B), it produces a file of user out-
put which includes a description of the states in the parser. For example, the output corresponding to the
above example might be:
23: shift/reduce Conflict (Shift 45, Reduce 18) on ELSE

State 23

stat : IF ( cond ) stat

stat : IF ( cond ) stat ELSE stat

ELSE shift 45
. reduce 18

The first line describes the conflict, giving the state and the input symbol. The state title follows, and a
brief description of the grammar rules which are active in this state. The underline ‘‘ ’’ describes the por-
tions of the grammar rules which have been seen. Thus in the example, in state 23 we have seen input
which corresponds to
IF ( cond ) stat
and the two grammar rules shown are active at this time. The actions possible are, if the input symbol is
ELSE, we may shift into state 45. In this state, we should find as part of the description a line of the form
stat : IF ( cond ) stat ELSE stat
because in this state we will have read and shifted the ELSE. Back in state 23, the alternative action,
described by ‘‘.’’, is to be done if the input symbol is not mentioned explicitly in the above actions; thus, in
this case, if the input symbol is not ELSE, we should reduce by grammar rule 18, which is presumably
stat : IF ´(´ cond ´)´ stat
Notice that the numbers following ‘‘shift’’ commands refer to other states, while the numbers following
‘‘reduce’’ commands refer to grammar rule numbers. In most states, there will be only one reduce action
possible in the state, and this will always be the default command. The user who encounters unexpected
shift/reduce conflicts will probably want to look at the verbose output to decide whether the default actions
are appropriate. In really tough cases, the user might need to know more about the behavior and construc-
tion of the parser than can be covered here; in this case, a reference such as [1] might be consulted; the ser-
vices of a local guru might also be appropriate.
There is one common situation where the rules given above for resolving conflicts are not sufficient;
this is in the area of arithmetic expressions. Most of the commonly used constructions for arithmetic
expressions can be naturally described by the notion of precedence levels for operators, together with infor-
mation about left or right associativity. It turns out that ambiguous grammars with appropriate disam-
biguating rules can be used to create parsers which are faster and easier to write than parsers constructed
from unambiguous grammars. The basic notion is to write grammar rules of the form
 - 11 -

expr : expr OP expr

and
expr : UNARY expr
for all binary and unary operators desired. This creates a very ambiguous grammar, with many parsing
conflicts. As disambiguating rules, the user specifies the precedence, or binding strength, of all the opera-
tors, and the associativity of the binary operators. This information is sufficient to allow Yacc to resolve the
parsing conflicts in accordance with these rules, and construct a parser which realizes the desired prece-
dences and associativities.
The precedences and associativities are attached to tokens in the declarations section. This is done
by a series of lines beginning with a Yacc keyword: %left, %right, or %nonassoc, followed by a list of
tokens. All of the tokens on the same line are assumed to have the same precedence level and associativity;
the lines are listed in order of increasing precedence or binding strength. Thus,
%left ´+´ ´−´
%left ´*´ ´/´
describes the precedence and associativity of the four arithmetic operators. Plus and minus are left associa-
tive, and have lower precedence than star and slash, which are also left associative. The keyword %right is
used to describe right associative operators, and the keyword %nonassoc is used to describe operators, like
the operator .LT. in Fortran, which may not associate with themselves; thus,
A .LT. B .LT. C
is illegal in Fortran, and such an operator would be described with the keyword %nonassoc in Yacc. As an
example of the behavior of these declarations, the description
%right ´=´
%left ´+´ ´−´
%left ´*´ ´/´

%%

expr :
expr ´=´ expr |
expr ´+´ expr |
expr ´−´ expr |
expr ´*´ expr |
expr ´/´ expr |
NAME ;
might be used to structure the input
a = b = c*d − e − f*g
as follows:
a = ( b = ( ((c*d)−e) − (f*g) ) )
When this mechanism is used, unary operators must, in general, be given a precedence. An interesting situ-
ation arises when we have a unary operator and a binary operator which have the same symbolic represen-
tation, but different precedences. An example is unary and binary ´−´; frequently, unary minus is given the
same strength as multiplication, or even higher, while binary minus has a lower strength than multiplica-
tion. We can indicate this situation by use of another keyword, %prec, to change the precedence level asso-
ciated with a particular grammar rule. %prec appears immediately after the body of the grammar rule,
before the action or closing semicolon, and is followed by a token name or literal; it causes the precedence
of the grammar rule to become that of the token name or literal. Thus, to make unary minus have the same
precedence as multiplication, we might write:
 - 12 -

%left ´+´ ´−´

%left ´*´ ´/´

%%

expr :
expr ´+´ expr |
expr ´−´ expr |
expr ´*´ expr |
expr ´/´ expr |
´−´ expr %prec ´*´ |
NAME ;

Notice that the precedences which are described by %left, %right, and %nonassoc are independent of
the declarations of token names by %token. A symbol can be declared by %token, and, later in the declara-
tions section, be given a precedence and associativity by one of the above methods. It is true, however, that
names which are given a precedence or associativity are also declared to be token names, and so in general
do not need to be declared by %token, although it does not hurt to do so.
As we mentioned above, the precedences and associativities are used by Yacc to resolve parsing con-
flicts; they give rise to disambiguating rules. Formally, the rules work as follows:
1. The precedences and associativities are recorded for those tokens and literals which have them.
2. A precedence and associativity is associated with each grammar rule; it is the precedence and asso-
ciativity of the last token or literal in the body of the rule. If the %prec construction is used, it over-
rides this default. Notice that some grammar rules may have no precedence and associativity associ-
ated with them.
3. When there is a reduce/reduce conflict, or there is a shift/reduce conflict and either the input symbol
or the grammar rule, or both, has no precedence and associativity associated with it, then the two dis-
ambiguating rules given at the beginning of the section are used, and the conflicts are reported.
4. If there is a shift/reduce conflict, and both the grammar rule and the input character have precedence
and associativity associated with them, then the conflict is resolved in favor of the action (shift or
reduce) associated with the higher precedence. If the precedences are the same, then the associativity
is used; left associative implies reduce, right associative implies shift, and nonassociating implies
error.
There are a number of points worth making about this use of disambiguation. There is no reporting
of conflicts which are resolved by this mechanism, and these conflicts are not counted in the number of
shift/reduce and reduce/reduce conflicts found in the grammar. This means that occasionally mistakes in
the specification of precedences disguise errors in the input grammar; it is a good idea to be sparing with
precedences, and use them in an essentially ‘‘cookbook’’ fashion, until some experience has been gained.
Frequently, not enough operators or precedences have been specified; this leads to a number of messages
about shift/reduce or reduce/reduce conflicts. The cure is usually to specify more precedences, or use the
%prec mechanism, or both. It is generally good to examine the verbose output file to ensure that the con-
flicts which are being reported can be validly resolved by precedence.

Section 5: Error Handling

Error handling is an extremely difficult area, and many of the problems are semantic ones. When an
error is found, for example, it may be necessary to reclaim parse tree storage, delete or alter symbol table
entries, and, typically, set switches to avoid putting out any further output.
It is generally not acceptable to stop all processing when an error is found; we wish to continue scan-
ning the input to find any further syntax errors. This leads to the problem of getting the parser ‘‘restarted’’
after an error. The general class of algorithms to do this involves reading ahead and discarding a number of
tokens from the input string, and attempting to adjust the parser so that input can continue.
 - 13 -

To allow the user some control over this process, Yacc provides a simple, but reasonably general, fea-
ture. The token name ‘‘error’’ is reserved for error handling. This name can be used in grammar rules; in
effect, it suggests places where errors are expected, and recovery might take place. The parser attempts to
find the last time in the input when the special token ‘‘error’’ is permitted. The parser then behaves as
though it saw the token name ‘‘error’’ as an input token, and attempts to parse according to the rule encoun-
tered. The token at which the error was detected remains the next input token after this error token is pro-
cessed. If no special error rules have been specified, the processing effectively halts when an error is
detected.
In order to prevent a cascade of error messages, the parser assumes that, after detecting an error, it
remains in error state until three tokens have been successfully read and shifted. If an error is detected
when the parser is already in error state, no error message is given, and the input token is quietly deleted.
As a common example, the user might include a rule of the form
statement : error ;
in his specification. This would, in effect, mean that on a syntax error the parser would attempt to skip over
the statement in which the error was seen. (Notice, however, that it may be difficult or impossible to tell the
end of a statement, depending on the other grammar rules). More precisely, the parser will scan ahead,
looking for three tokens that might legally follow a statement, and start processing at the first of these; if the
beginnings of statements are not sufficiently distinctive, it may make a false start in the middle of a state-
ment, and end up reporting a second error where there is in fact no error.
The user may supply actions after these special grammar rules, just as after the other grammar rules.
These actions might attempt to reinitialize tables, reclaim symbol table space, etc.
The above form of grammar rule is very general, but somewhat difficult to control. Somewhat easier
to deal with are rules of the form
statement : error ´;´ ;
Here, when there is an error, the parser will again attempt to skip over the statement, but in this case will do
so by skipping to the next ‘‘;’’. All tokens after the error and before the next ‘‘;’’ give syntax errors, and are
discarded. When the ‘‘;’’ is seen, this rule will be reduced, and any ‘‘cleanup’’ action associated with it will
be performed.
Still another form of error rule arises in interactive applications, where we may wish to prompt the
user who has incorrectly input a line, and allow him to reenter the line. In C we might write:
inputline: error ´\n´ prompt inputline
= { $$ = $4; };

prompt: /* matches no input */

= { printf( "Reenter last line: " ); };
There is one difficulty with this approach; the parser must correctly process three input tokens before it is
prepared to admit that it has correctly resynchronized after the error. Thus, if the reentered line contains
errors in the first two tokens, the parser will simply delete the offending tokens, and give no message; this is
clearly unacceptable. For this reason, there is a mechanism in both C and Ratfor which can be used to force
the parser to believe that resynchronization has taken place. One need only include a statement of the form
yyerrok ;
in his action after such a grammar rule, and the desired effect will take place; this name will be expanded,
using the ‘‘# define’’ mechanism of C or the ‘‘define’’ mechanism of Ratfor, into an appropriate code
sequence. For example, in the situation discussed above where we want to prompt the user to produce
input, we probably want to consider that the original error has been recovered when we have thrown away
the previous line, including the newline. In this case, we can reset the error state before putting out the
prompt message. The grammar rule for the nonterminal symbol prompt becomes:
 - 14 -

prompt: /* matches no input */

={
yyerrok;
printf( "Reenter last line: " );
};

There is another special feature which the user may wish to use in error recovery. As mentioned
above, the token seen immediately after the ‘‘error’’ symbol is the input token at which the error was dis-
covered. Sometimes, this is seen to be inappropriate; for example, an error recovery action might take upon
itself the job of finding the correct place to resume input. In this case, the user wishes a way of clearing the
previous input token held in the parser. One need only include a statement of the form
yyclearin ;
in his action; again, this expands, in both C and Ratfor, to the appropriate code sequence. For example,
suppose the action after error were to call some sophisticated resynchronization routine, supplied by the
user, which attempted to advance the input to the beginning of the next valid statement. After this routine
was called, the next token returned by yylex would presumably be the first token in a legal statement; we
wish to throw away the old, illegal token, and reset the error state. We might do this by the sequence:
statement : error
={
resynch( );
yyerrok ;
yyclearin ;
};

These mechanisms are admittedly crude, but do allow for a simple, fairly effective recovery of the
parser from many errors, and have the virtue that the user can get ‘‘handles’’ by which he can deal with the
error actions required by the lexical and output portions of the system.

Section 6C: The C Language Yacc Environment

The default mode of operation in Yacc is to write actions and the lexical analyzer in C. This has a
number of advantages; primarily, it is easier to write character handling routines, such as the lexical ana-
lyzer, in a language which supports character-by-character I/O, and has shifting and masking operators.
When the user inputs a specification to Yacc, the output is a file of C programs, called ‘‘y.tab.c’’.
These are then compiled, and loaded with a library; the library has default versions of a number of useful
routines. This section discusses these routines, and how the user can write his own routines if desired. The
name of the Yacc library is system dependent; see Appendix B.
The subroutine produced by Yacc is called ‘‘yyparse’’; it is an integer valued function. When it is
called, it in turn repeatedly calls ‘‘yylex’’, the lexical analyzer supplied by the user (see Section 3), to
obtain input tokens. Eventually, either an error is detected, in which case (if no error recovery is possible)
yyparse returns the value 1, or the lexical analyzer returns the endmarker token (type number 0), and the
parser accepts. In this case, yyparse returns the value 0.
Three of the routines on the Yacc library are concerned with the ‘‘external’’ environment of yyparse.
There is a default ‘‘main’’ program, a default ‘‘initialization’’ routine, and a default ‘‘accept’’ routine,
respectively. They are so simple that they will be given here in their entirety:
 - 15 -

main( argc, argv )

int argc;
char *argv[ ]
{
yyinit( argc, argv );
if( yyparse( ) )
return;
yyaccpt( );
}

yyinit( ) { }

yyaccpt( ) { }
By supplying his own versions of yyinit and/or yyaccpt, the user can get control either before the parser is
called (to set options, open input files, etc.) or after the accept action has been done (to close files, call the
next pass of the compiler, etc.). Note that yyinit is called with the two ‘‘command line’’ arguments which
have been passed into the main program. If neither of these routines is redefined, the default situation sim-
ply looks like a call to the parser, followed by the termination of the program. Of course, in many cases the
user will wish to supply his own main program; for example, this is necessary if the parser is to be called
more than once.
The other major routine on the library is called ‘‘yyerror’’; its main purpose is to write out a message
when a syntax error is detected. It has a number of hooks and handles which attempt to make this error
message general and easy to understand. This routine is somewhat more complex, but still approachable:
extern int yyline; /* input line number */

yyerror(s)
char *s;
{
extern int yychar;
extern char *yysterm[ ];

printf("\n%s", s );
if( yyline )
printf(", line %d,", yyline );
printf(" on input: ");
if( yychar >= 0400 )
printf("%s\n", yysterm[yychar−0400] );
else switch ( yychar ) {
case ´\t´: printf( "\\t\n" ); return;
case ´\n´: printf( "\\n\n" ); return;
case ´\0´: printf( "$end\n" ); return;
default: printf( "%c\n" , yychar ); return;
}
}
The argument to yyerror is a string containing an error message; most usually, it is ‘‘syntax error’’. yyerror
also uses the external variables yyline, yychar, and yysterm. yyline is a line number which, if set by the
user to a nonzero number, will be printed out as part of the error message. yychar is a variable which con-
tains the type number of the current token. yysterm has the names, supplied by the user, for all the tokens
which have names. Thus, the routine spends most of its time trying to print out a reasonable name for the
input token. The biggest problem with the routine as given is that, on Unix, the error message does not go
out on the error file (file 2). This is hard to arrange in such a way that it works with both the portable I/O
library and the system I/O library; if a way can be worked out, the routine will be changed to do this.
 - 16 -

Beware: This routine will not work if any token names have been given redefined type numbers. In this
case, the user must supply his own yyerror routine. Hopefully, this ‘‘feature’’ will disappear soon.
Finally, there is another feature which the C user of Yacc might wish to use. The integer variable
yydebug is normally set to 0. If it is set to 1, the parser will output a verbose description of its actions,
including a discussion of which input symbols have been read, and what the parser actions are. Depending
on the operating environment, it may be possible to set this variable by using a debugging system.

Section 6R: The Ratfor Language Yacc Environment

For reasons of portability or compatibility with existing software, it may be desired to use Yacc to
generate parsers in Ratfor, or, by extension, in portable Fortran. The user is likely to work considerably
harder doing this than he might if he were to use C.
When the user inputs a specification to Yacc, and specifies the Ratfor option (see Appendix B), the
output is a file of Ratfor programs called ‘‘y.tab.r’’. These programs are then compiled, and provide the
desired subroutine.
The subroutine produced by Yacc which does the input process is an integer function called
‘‘yypars’’. When it is called, it in turn repeatedly calls ‘‘yylex’’, the lexical analyzer supplied by the user
(see Section 3). Eventually, either an error is detected, in which case (if no error recovery is possible)
yypars returns the value 1, or the lexical analyzer returns the endmarker (type number 0), and the parser
accepts. In this case, yypars returns 0.
Unlike the C program situation (see Section 6C) there is no library of Ratfor routines which must be
used in the loading process. As a side effect of this, the user must supply a main program which calls
yypars. A suggested Ratfor main program is
integer yypars
n = yypars(0)
if( n .EQ. 0 ) {
. . . here if the program accepted
} else {
. . . here if there were unrecoverable errors
}
end
Notice that there is no easy way for the user to get control when an error is detected, since the Fortran lan-
guage provides only a very crude character string capability.
There is another feature which the Ratfor user might wish to use. The argument to yypars is nor-
mally 0. If it is set to 1, the parser will output a verbose description of its actions, including a discussion of
which input symbols have been read, and what the parser actions are. During the input process, the value
of this debug flag is kept in a common variable yydebu, which is available to the actions and may be set and
reset at will.
Statement labels 1 through 1000 are reserved for the parser, and may not appear in actions; note that,
because Ratfor has a more modern control structure than Fortran, it is rarely necessary to use statement
labels at all; the most frequent use of labels in Ratfor is in formatted I/O.
Because Fortran has no standard character set and not even a standard character width, it is difficult
to produce a lexical analyzer in portable Fortran The usual solution is to provide a routine which does a
table search to get the internal type number for each input character, with the understanding that such a rou-
tine can be recoded to run far faster for any particular machine.
Finally, we must warn the user that the Ratfor feature of Yacc has been operational for a much shorter
time than the other portions of the system. If past experience is any guide, the Ratfor support will develop
and become more powerful and better human engineered in response to user complaints and requirements.
Thus, the potential Ratfor user might do well to contact the author to discuss his own particular needs.
 - 17 -

Section 7. Hints for Preparing Specifications

This section contains miscellaneous hints on preparing efficient, easy to change, and clear specifica-
tions. The individual subsections are, more or less, independent; the reader seeing Yacc for the first time
may well find that this entire section could be omitted.

Input Style
It is difficult to input rules with substantial actions and still have a readable specification file. The
following style hints owe much to Brian Kernighan, and are officially endorsed by the author.
a. Use all capital letters for token names, all lower case letters for nonterminal names. This rule comes
under the heading of ‘‘knowing who to blame when things go wrong.‘‘
b. Put grammar rules and actions on separate lines. This allows either to be changed without an auto-
matic need to change the other.
c. Put all rules with the same left hand side together. Put the left hand side in only once, and let all fol-
lowing rules begin with a vertical bar.
d. Indent rule bodies by one tab stop, and action bodies by two tab stops.
The example in Appendix A is written following this style, as are the examples in the text of this
paper (where space permits). The user must make up his own mind about these stylistic questions; the cen-
tral problem, however, is to make the rules visible through the morass of action code.

Common Actions
When several grammar rules have the same action, the user might well wish to provide only one code
sequence. A simple, general mechanism is, of course, to use subroutine calls. It is also possible to put a
label on the first statement of an action, and let other actions be simply a goto to this label. Thus, if the user
had a routine which built trees, he might wish to have only one call to it, as follows:
expr :
expr ´+´ expr =
{ binary:
$$ = btree( $1, $2, $3 );
}|
expr ´−´ expr =
{
goto binary;
}|
expr ´*´ expr =
{
goto binary;
};

Left Recursion
The algorithm used by the Yacc parser encourages so called ‘‘left recursive’’ grammar rules: rules of
the form
name : name rest of rule ;
These rules frequently arise when writing specifications of sequences and lists:
list :
item |
list ´,´ item ;
and
 - 18 -

sequence :
item |
sequence item ;
Notice that, in each of these cases, the first rule will be reduced for the first item only, and the second rule
will be reduced for the second and all succeeding items.
If the user were to write these rules right recursively, such as
sequence :
item |
item sequence ;
the parser would be a bit bigger, and the items would be seen, and reduced, from right to left. More seri-
ously, an internal stack in the parser would be in danger of overflowing if a very long sequence were read.
Thus, the user should use left recursion wherever reasonable.
The user should also consider whether a sequence with zero elements has any meaning, and if so,
consider writing the sequence specification with an empty rule:
sequence :
| /* empty */
sequence item ;
Once again, the first rule would always be reduced exactly once, before the first item was read, and then the
second rule would be reduced once for each item read. Experience suggests that permitting empty
sequences leads to increased generality, which frequently is not evident at the time the rule is first written.
There are cases, however, when the Yacc algorithm can fail when such a change is made. In effect, con-
flicts might arise when Yacc is asked to decide which empty sequence it has seen, when it hasn´t seen
enough to know! Nevertheless, this principle is still worth following wherever possible.

Lexical Tie-ins
Frequently, there are lexical decisions which depend on the presence of various constructions in the
specification. For example, the lexical analyzer might want to delete blanks normally, but not within
quoted strings. Or names might be entered into a symbol table in declarations, but not in expressions.
One way of handling these situations is to create a global flag which is examined by the lexical ana-
lyzer, and set by actions. For example, consider a situation where we have a program which consists of 0 or
more declarations, followed by 0 or more statements. We declare a flag called ‘‘dflag’’, which is 1 during
declarations, and 0 during statements. We may do this as follows:
 - 19 -

%{
int dflag ;
%}
%%
program :
decls stats ;

decls :
= /* empty */
{
dflag = 1;
}|
decls declaration ;

stats :
= /* empty */
{
dflag = 0;
}|
stats statement ;

. . . other rules . . .
The flag dflag is now set to zero when reading statements, and 1 when reading declarations, except for the
first token in the first statement. This token must be seen by the parser before it can tell that the declaration
section has ended and the statements have begun. Frequently, however, this single token exception does not
affect the lexical scan required.
Clearly, this kind of ‘‘backdoor’’ approach can be elaborated on to a noxious degree. Nevertheless, it
represents a way of doing some things that are difficult, if not impossible, to do otherwise.

Bundling
Bundling is a technique for collecting together various character strings so that they can be output at
some later time. It is derived from a feature of the same name in the compiler/compiler TMG [6].
Bundling has two components − a nice user interface, and a clever implementation trick. They will
be discussed in that order.
The user interface consists of two routines, ‘‘bundle’’ and ‘‘bprint’’.
bundle( a1, a2, . . ., an )
accepts a variable number of arguments which are either character strings or bundles, and returns a bundle,
whose value will be the concatenation of the values of a1, . . ., an.
bprint( b )
accepts a bundle as argument and outputs its value.
For example, suppose that we wish to read arithmetic expressions, and output function calls to rou-
tines called ‘‘add’’, ‘‘sub’’, ‘‘mul’’, ‘‘div’’, and ‘‘assign’’. Thus, we wish to translate
a = b − c*d
into
assign(a,sub(b,mul(c,d)))

A Yacc specification file which does this is given in Appendix D; this includes an implementation of
the bundle and bprint routines. A rule and action of the form
 - 20 -

expr:
expr ´+´ expr =
{
$$ = bundle( "add(", $1, ",", $3, ")" );
}
causes the returned value of expr to be come a bundle, whose value is the character string containing the
desired function call. Each NAME token has a value which is a pointer to the actual name which has been
read. Finally, when the entire input line has been read and the value has been bundled, the value is written
out and the bundles and names are cleared, in preparation for the next input line.
Bundles are implemented as arrays of pointers, terminated by a zero pointer. Each pointer either
points to a bundle or to a character string. There is an array, called bundle space, which contains all the
bundles.
The implementation trick is to check the values of the pointers in bundles − if the pointer points into
bundle space, it is assumed to point to a bundle; otherwise it is assumed to point to a character string.
The treatment of functions with a variable number of arguments, like bundle, is likely to differ from
one implementation of C to another.
In general, one may wish to have a simple storage allocator which allocates and frees bundles, in
order to handle situations where it is not appropriate to completely clear all of bundle space at one time.

Reserved Words
Some programming languages permit the user to use words like ‘‘if ’’, which are normally reserved,
as label or variable names, provided that such use does not conflict with the legal use of these names in the
programming language. This is extremely hard to do in the framework of Yacc, since it is difficult to pass
the required information to the lexical analyzer which tells it ‘‘this instance of if is a keyword, and that
instance is a variable’’. The user can make a stab at it, using the mechanism described in the last subsec-
tion, but it is difficult.
A number of ways of making this easier are under advisement, and one will probably be supported
eventually. Until this day comes, I suggest that the keywords be reserved; that is, be forbidden for use as
variable names. There are powerful stylistic reasons for preferring this, anyway (he said weakly . . . ).

Non-integer Values
Frequently, the user wishes to have values which are bigger than integers; again, this is an area where
Yacc does not make the job as easy as it might, and some additional support is likely. Nevertheless, at the
cost of writing a storage manager, the user can return pointers or indices to blocks of storage big enough to
contain the full values desired.

Previous Work
There have been many previous applications of Yacc. The user who is contemplating a big applica-
tion might well find that others have developed relevant techniques, or even portions of grammars. Yacc
specifications appear to be easier to change than the equivalent computer programs, so that the ‘‘prior art’’
is more relevant here as well.

Section 8: User Experience, Summary, and Acknowledgements

Yacc has been used in the construction of a C compiler for the Honeywell 6000, a system for typeset-
ting mathematical equations, a low level implementation language for the PDP 11, APL and Basic compil-
ers to run under the UNIX system, and a number of other applications.
To summarize, Yacc can be used to construct parsers; these parsers can interact in a fairly flexible
way with the lexical analysis and output phases of a larger system. The system also provides an indication
of ambiguities in the specification, and allows disambiguating rules to be supplied to resolve these ambigui-
ties.
 - 21 -

Because the output of Yacc is largely tables, the system is relatively language independent. In the
presence of reasonable applications, Yacc could be modified or adapted to produce subroutines for other
machines and languages. In addition, we continue to seek better algorithms to improve the lexical analysis
and code generation phases of compilers produced using Yacc.
This document would be incomplete if I did not give credit to a most stimulating collection of users,
who have goaded me beyond my inclination, and frequently beyond my ability, in their endless search for
‘‘one more feature’’. Their irritating unwillingness to learn how to do things my way has usually led to my
doing things their way; most of the time, they have been right. B. W. Kernighan, P. J. Plauger, S. I. Feld-
man, C. Imagna, M. E. Lesk, and A. Snyder will recognize some of their ideas in the current version of
Yacc. Al Aho also deserves recognition for bringing the mountain to Mohammed, and other favors.
 - 22 -

References
1 Aho, A.V. and Johnson, S.C., ‘‘LR Parsing’’, Computing Surveys, Vol 6, No 2, June 1974, pp.
99-124.
2 Aho, A.V., Johnson, S.C., and Ullman, J.D., ‘‘Deterministic Parsing of Ambiguous Grammars’’, Pro-
ceedings of the A.C.M. Symposium on Principles of Programming Languages, October 1973, pp.
1-21; to appear in CACM.
3 Aho, A.V. and Ullman, J.D., Theory of Parsing, Translation, and Compiling. Volume 1 (1972) and
Volume 2 (1973), Prentice-Hall, Englewood Cliffs, N.J.
4 Kernighan, B. W., Ratfor, a Rational Fortran
5 Ryder, B. B., ‘‘The PFORT Verifier,’’ Software−Practice and Experience, Vol 4 (1974), pp 359-377.
6 McIlroy, M. D., A Manual for the TMG Compiler-writing Language
7 Ritchie, D. M., C Reference Manual
 - 23 -

Appendix A: A Simple Example

This example gives the complete Yacc specification for a small desk calculator; the desk calculator
has 26 registers, labeled a through z, and accepts arithmetic expressions made up of the operators +, −, *, /,
% (mod operator), & (bitwise and), | (bitwise or), and assignment. If an expression is an assignment at the
top level, the value is not printed; otherwise it is. As in C, an integer which begins with 0 (zero) is assumed
to be octal; otherwise, it is assumed to be decimal.
As an example of a Yacc specification, the desk calculator does a reasonable job of showing the way
that precedences and ambiguities are used, as well as showing how simple error recovery operates. The
major oversimplifications are that the lexical analysis phase is much simpler than for most applications, and
the output is produced immediately, line by line. Note the way that decimal and octal integers are read in
by the grammar rules; frequently, this job is better done by the lexical analyzer.

%token DIGIT LETTER/* these are token names */

%left ´|´ /* declarations of operator precedences */
%left ´&´
%left ´+´ ´−´
%left ´*´ ´/´ ´%´
%left UMINUS /* supplies precedence for unary minus */
%{ /* declarations used by the actions */
int base;
int regs[26];
%}

%% /* beginning of rules section */

list : /* list is the start symbol */

| /* empty */
list stat ´\n´ |
list error ´\n´ =
{
yyerrok ;
};

stat :
expr =
{
printf("%d\n", $1) ;
}|
LETTER ´=´ expr =
{
regs[$1] = $3 ;
};

expr :
´(´ expr ´)´ =
{
$$ = $2 ;
}|
expr ´+´ expr =
{
$$ = $1 + $3 ;
}|
 - 24 -

expr ´−´ expr =

{
$$ = $1 − $3 ;
}|
expr ´*´ expr =
{
$$ = $1 * $3 ;
}|
expr ´/´ expr =
{
$$ = $1 / $3 ;
}|
expr ´%´ expr =
{
$$ = $1 % $3 ;
}|
expr ´&´ expr
{
$$ = $1 & $3 ;
}|
expr ´|´ expr
{
$$ = $1 | $3 ;
}|
´−´ expr %prec UMINUS
{
$$ = − $2 ;
}|
LETTER
{
$$ = regs[$1] ;
}|
number ;

number :
DIGIT =
{
$$ = $1 ;
base = 10 ;
if( $1 == 0 )
base = 8 ;
}|
number DIGIT =
{
$$ = base * $1 + $2 ;
};

%% /* start of programs */

yylex( ) /* lexaical analysis routine */

{
/* returns LETTER for a lower case letter, yylval = 0 through 25 */
/* return DIGIT for a digit, yylval = 0 through 9 */
/* all other characters are returned immediately */
 - 25 -

int c ;

while( (c=getchar( )) == ´ ´ )
;
if( c >= ´a´ && c <= ´z´ ) {
yylval = c − ´a´ ;
return( LETTER ) ;
}
if( c >= ´0´ && c <= ´9´ ) {
yylval = c − ´0´ ;
return( DIGIT ) ;
}
return( c ) ;
}
 - 26 -

Appendix B: Use of Yacc on Unix

Suppose that the Yacc specification is on a file called yfile. If the actions are in C, Yacc is invoked by
yacc yfile
The output appears on file y.tab.c To compile the parser and load it with the Yacc library, use the command
cc y.tab.c −ly
If Yacc is invoked with the option −v:
yacc −v yfile
a verbose description of the parser is produced on file y.output. The C user should consult section 6C for
more information about the run time environment.
If the actions are in Ratfor, the user should invoke Yacc with the option −r:
yacc −r yfile
The Ratfor output appears on file y.tab.r It may be compiled by
rc −2 y.tab.r
Note that when Yacc is used to produce Ratfor programs, there is no need to load these programs with any
library.
If the −v action is also invoked:
yacc −rv yfile
a verbose description of the parser is produced on file y.output. The Ratfor user should consult section 6R
for more information about the run time environment.

Appendix C: Old Features Supported but not Encouraged

This appendix mentions synonyms and features which are supported for historical continuity, but, for
various reasons, are not encouraged.
1. Literals may be delimited by double quotes ‘‘"’’ as well as single quotes ‘‘´’’.
2. Literals may be more than one character long. If all the characters are alphabetic, numeric, or , the
type number of the literal is defined, just as if the literal did not have the quotes around it. Otherwise,
it is difficult to find the value for such literals.
The use of multi-character literals is likely to mislead those unfamiliar with Yacc, since it suggests
that Yacc is doing a job which must be actually done by the lexical analyzer.
3. Most places where % is legal, backslash ‘‘\’’ may be used. In particular, \\ is the same as %%, \left
the same as %left, etc.
4. There are a number of other synonyms:
%< is the same as %left
%> is the same as %right
%binary and %2 are the same as %nonassoc
%0 and %term are the same as %token
%= is the same as %prec

5. The curly braces ‘‘{’’ and ‘‘}’’ around an action are optional if the action consists of a single C state-
ment. (They are always required in Ratfor).
 - 27 -

Appendix D: An Example of Bundling

The following program is an example of the technique of bundling; this example is discussed in Sec-
tion 7.

/* warnings:
1. This works on Unix; the handling of functions with a variable number of arguments is different on
different systems.
2. A number of checks for array bounds have been left out to avoid obscuring the basic ideas, but
should be there in a practical program.
*/

%token NAME

%right ´=´
%left ´+´ ´−´
%left ´*´ ´/´

%%

lines :
= /* empty */
{
bclear( ) ;
}|
lines expr ´\n´ =
{
bprint( $2 ) ;
printf( "\n" ) ;
bclear( ) ;
}|
lines error ´\n´ =
{
bclear( ) ;
yyerrok;
};

expr :
expr ´+´ expr =
{
$$ = bundle( "add(", $1, ",", $3, ")" );
}|
expr ´−´ expr =
{
$$ = bundle( "sub(", $1, ",", $3, ")" );
}|
expr ´*´ expr =
{
$$ = bundle( "mul(", $1, ",", $3, ")" );
}|
expr ´/´ expr =
 - 28 -

{
$$ = bundle( "div(", $1, ",", $3, ")" );
}|
´(´ expr ´)´ =
{
$$ = $2;
}|
NAME ´=´ expr =
$$ = bundle( "assign(", $1, ",", $3, ")" );
}|
NAME ;

%%

#define nsize 200

char names[nsize], *nptr { names };

#define bsize 500

int bspace[bsize], *bptr { bspace };

yylex( )
{
int c;

c = getchar( );
while( c == ´ ´ )
c = getchar( );
if( c>=´a´ && c<=´z´ ) {
yylval = nptr;
for( ; c>=´a´ && c<=´z´; c=getchar( ) )
*nptr++ = c;
ungetc( c );
*nptr++ = ´\0´;
return( NAME );
}
return( c );
}

bclear( )
{

nptr = names;
bptr = bspace;
}

bundle( a1,a2,a3,a4,a5 )
{
int i, j, *p, *obp;

p = &a1;
i = nargs( );
obp = bptr;
 - 29 -

for( j=0; j<i; ++j )

*bptr++ = *p++;
*bptr++ = 0;
return( obp );
}

bprint( p )
int *p;
{

if( p>=bspace && p< &bspace[bsize] ) /* bundle */

while( *p != 0 )
bprint( *p++ );
else printf( "%s", p );
}
-

The Unix I/O System

Dennis M. Ritchie
Bell Telephone Laboratories

This paper gives an overview of the workings of the Unix I/O system. It was written with an eye toward providing
guidance to writers of device driver routines, and is oriented more toward describing the environment and nature of
device drivers than the implementation of that part of the file system which deals with ordinary files.
It is assumed that the reader has a good knowledge of the overall structure of the file system as discussed in the pa-
per ‘‘The Unix Time-sharing System.’’ Moreover the present document is intended to be used in conjunction with a
copy of the system code, since it is basically an exegesis of that code.

Device Classes
There are two classes of device: block and character. The block interface is suitable for devices like disks, tapes,
and DECtape which work, or can work, with addressible 512-byte blocks. Ordinary magnetic tape just barely fits in
this category, since by use of forward backward spacing any block can be read, even though blocks can be written
only at the end of the tape. Block devices can at least potentially contain a mounted file system. The interface to
block devices is very highly structured; the drivers for these devices share a great many routines as well as a pool of
buffers.
Character-type devices have a much more straightforward interface, although more work must be done by the driver
itself.
Devices of both types are named by a major and a minor device number. These numbers are generally stored as a
word with the minor device number as the low byte and the major device number as the high byte. The major device
number selects which driver will deal with the device; the minor device number is not used by the rest of the system
but is passed to the driver at appropriate times. Typically the minor number selects a subdevice attached to a given
controller, or one of several similar hardware interfaces.
The major device numbers for block and character devices are used as indices in separate tables; they both start at 0
and therefore overlap.

Overview of I/O
The purpose of the open and creat system calls is to set up entries in three separate system tables. The first of these
is the u ofile table, which is stored in the system’s per-process data area u. This table is indexed by the file descrip-
tor returned by the open or creat, and is accessed during a read, write, or other operation on the open file. An entry
contains only a pointer to the corresponding entry of the file table, which is a per-system data base. There is one en-
try in the file table for each instance of open or creat. This table is per-system because the same instance of an open
file must be shared among the several processes which can result from forks after the file is opened. A file table en-
try contains flags which indicate whether the file was open for reading or writing or is a pipe, and a count which is
used to decide when all processes using the entry have terminated or closed the file (so the entry can be abandoned).
There is also a 32-bit file offset which is used to indicate where in the file the next read or write will take place. Fi-
nally, there is a pointer to the entry for the file in the inode table, which contains a copy of the file’s i-node. Notice
that an entry in the file table corresponds precisely to an instance of open or creat; if the same file is opened several
times, it will have several entries in this table. However, there is at most one entry in the inode table for a given file.
Also, a file may enter the inode table not only because it is open, but also because it is the current directory of some
process or because it is a special file containing a currently-mounted file system.
An entry in the inode table differs somewhat from the corresponding i-node as stored on the disk; the modified and
accessed times are not stored, and the entry is augmented by a flag word containing information about the entry, a
count used to determine when it may be allowed to disappear, and the device and i-number whence the entry came.
-

2 - Unix I/O System

During the processing of an open or creat call for a special file, the system always calls the device’s open routine to
allow for any special processing required (rewinding a tape, turning on the data-terminal-ready lead of a modem,
etc.). However, the close routine is called only when the last process closes a file, that is, when the i-node table en-
try is being deallocated. Thus it is not feasible for a device to maintain, or depend on, a count of its users, although
it is quite possible to implement an exclusive-use device which cannot be reopened until it has been closed.
When a read or write takes place, the user’s arguments and the file table entry are used to set up the variables
u.u base, u.u count, and u.u offset which respectively contain the (user) address of the I/O target area, the byte-
count for the transfer, and the current location in the file. If the file referred to is a character-type special file, the ap-
propriate read or write routine is called; it is responsible for transferring data and updating the count and current lo-
cation appropriately as discussed below. Otherwise, the current location is used to calculate a logical block number
in the file. If the file is an ordinary file the logical block number must be mapped (possibly using an indirect block)
to a physical block number; a block-type special file need not be mapped. In any event, the resulting physical block
number is used, as discussed below, to read or write the appropriate device.

Character device drivers

The cdevsw table specifies the interface routines present for character devices. Each device provides five routines:
open, close, read, write, and special-function. Any of these may be missing. If a call on the routine should be ig-
nored, (e.g. open on non-exclusive devices which require no setup) the cdevsw entry can be given as nulldev; if it
should be considered an error, (e.g. write on read-only devices) nodev is used.
The open routine is called each time the file is opened with the full device number as argument. The second argu-
ment is a flag which is non-zero only if the device is to be written upon.
The close routine is called only when the file is closed for the last time, that is when the very last process in which
the file is open closes it. This means it is not possible for the driver to maintain its own count of its users. The first
argument is the device number; the second is a flag which is non-zero if the file was open for writing in the process
which performs the final close.
When write is called, it is supplied the device as argument. The per-user variable u.u count has been set to the num-
ber of characters indicated by the user; for character devices, this number may be 0 initially. u.u base is the address
supplied by the user from which to start taking characters. The system may call the routine internally, so the flag
u.u segflg is supplied which indicates, if on, that u.u base refers to the system address space instead of the user’s.
The write routine should copy up to u.u count characters from the user’s buffer to the device, decrementing
u.u count for each character passed. For most drivers, which work one character at a time, the routine
cpass( )
is used to pick up characters from the user’s buffer. Successive calls on it return the characters to be written until
u.u count goes to 0 or an error occurs, when it returns −1. Cpass takes care of interrogating u.u segflg and updating
u.u count.
Write routines which want to transfer a probably large number of characters into an internal buffer may also use the
routine
iomove(buffer, offset, count, flag)
which is faster when many characters must be moved. Iomove transfers up to count characters into the buffer start-
ing offset bytes from the start of the buffer; flag should be B WRITE (which is 0) in the write case. Caution: the
caller is responsible for making sure the count is not too large and is non-zero. As an efficiency note, iomove is
much slower if any of buffer+offset, count or u.u base is odd.
The device’s read routine is called under conditions similar to write, except that u.u count is guaranteed to be non-
zero. To return characters to the user, the routine
passc(c)
is available; it takes care of housekeeping like cpass and returns −1 as the last character specified by u.u count is re-
turned to the user; before that time, 0 is returned. Iomove is also usable as with write; the flag should be B READ
but the same cautions apply.
The ‘‘special-functions’’ routine is invoked by the stty and gtty system calls as follows:
-

Unix I/O System - 3

(*p) (dev, v)
where p is a pointer to the device’s routine, dev is the device number, and v is a vector. In the gtty case, the device is
supposed to place up to 3 words of status information into the vector; this will be returned to the caller. In the stty
case, v is 0; the device should take up to 3 words of control information from the array u.u arg[0...2].
Finally, each device should have appropriate interrupt-time routines. When an interrupt occurs, it is turned into a C-
compatible call on the devices’s interrupt routine. The interrupt-catching mechanism makes the low-order four bits
of the ‘‘new PS’’ word in the trap vector for the interrupt available to the interrupt handler. This is conventionally
used by drivers which deal with multiple similar devices to encode the minor device number. After the interrupt has
been processed, a return from the interrupt handler will return from the interrupt itself.
A number of subroutines are available which are useful to character device drivers. Most of these handlers, for ex-
ample, need a place to buffer characters in the internal interface between their ‘‘top half’’ (read/write) and ‘‘bottom
half ’’ (interrupt) routines. For relatively low data-rate devices, the best mechanism is the character queue main-
tained by the routines getc and putc. A queue header has the structure
struct {
int c cc; /* character count */
char *c cf; /* first character */
char *c cl; /* last character */
} queue;
A character is placed on the end of a queue by
putc(c, &queue)
where c is the character and queue is the queue header. The routine returns −1 if there is no space to put the charac-
ter, 0 otherwise. The first character on the queue may be retrieved by
getc(&queue)
which returns either the (non-negative) character or −1 if the queue is empty.
Notice that the space for characters in queues is shared among all devices in the system and in the standard system
there are only some 600 character slots available. Thus device handlers, especially write routines, must take care to
avoid gobbling up excessive numbers of characters.
The other major help available to device handlers is the sleep-wakeup mechanism. The call
sleep(event, priority)
causes the process to wait (allowing other processes to run) until the event occurs; at that time, the process is marked
ready-to-run and the call will return when there is no process with higher priority.
The call
wakeup(event)
indicates that the event has happened, that is, causes processes sleeping on the event to be awakened. The event is an
arbitrary quantity agreed upon by the sleeper and the waker-up. By convention, it is the address of some data area
used by the driver, which guarantees that events are unique.
Processes sleeping on an event should not assume that the event has really happened; they should check that the con-
ditions which caused them to sleep no longer hold.
Priorities can range from 127 to −127; a higher numerical value indicates a less-favored scheduling situation. A pro-
cess sleeping at negative priority cannot be terminated for any reason, although it is conceivable that it may be
swapped out. Thus it is a bad idea to sleep with negative priority on an event which might never occur. On the other
hand, calls to sleep with non-negative priority may never return if the process is terminated by some signal in the
meantime. Incidentally, it is a gross error to call sleep in a routine called at interrupt time, since the process which is
running is almost certainly not the process which should go to sleep. Likewise, none of the variables in the user area
‘‘u.’’ should be touched, let alone changed, by an interrupt routine.
If a device driver wishes to wait for some event for which it is inconvenient or impossible to supply a wakeup, (for
example, a device going on-line, which does not generally cause an interrupt), the call
-

4 - Unix I/O System

sleep(&lbolt, priority)
may be given. Lbolt is an external cell whose address is awakened once every 4 seconds by the clock interrupt rou-
tine.
The routines
spl4( ), spl5( ), spl6( ), spl7( )
are available to set the processor priority level as indicated to avoid inconvenient interrupts from the device.
If a device needs to know about real-time intervals, then
timeout(func, arg, interval)
will be useful. This routine arranges that after interval sixtieths of a second, the func will be called with arg as argu-
ment, in the style
(*func)(arg)
Timeouts are used, for example, to provide real-time delays after function characters like new-line and tab in type-
writer output, and to terminate an attempt to read the 201 Dataphone dp if there is no response within a specified
number of seconds. Notice that the number of sixtieths of a second is limited to 32767, since it must appear to be
positive, and that only a bounded number of timeouts can be going on at once. Also, the specified func is called at
clock-interrupt time, so it should conform to the requirements of interrupt routines in general.

An example
The driver for the paper-tape reader/punch is worth examining as a fairly simple example of many of the techniques
used in writing character device handlers. The pc11 structure contains a state (used for the reader), an input queue,
and an output queue. A structure, rather than three individual variables, was used to cut down on the number of ex-
ternal symbols which might be confused with symbols in other routines.
When the file is opened for reading, the open routine checks to see if its state is not CLOSED; if so an error is re-
turned since it is considered a bad idea to let several people read one tape simultaneously. The state is set to WAIT-
ING, the interrupt is enabled, and a character is requested. The reason for this gambit is that there is no direct way
to determine if there is any tape in the reader or if the reader is on-line. In these situations an interrupt will occur
immediately and an error indicated. As will be seen, the interrupt routine ignores errors if the state is WAITING, but
if a good character comes in while in the WAITING state the interrupt routine sets the state to READING. Thus open
loops until the state changes, meanwhile sleeping on the ‘‘lightning bolt’’ lbolt. If it did not sleep at all, it would
prevent any other process from running until the reader came on-line; if it depended on the interrupt routine to wake
it up, the effect would be the same, since the error interrupt is almost instantaneous.
The open-write case is much simpler; the punch is enabled and a 100-character leader is punched using pcleader.
The close routine is also simple; if the reader was open, any uncollected characters are flushed, the interrupt is
turned off, and the state is set to CLOSED. In the write case a 100-character trailer is punched. The routine has a
bug in that if both the reader and punch are open close will be called only once, so that either the leftover characters
are flushed or the trailer is punched, but not both. It is hard to see how to fix this problem except by making the
reader and punch separate devices.
The pcread routine tries to pick up characters from the input queue and passes them back until the user’s read call is
satisfied. If there are no characters it checks whether the state has gone to EOF, which means that the interrupt rou-
tine detected an error in the READ state (assumed to indicate the end of the tape). If so, pcread returns; either during
this call or the next one no characters will be passed back, indicating an end-of-file. If the state is still READING the
routine enables another character by fiddling the device’s reader control register, provided it is not active, and goes to
sleep.
When a reader interrupt occurs and the state is WAITING, and the device’s error bit is set, the interrupt is ignored; if
there is no error the state is set to READING, as indicated in the discussion of pcread. If the state is READING and
there is an error, the state is set to EOF; it is assumed that the error represents the end of the tape. If there is no er-
ror, the character is picked up and stored in the input queue. Then, provided the number of characters already in the
queue is less than the high-water mark PCIHWAT, the reader is enabled again to read another character. This strate-
gy keeps the tape moving without flooding the input queue with unread characters. Finally, the top half is awakened.
-

Unix I/O System - 5

Looking again at pcread, notice that the priority level is raised by spl4( ) to prevent interrupts during the loop. This
is done because of the possibility that the input queue is empty, and just after the EOF test is made an error interrupt
occurs because the tape runs out. Then sleep will be called with no possibility of a wakeup. In general the processor
priority should be raised when a routine is about to sleep awaiting some condition where the presence of the condi-
tion, and the consequent wakeup, is indicated by an interrupt. The danger is that the interrupt might occur between
the test for the condition and the call to sleep, so that the wakeup apparently never happens.
At the same time it is a bad idea to raise the processor priority level for an extended period of time, since devices
with real-time requirements may be shut out so long as to cause an error. The pcread routine is perhaps overzealous
in this respect, although since most devices have a priority level higher than 4 this difficulty is not very important.
The pcwrite routine simply gets characters from the user and passes them to pcoutput, which is separated out so that
pcleader can call it also. Pcoutput checks for errors (like out-of-tape) and if none are present makes sure that the
number of characters in the output queue does not exceed PCOHWAT; if it does, sleep is called. Then the character
is placed on the output queue. There is a small bug here in that pcoutput does not check that the character was suc-
cessfully put on the queue (all character-queue space might be empty); perhaps in this case it might be a good idea to
sleep on the lightning-bolt until things quiet down. Finally pcstart is called, which checks to see if the punch is cur-
rently busy, and if not starts the punching of the first character on the output queue.
When punch interrupts occur, pcpint is called; it starts the punching of the next character on the output queue, and if
the number of characters remaining on the queue is less than the low-water mark PCOLWAT it wakes up the write
routine, which is presumably waiting for the queue to empty.

The Block-device Interface

Handling of block devices is mediated by a collection of routines which manage a set of buffers containing the im-
ages of blocks of data on the various devices. The most important purpose of these routines is to assure that several
processes which access the same block of the same device in multiprogrammed fashion maintain a consistent view
of the data in the block. A secondary but still important purpose is to increase the efficiency of the system by keep-
ing in-core copies of blocks which are being accessed frequently. The main data base for this mechanism is the table
of buffers buf. Each buffer header contains a pair of pointers (b forw, b back) which maintain a doubly-linked list
of the buffers associated with a particular block device, and a pair of pointers (av forw, av back) which generally
maintain a doubly-linked list of blocks which are ‘‘free,’’ that is, eligible to be reallocated for another transaction.
Buffers which have I/O in progress or are busy for other purposes do not appear in this list. The buffer header also
contains the device and block number to which the buffer refers, and a pointer to the actual storage associated with
the buffer. There is a word count which is the negative of the number of words to be transferred to or from the
buffer; there is also an error byte and a residual word count used to communicate information from an I/O routine to
its caller. Finally, there is a flag word with bits indicating the status of the buffer. These flags will be discussed be-
low.
Six routines constitute the most important part of the interface with the rest of the system. Given a device and block
number, both bread and getblk return a pointer to a buffer header for the block; the difference is that bread is guaran-
teed to return a buffer actually containing the current data for the block, while getblk returns a buffer which contains
the data in the block only if it is already in core (whether it is or not is indicated by the B DONE bit; see below). In
either case the buffer, and the corresponding device block, is made ‘‘busy,’’ so that other processes referring to it are
obliged to wait until it becomes free. Getblk is used, for example, when a block is about to be totally rewritten, so
that its previous contents are not useful; still, no other process can be allowed to refer to the block until the new data
is placed into it.
Given a pointer to a buffer, the brelse routine makes the buffer again available to other processes. It is called, for ex-
ample, after data has been extracted following a bread. There are three subtly-different write routines, all of which
take a buffer pointer as argument, and all of which logically release the buffer for use by others and place it on the
free list. Bwrite puts the buffer on the appropriate device queue, waits for the write to be done, and sets the user’s er-
ror flag if required.

Bawrite places the buffer on the device’s queue, but does not wait for completion, so that errors cannot be reflected
directly to the user. Bdwrite does not start any I/O operation at all, but merely marks the buffer so that if it happens
to be grabbed from the free list to contain data from some other block, the data in it will first be written out.
Bwrite is used when one wants to be sure that I/O takes place correctly, and that errors are reflected to the proper us-
er; it is used, for example, when updating i-nodes. Bawrite is useful when more efficiency is desired (because no
-

6 - Unix I/O System

wait is required for I/O to finish) but when it is reasonably certain that the write is really required. Bdwrite is used
when there is doubt that the write is needed at the moment. For example, bdwrite is called when the last byte of a
write system call falls short of the end of a block, on the assumption that another write will be given soon which will
re-use the same block. On the other hand, as the end of a block is passed, bawrite is called, since probably the block
will not be accessed again soon and one might as well start the writing process as soon as possible.
In any event, notice that the routines getblk and bread dedicate the given block exclusively to the use of the caller,
and make others wait, while one of brelse, bwrite, bawrite, or bdwrite must eventually be called to free the block for
use by others.
As mentioned, each buffer header contains a flag word which indicates the status of the buffer. Since they provide
one important channel for information between the drivers and the block I/O system, it is important to understand
these flags. The following names are manifest constants which select the associated flag bits.

B READ
This bit is set when the buffer is handed to the device strategy routine (see below) to indicate a read operation. The
symbol B WRITE is defined as 0 and does not define a flag; it is provided as a mnemonic convenience to callers of
routines like swap which have a separate argument which indicates read or write.

B DONE
This bit is set to 0 when a block is handed to the the device strategy routine and is turned on when the operation
completes, whether normally as the result of an error. It is also used as part of the return argument of getblk to indi-
cate if 1 that the returned buffer actually contains the data in the requested block.

B ERROR
This bit may be set to 1 when B DONE is set to indicate that an I/O or other error occurred. If it is set the b error
byte of the buffer header may contain an error code if it is non-zero. If b error is 0 the nature of the error is not
specified. Actually no driver at present sets b error; the latter is provided for a future improvement whereby a more
detailed error-reporting scheme may be implemented.

B BUSY
This bit indicates that the buffer header is not on the free list, i.e. is dedicated to someone’s exclusive use. The
buffer still remains attached to the list of blocks associated with its device, however. When getblk (or bread, which
calls it) searches the buffer list for a given device and finds the requested block with this bit on, it sleeps until the bit
clears.

B WANTED
This flag is used in conjunction with the B BUSY bit. Before sleeping as described just above, getblk sets this flag.
Conversely, when the block is freed and the busy bit goes down (in brelse) a wakeup is given for the block header
whenever B WANTED is on. This strategem avoids the overhead of having to call wakeup every time a buffer is
freed on the chance that someone might want it.

B ASYNC
This bit is set by bawrite to indicate to the appropriate device driver that the buffer should be released when the write
has been finished, usually at interrupt time. The difference between bwrite and bawrite is that the former starts I/O,
waits until it is done, and frees the buffer. The latter merely sets this bit and starts I/O. The bit indicates that relse
should be called for the buffer on completion.

B DELWRI
This bit is set by bdwrite before releasing the buffer. When getblk, while searching for a free block, discovers the bit
is 1 in a buffer it would otherwise grab, it causes the block to be written out before reusing it.

B XMEM
This is actually a mask for the pair of bits which contain the high-order two bits of the physical address of the origin
-

Unix I/O System - 7

of the buffer; these bits are an extension of the 16 address bits elsewhere in the buffer header.

B RELOC
This bit is currently unused; it previously indicated that a system-wide relocation constant was to be added to the
buffer address. It was needed during a period when addresses of data in the system (including the buffers) were
mapped by the relocation hardware to a physical address differing from its virtual address.

Block Device Drivers

The bdevsw table contains the names of the interface routines and that of a table for each block device.
Just as for character devices, block device drivers may supply an open and a close routine called respectively on
each open and on the final close of the device. Instead of separate read and write routines, each block device driver
has a strategy routine which is called with a pointer to a buffer header as argument. As discussed, the buffer header
contains a read/write flag, the core address (including extended-memory bits), the block number, a (negative) word
count, and the major and minor device number. The rôle of the strategy routine is to carry out the operation as re-
quested by the information in the buffer header. When the transaction is complete the B DONE (and possibly the
B ERROR) bits should be set. Then if the B ASYNC bit is set, brelse should be called; otherwise, wakeup. In cases
where the device is capable, under error-free operation, of transferring fewer words than requested, the device’s
word-count register should be placed in the residual count slot of the buffer header; otherwise, the residual count
should be set to 0. This particular mechanism is really for the benefit of the magtape driver; when reading this de-
vice records shorter than requested are quite normal, and the user should be told the actual length of the record.
[However the mechanism has not been integrated into normal I/O even on magtape and is used only in ‘‘raw’’ I/O as
discussed below.]
Notice that although the most usual argument to the strategy routines is a genuine buffer header allocated as dis-
cussed above, all that is actually required is that the argument be a pointer to a place containing the appropriate in-
formation. For example the swap routine, which manages movement of core images to and from the swapping de-
vice, uses the strategy routine for this device. Care has to be taken that no extraneous bits get turned on in the flag
word.
The device’s table specified by bdevsw has a byte to contain an active flag and an error count, a pair of links which
constitute the head of the chain of buffers for the device (b forw, b back), and a first and last pointer for a device
queue. Of these things, all are used solely by the device driver itself except for the buffer-chain pointers. Typically
the flag encodes the state of the device, and is used at a minimum to indicate that the device is currently engaged in
transferring information and no new command should be issued. The error count is useful for counting retries when
errors occur. The device queue is used to remember stacked requests; in the simplest case it may be maintained as a
first-in first-out list. Since buffers which have been handed over to the strategy routines are never on the list of free
buffers, the pointers in the buffer which maintain the free list (av forw, av back) are also used to contain the point-
ers which maintain the device queues.
A couple of routines are provided which are useful to block device drivers.
iodone(bp) ,
arranges that the buffer to which bp points be released or awakened, as appropriate, when the strategy module has
finished with the buffer, either normally or after an error. (In the latter case the B ERROR bit has presumably been
set.)
When the device conforms to some rather loose standards adhered to by certain DEC hardware, the routine
devstart(bp, devloc, devblk, hbcom)
is useful. Here bp is the address of the buffer header, devloc is the address of the slot in the device registers which
accepts a perhaps-encoded device block number, devblk is the block number, and hbcom is a quantity to be stored in
the high byte of the device’s command register. It is understood, when using this routine, that the device registers
are laid out in the order
command register
word count
core address
block (or track or sector)
-

8 - Unix I/O System

where the address of the last corresponds to devloc. Moreover, the device should correspond to the RP, RK, and RF
devices with respect to its layout of extended-memory bits and structure of read and write commands.
The routine
geterror(bp)
can be used to examine the error bit in a buffer header and arrange that any error indication found therein is reflected
to the user. It may be called only in the non-interrupt part of a driver when I/O has completed (B DONE has been
set).

An example
The RF disk driver is worth studying as the simplest example of a block I/O device. Its strategy routine checks to
see if the requested block lies beyond the end of the device; the size of the disk, in this instance, is indicated by the
minor device number. If the request is plausible, the buffer is placed at the end of the device queue, and if the disk is
not running, rfstart is called.
Rfstart merely returns if there is nothing to do, but otherwise sets the device-active flag, loads the address extension
register, and calls devstart to perform the remaining tasks attendant on beginning a data transfer.
When a completion or error interrupt occurs, rfintr is called. If an error is indicated, and if the error count has not
exceeded 10, the same transaction is reattempted; otherwise the error bit is set. If there was no error or if 10 failing
transfers have been issued the queue is advanced and rfstart is called to begin another transaction.

Raw Block-device I/O

A scheme has been set up whereby block device drivers may provide the ability to transfer information directly be-
tween the user’s core image and the device without the use of buffers and in blocks as large as the caller requests.
The method involves setting up a character-type special file corresponding to the raw device and providing read and
write routines which set up what is usually a private, non-shared buffer header with the appropriate information and
call the device’s strategy routine. If desired, separate open and close routines may be provided but this is usually un-
necessary. A special-function routine might come in handy, especially for magtape.
A great deal of work has to be done to generate the ‘‘appropriate information’’ to put in the argument buffer for the
strategy module; the worst part is to map relocated user addresses to physical addresses. Most of this work is done
by
physio(strat, bp, dev, rw)
whose arguments are the name of the strategy routine strat, the buffer pointer bp, the device number dev, and a read-
write flag rw whose value is either B READ or B WRITE. Physio makes sure that the user’s base address and count
are even (because most devices work in words) and that the core area affected is contiguous in physical space; it de-
lays until the buffer is not busy, and makes it busy while the operation is in progress; and it sets up user error return
information.
The magtape driver is the only one which as of this writing provides a raw I/O capability; given physio, the work in-
volved is trivial, and amounts to passing back to the user information on the length of the record read or written.
(There is some funniness because the magtape, uniquely among DEC devices, works in bytes, not words.) Putting in
raw I/O for disks should be equally trivial except that the disk address has to be carefully checked to make sure it
does not overflow from one logical device to another on which the caller may not have write permission.
 On the Security of UNIX

Dennis M. Ritchie

Bell Laboratories, Murray Hill, N. J.

Recently there has been much interest in the security aspects of operating systems and software. At issue is
the ability to prevent undesired disclosure of information, destruction of information, and harm to the functioning of
the system. This paper discusses the degree of security which can be provided under the UNIX system and offers a
number of hints on how to improve security.
The first fact to face is that UNIX was not developed with security, in any realistic sense, in mind; this fact
alone guarantees a vast number of holes. (Actually the same statement can be made with respect to most systems.)
The area of security in which UNIX is theoretically weakest is in protecting against crashing or at least crippling the
operation of the system. The problem here is not mainly in uncritical acceptance of bad parameters to system
calls— there may be bugs in this area, but none are known— but rather in the lack of any checks for excessive con-
sumption of resources. Most notably, there is no limit on the amount of disk storage used, either in total space allo-
cated or in the number of files or directories. Here is a particularly ghastly shell sequence guaranteed to stop the
system:

: loop
mkdir x
chdir x
goto loop

Either a panic will occur because all the i-nodes on the device are used up or all the disk blocks will be consumed,
thus preventing anyone from writing files on the device.
Processes are another resource on which the only limit is total exhaustion. For example, the sequence

command&
command&
command&

if continued long enough will use up all the slots in the system’s process table and prevent anyone from executing
any commands. Alternatively, if the commands use much core, swap space may run out, causing a panic. Incident-
ly, because of the implementation of process termination, the above sequence is effective in stopping the system no
matter how short a time it takes each command to terminate. (The process-table slot is not freed until the terminated
process is waited for; if no commands without ‘‘&’’ are executed, the Shell never executes a ‘‘wait.’’)
It should be evident that unbounded consumption of disk space, files, swap space, and processes can easily oc-
cur accidentally in malfunctioning programs as well as at command level. In fact UNIX is essentially defenseless
against this kind of abuse, nor is there any easy fix. The best that can be said is that it is generally fairly easy to de-
tect what has happened when disaster strikes, to identify the user responsible, and take appropriate action. In prac-
tice, we have found that difficulties in this area are rather rare, but we have not been faced with malicious users, and
enjoy a fairly generous supply of resources which have served to cushion us against accidental overconsumption.
The picture is considerably brighter in the area of protection of information from unauthorized perusal and de-
struction. Here the degree of security seems (almost) adequate theoretically, and the problems lie more in the neces-
sity for care in the actual use of the system.
Each UNIX file has associated with it eleven bits of protection information together with a user identification
number and a user-group identification number (UID and GID). Nine of the protection bits are used to specify inde-
pendently permission to read, to write, and to execute the file to the user himself, to members of the user’s group,
and to all other users. Each process generated by or for a user has associated with it an effective UID and a real
UID, and an effective and real GID. When an attempt is made to access the file for reading, writing, or execution,
the user process’s effective UID is compared against the file’s UID; if a match is obtained, access is granted provided
the read, write, or execute bit respectively for the user himself is present. If the UID for the file and for the process
fail to match, but the GID’s do match, the group bits are used; if the GID’s do not match, the bits for other users are
 UNIX Security - 2

tested. The last two bits of each file’s protection information, called the set-UID and set-GID bits, are used only
when the file is executed as a program. If, in this case, the set-UID bit is on for the file, the effective UID for the
process is changed to the UID associated with the file; the change persists until the process terminates or until the
UID changed again by another execution of a set-UID file. Similarly the effective group ID of a process is changed
to the GID associated with a file when that file is executed and has the set-GID bit set. The real UID and GID of a
process do not change when any file is executed, but only as the result of a privileged system call.
The basic notion of the set-UID and set-GID bits is that one may write a program which is executable by oth-
ers and which maintains files accessible to others only by that program. The classical example is the game-playing
program which maintains records of the scores of its players. The program itself has to read and write the score file,
but no one but the game’s sponsor can be allowed unrestricted access to the file lest they manipulate the game to
their own advantage. The solution is to turn on the set-UID bit of the game program. When, and only when, it is in-
voked by players of the game, it may update the score file ordinary programs executed by others cannot access the
score.
There are a number of special cases involved in determining access permissions. Since executing a directory
as a program is a meaningless operation, the execute-permission bit, for directories, is taken instead to mean permis-
sion to search the directory for a given file during the scanning of a path name; thus if a directory has execute per-
mission but no read permission for a given user, he may access files with known names in the directory, but may not
read (list) the entire contents of the directory. Write permission on a directory is interpreted to mean that the user
may create and delete files in that directory; it is impossible for any user to write directly into any directory.
Another, and from the point of view of security, much more serious special case is that there is a ‘‘super user’’
who is able to read any file and write any non-directory. The super-user is also able to change the protection mode
and the owner UID and GID of any file and to invoke privileged system calls. It must be recognized that the mere
notion of a super-user is a theoretical, and usually practical, blemish on any protection scheme.
The first necessity for a secure system is of course arranging that all files and directories have the proper pro-
tection modes. Unfortunately, UNIX software is exceedingly permissive in this regard; essentially all commands
create files readable and writable by everyone. This means that more or less continuous attention must be paid to
adjusting modes properly. If one wants to keep one’s files completely secret, it is possible to remove all permissions
from the directory in which they live, which is easy and effective; but if it is desired to give general read permission
while preventing writing, things are more complicated. The main problem is that write permission in a directory
means precisely that; it has nothing to do with write permission for a file in that directory. Thus a writeable file in a
read-only directory may be changed, or even truncated, though not removed. This fact is perfectly logical, though in
this case unfortunate. A case can be made for requiring write permission for the directory of a file as well as for the
file itself before allowing writing. (This possibility is more complicated than it seems at first; the system has to al-
low users to change their own directories while forbidding them to change the user-directory directory.)
A situation converse to the above-discussed difficulty is also present— it is possible to delete a file if one has
write permission for its directory independently of any permissions for the file. This problem is related more to self-
protection than protection from others. It is largely mitigated by the fact that the two major commands which delete
named files (mv and rm) ask confirmation before deleting unwritable files.
It follows from this discussion that to maintain both data privacy and data integrity, it is necessary, and largely
sufficient, to make one’s directory inaccessible to others. The lack of sufficiency could follow from the existence of
set-UID programs created by the user and the possibility of total breach of system security in one of the ways dis-
cussed below (or one of the ways not discussed below).
Needless to say, the system administrators must be at least as careful as their most demanding user to place the
correct protection mode on the files under their control. In particular, it is necessary that special files be protected
from writing, and probably reading, by ordinary users when they store sensitive files belonging to other users. It is
easy to write programs that examine and change files by accessing the device on which the files live.
On the issue of password security, UNIX is probably better than most systems. Passwords are stored in an en-
crypted form which, in the absence of serious attention from specialists in the field, appears reasonably secure, pro-
vided its limitations are understood. Since both the encryption algorithm and the encrypted passwords are available,
exhaustive enumeration of potential passwords is feasible up to a point. As a practical test of the possibilities in this
 UNIX Security - 3

area, 67 encrypted passwords were collected from 10 UNIX installations. These were tested against all five-letter
combinations, all combinations of letters and digits of length four or less, and all words in Webster’s Second
unabridged dictionary; 60 of the 67 passwords were found. The whole process took about 12 hours of machine time.
This experience suggests that passwords should be at least six characters long and randomly chosen from an alpha-
bet which includes digits and special characters.
Of course there also exist feasible non-cryptanalytic ways of finding out passwords. For example: write a pro-
gram which types out ‘‘login: ’’ on the typewriter and copies whatever is typed to a file of your own. Then invoke
the command and go away until the victim arrives. (It is this kind of possibility that makes it evident that UNIX was
not designed to be secure.)
The set-UID (set-GID) notion must be used carefully if any security is to be maintained. The first thing to
keep in mind is that a writable set-UID file can have another program copied onto it. For example, if the super-user
(su) command is writable, anyone can copy the shell onto it and get a password-free version of su. A more subtle
problem can come from set-UID programs which are not sufficiently careful of what is fed into them. In some sys-
tems, for example, the mail command is set-UID and owned by the super-user. The notion is that one should be able
to send mail to anyone even if they want to protect their directories from writing. The trouble is that mail is rather
dumb: anyone can mail someone else’s private file to himself. Much more serious, is the following scenario: make a
file with a line like one in the password file which allows one to log in as the super-user. Then make a link named
‘‘.mail’’ to the password file in some writeable directory on the same device as the password file (say /tmp). Finally
mail the bogus login line to /tmp/.mail; You can then login as the super-user, clean up the incriminating evidence,
and have your will.
The fact that users can mount their own disks and tapes as file systems can be another way of gaining super-
user status. Once a disk pack is mounted, the system believes what is on it. Thus one can take a blank disk pack,
put on it anything desired, and mount it. There are obvious and unfortunate consequences. For example: a mounted
disk with garbage on it will crash the system; one of the files on the mounted disk can easily be a password-free ver-
sion of su; other files can be unprotected entries for special files. The only easy fix for this problem is to forbid the
use of mount to unprivileged users. A partial solution, not so restrictive, would be to have the mount command ex-
amine the special file for bad data, set-UID programs owned by others, and accessible special files, and balk at un-
privileged invokers.
 DC − An Interactive Desk Calculator

Robert Morris
Lorinda Cherry

ABSTRACT

DC is an interactive desk calculator program implemented on the UNIX time-shar-

ing system to do arbitrary-precision integer arithmetic. It has provision for manipulating
scaled fixed-point numbers and for input and output in bases other than decimal.
The size of numbers that can be manipulated is limited only by available core stor-
age. On typical implementations of UNIX, the size of numbers that can be handled varies
from several hundred digits on the smallest systems to several thousand on the largest.

January 6, 1998
 DC − An Interactive Desk Calculator

Robert Morris
Lorinda Cherry

DC is an arbitrary precision arithmetic package implemented on the UNIX time-sharing system in

the form of an interactive desk calculator. It works like a stacking calculator using reverse Polish notation.
Ordinarily DC operates on decimal integers, but one may specify an input base, output base, and a number
of fractional digits to be maintained.
A language called BC [1] has been developed which accepts programs written in the familiar style of
higher-level programming languages and compiles output which is interpreted by DC. Some of the com-
mands described below were designed for the compiler interface and are not easy for a human user to
manipulate.
Numbers that are typed into DC are put on a push-down stack. DC commands work by taking the
top number or two off the stack, performing the desired operation, and pushing the result on the stack. If an
argument is given, input is taken from that file until its end, then from the standard input.

SYNOPTIC DESCRIPTION
Here we describe the DC commands that are intended for use by people. The additional commands
that are intended to be invoked by compiled output are described in the detailed description.
Any number of commands are permitted on a line. Blanks and new-line characters are ignored
except within numbers and in places where a register name is expected.
The following constructions are recognized:

number
The value of the number is pushed onto the main stack. A number is an unbroken string of the digits
0-9 and the capital letters A−F which are treated as digits with values 10−15 respectively. The num-
ber may be preceded by an underscore _ to input a negative number. Numbers may contain decimal
points.

+ − * % ˆ
The top two values on the stack are added (+), subtracted (−), multiplied (*), divided (/), remaindered
(%), or exponentiated (ˆ). The two entries are popped off the stack; the result is pushed on the stack
in their place. The result of a division is an integer truncated toward zero. See the detailed descrip-
tion below for the treatment of numbers with decimal points. An exponent must not have any digits
after the decimal point.

sx
The top of the main stack is popped and stored into a register named x, where x may be any character.
If the s is capitalized, x is treated as a stack and the value is pushed onto it. Any character, even
blank or new-line, is a valid register name.

lx
The value in register x is pushed onto the stack. The register x is not altered. If the l is capitalized,
register x is treated as a stack and its top value is popped onto the main stack.
 -2-

All registers start with empty value which is treated as a zero by the command l and is treated as an error by
the command L.

d
The top value on the stack is duplicated.

p
The top value on the stack is printed. The top value remains unchanged.

f
All values on the stack and in registers are printed.

x
treats the top element of the stack as a character string, removes it from the stack, and executes it as a
string of DC commands.

[ ... ]
puts the bracketed character string onto the top of the stack.

q
exits the program. If executing a string, the recursion level is popped by two. If q is capitalized, the
top value on the stack is popped and the string execution level is popped by that value.

<x >x =x !<x !>x !=x

The top two elements of the stack are popped and compared. Register x is executed if they obey the
stated relation. Exclamation point is negation.

v
replaces the top element on the stack by its square root. The square root of an integer is truncated to
an integer. For the treatment of numbers with decimal points, see the detailed description below.

!
interprets the rest of the line as a UNIX command. Control returns to DC when the UNIX command
terminates.

c
All values on the stack are popped; the stack becomes empty.

i
The top value on the stack is popped and used as the number radix for further input. If i is capital-
ized, the value of the input base is pushed onto the stack. No mechanism has been provided for the
input of arbitrary numbers in bases less than 1 or greater than 16.

o
The top value on the stack is popped and used as the number radix for further output. If o is capital-
ized, the value of the output base is pushed onto the stack.
 -3-

k
The top of the stack is popped, and that value is used as a scale factor that influences the number of
decimal places that are maintained during multiplication, division, and exponentiation. The scale
factor must be greater than or equal to zero and less than 100. If k is capitalized, the value of the
scale factor is pushed onto the stack.

z
The value of the stack level is pushed onto the stack.

?
A line of input is taken from the input source (usually the console) and executed.

DETAILED DESCRIPTION

Internal Representation of Numbers

Numbers are stored internally using a dynamic storage allocator. Numbers are kept in the form of a
string of digits to the base 100 stored one digit per byte (centennial digits). The string is stored with the
low-order digit at the beginning of the string. For example, the representation of 157 is 57,1. After any
arithmetic operation on a number, care is taken that all digits are in the range 0−99 and that the number has
no leading zeros. The number zero is represented by the empty string.
Negative numbers are represented in the 100’s complement notation, which is analogous to two’s
complement notation for binary numbers. The high order digit of a negative number is always −1 and all
other digits are in the range 0−99. The digit preceding the high order −1 digit is never a 99. The represen-
tation of −157 is 43,98,−1. We shall call this the canonical form of a number. The advantage of this kind
of representation of negative numbers is ease of addition. When addition is performed digit by digit, the
result is formally correct. The result need only be modified, if necessary, to put it into canonical form.
Because the largest valid digit is 99 and the byte can hold numbers twice that large, addition can be
carried out and the handling of carries done later when that is convenient, as it sometimes is.
An additional byte is stored with each number beyond the high order digit to indicate the number of
assumed decimal digits after the decimal point. The representation of .001 is 1,3 where the scale has been
italicized to emphasize the fact that it is not the high order digit. The value of this extra byte is called the
scale factor of the number.

The Allocator
DC uses a dynamic string storage allocator for all of its internal storage. All reading and writing of
numbers internally is done through the allocator. Associated with each string in the allocator is a four-word
header containing pointers to the beginning of the string, the end of the string, the next place to write, and
the next place to read. Communication between the allocator and DC is done via pointers to these headers.
The allocator initially has one large string on a list of free strings. All headers except the one point-
ing to this string are on a list of free headers. Requests for strings are made by size. The size of the string
actually supplied is the next higher power of 2. When a request for a string is made, the allocator first
checks the free list to see if there is a string of the desired size. If none is found, the allocator finds the next
larger free string and splits it repeatedly until it has a string of the right size. Left-over strings are put on
the free list. If there are no larger strings, the allocator tries to coalesce smaller free strings into larger ones.
Since all strings are the result of splitting large strings, each string has a neighbor that is next to it in core
and, if free, can be combined with it to make a string twice as long. This is an implementation of the
‘buddy system’ of allocation described in [2].
Failing to find a string of the proper length after coalescing, the allocator asks the system for more
space. The amount of space on the system is the only limitation on the size and number of strings in DC.
If at any time in the process of trying to allocate a string, the allocator runs out of headers, it also asks the
system for more space.
 -4-

There are routines in the allocator for reading, writing, copying, rewinding, forward-spacing, and
backspacing strings. All string manipulation is done using these routines.
The reading and writing routines increment the read pointer or write pointer so that the characters of
a string are read or written in succession by a series of read or write calls. The write pointer is interpreted
as the end of the information-containing portion of a string and a call to read beyond that point returns an
end-of-string indication. An attempt to write beyond the end of a string causes the allocator to allocate a
larger space and then copy the old string into the larger block.

Internal Arithmetic
All arithmetic operations are done on integers. The operands (or operand) needed for the operation
are popped from the main stack and their scale factors stripped off. Zeros are added or digits removed as
necessary to get a properly scaled result from the internal arithmetic routine. For example, if the scale of
the operands is different and decimal alignment is required, as it is for addition, zeros are appended to the
operand with the smaller scale. After performing the required arithmetic operation, the proper scale factor
is appended to the end of the number before it is pushed on the stack.
A register called scale plays a part in the results of most arithmetic operations. scale is the bound on
the number of decimal places retained in arithmetic computations. scale may be set to the number on the
top of the stack truncated to an integer with the k command. K may be used to push the value of scale on
the stack. scale must be greater than or equal to 0 and less than 100. The descriptions of the individual
arithmetic operations will include the exact effect of scale on the computations.

Addition and Subtraction

The scales of the two numbers are compared and trailing zeros are supplied to the number with the
lower scale to give both numbers the same scale. The number with the smaller scale is multiplied by 10 if
the difference of the scales is odd. The scale of the result is then set to the larger of the scales of the two
operands.
Subtraction is performed by negating the number to be subtracted and proceeding as in addition.
Finally, the addition is performed digit by digit from the low order end of the number. The carries
are propagated in the usual way. The resulting number is brought into canonical form, which may require
stripping of leading zeros, or for negative numbers replacing the high-order configuration 99,−1 by the digit
−1. In any case, digits which are not in the range 0−99 must be brought into that range, propagating any
carries or borrows that result.

Multiplication
The scales are removed from the two operands and saved. The operands are both made positive.
Then multiplication is performed in a digit by digit manner that exactly mimics the hand method of multi-
plying. The first number is multiplied by each digit of the second number, beginning with its low order
digit. The intermediate products are accumulated into a partial sum which becomes the final product. The
product is put into the canonical form and its sign is computed from the signs of the original operands.
The scale of the result is set equal to the sum of the scales of the two operands. If that scale is larger
than the internal register scale and also larger than both of the scales of the two operands, then the scale of
the result is set equal to the largest of these three last quantities.

Division
The scales are removed from the two operands. Zeros are appended or digits removed from the divi-
dend to make the scale of the result of the integer division equal to the internal quantity scale. The signs
are removed and saved.
Division is performed much as it would be done by hand. The difference of the lengths of the two
numbers is computed. If the divisor is longer than the dividend, zero is returned. Otherwise the top digit of
the divisor is divided into the top two digits of the dividend. The result is used as the first (high-order) digit
of the quotient. It may turn out be one unit too low, but if it is, the next trial quotient will be larger than 99
 -5-

and this will be adjusted at the end of the process. The trial digit is multiplied by the divisor and the result
subtracted from the dividend and the process is repeated to get additional quotient digits until the remaining
dividend is smaller than the divisor. At the end, the digits of the quotient are put into the canonical form,
with propagation of carry as needed. The sign is set from the sign of the operands.

Remainder
The division routine is called and division is performed exactly as described. The quantity returned
is the remains of the dividend at the end of the divide process. Since division truncates toward zero,
remainders have the same sign as the dividend. The scale of the remainder is set to the maximum of the
scale of the dividend and the scale of the quotient plus the scale of the divisor.

Square Root
The scale is stripped from the operand. Zeros are added if necessary to make the integer result have a
scale that is the larger of the internal quantity scale and the scale of the operand.
The method used to compute sqrt(y) is Newton’s method with successive approximations by the rule

The initial guess is found by taking the integer square root of the top two digits.

Exponentiation
Only exponents with zero scale factor are handled. If the exponent is zero, then the result is 1. If the
exponent is negative, then it is made positive and the base is divided into one. The scale of the base is
removed.
The integer exponent is viewed as a binary number. The base is repeatedly squared and the result is
obtained as a product of those powers of the base that correspond to the positions of the one-bits in the
binary representation of the exponent. Enough digits of the result removed to make the scale of the result
the same as if the indicated multiplication had been performed.

Input Conversion and Base

Numbers are converted to the internal representation as they are read in. The scale stored with a
number is simply the number of fractional digits input. Negative numbers are indicated by preceding the
number with a . The hexadecimal digits A−F correspond to the numbers 10−15 regardless of input base.
The i command can be used to change the base of the input numbers. This command pops the stack, trun-
cates the resulting number to an integer, and uses it as the input base for all further input. The input base is
initialized to 10 but may, for example be changed to 8 or 16 to do octal or hexadecimal to decimal conver-
sions. The command I will push the value of the input base on the stack.

Output Commands
The command p causes the top of the stack to be printed. It does not remove the top of the stack. All
of the stack and internal registers can be output by typing the command f. The o command can be used to
change the output base. This command uses the top of the stack, truncated to an integer as the base for all
further output. The output base in initialized to 10. It will work correctly for any base. The command O
pushes the value of the output base on the stack.

Output Format and Base

The input and output bases only affect the interpretation of numbers on input and output; they have
no effect on arithmetic computations. Large numbers are output with 70 characters per line; a \ indicates a
continued line. All choices of input and output bases work correctly, although not all are useful. A particu-
larly useful output base is 100000, which has the effect of grouping digits in fives. Bases of 8 and 16 can
be used for decimal-octal or decimal-hexadecimal conversions.
 -6-

Internal Registers
Numbers or strings may be stored in internal registers or loaded on the stack from registers with the
commands s and l. The command sx pops the top of the stack and stores the result in register x. x can be
any character. lx puts the contents of register x on the top of the stack. The l command has no effect on the
contents of register x. The s command, however, is destructive.

Stack Commands
The command c clears the stack. The command d pushes a duplicate of the number on the top of the
stack on the stack. The command z pushes the stack size on the stack. The command X replaces the num-
ber on the top of the stack with its scale factor. The command Z replaces the top of the stack with its
length.

Subroutine Definitions and Calls

Enclosing a string in [] pushes the ascii string on the stack. The q command quits or in executing a
string, pops the recursion levels by two.

Internal Registers − Programming DC

The load and store commands together with [] to store strings, x to execute and the testing commands
‘<’, ‘>’, ‘=’, ‘!<’, ‘!>’, ‘!=’ can be used to program DC. The x command assumes the top of the stack is an
string of DC commands and executes it. The testing commands compare the top two elements on the stack
and if the relation holds, execute the register that follows the relation. For example, to print the numbers
0-9,
[lip1+ si li10>a]sa
0si lax

Push-Down Registers and Arrays

These commands were designed for used by a compiler, not by people. They involve push-down reg-
isters and arrays. In addition to the stack that commands work on, DC can be thought of as having individ-
ual stacks for each register. These registers are operated on by the commands S and L. Sx pushes the top
value of the main stack onto the stack for the register x. Lx pops the stack for register x and puts the result
on the main stack. The commands s and l also work on registers but not as push-down stacks. l doesn’t
effect the top of the register stack, and s destroys what was there before.
The commands to work on arrays are : and ;. :x pops the stack and uses this value as an index into
the array x. The next element on the stack is stored at this index in x. An index must be greater than or
equal to 0 and less than 2048. ;x is the command to load the main stack from the array x. The value on the
top of the stack is the index into the array x of the value to be loaded.

Miscellaneous Commands
The command ! interprets the rest of the line as a UNIX command and passes it to UNIX to execute.
One other compiler command is Q. This command uses the top of the stack as the number of levels of
recursion to skip.

DESIGN CHOICES
The real reason for the use of a dynamic storage allocator was that a general purpose program could
be (and in fact has been) used for a variety of other tasks. The allocator has some value for input and for
compiling (i.e. the bracket [...] commands) where it cannot be known in advance how long a string will be.
The result was that at a modest cost in execution time, all considerations of string allocation and sizes of
strings were removed from the remainder of the program and debugging was made easier. The allocation
method used wastes approximately 25% of available space.
The choice of 100 as a base for internal arithmetic seemingly has no compelling advantage. Yet the
base cannot exceed 127 because of hardware limitations and at the cost of 5% in space, debugging was
 -7-

made a great deal easier and decimal output was made much faster.
The reason for a stack-type arithmetic design was to permit all DC commands from addition to sub-
routine execution to be implemented in essentially the same way. The result was a considerable degree of
logical separation of the final program into modules with very little communication between modules.
The rationale for the lack of interaction between the scale and the bases was to provide an under-
standable means of proceeding after a change of base or scale when numbers had already been entered. An
earlier implementation which had global notions of scale and base did not work out well. If the value of
scale were to be interpreted in the current input or output base, then a change of base or scale in the midst
of a computation would cause great confusion in the interpretation of the results. The current scheme has
the advantage that the value of the input and output bases are only used for input and output, respectively,
and they are ignored in all other operations. The value of scale is not used for any essential purpose by any
part of the program and it is used only to prevent the number of decimal places resulting from the arith-
metic operations from growing beyond all bounds.
The design rationale for the choices for the scales of the results of arithmetic were that in no case
should any significant digits be thrown away if, on appearances, the user actually wanted them. Thus, if the
user wants to add the numbers 1.5 and 3.517, it seemed reasonable to give him the result 5.017 without
requiring him to unnecessarily specify his rather obvious requirements for precision.
On the the other hand, multiplication and exponentiation produce results with many more digits than
their operands and it seemed reasonable to give as a minimum the number of decimal places in the
operands but not to give more than that number of digits unless the user asked for them by specifying a
value for scale. Square root can be handled in just the same way as multiplication. The operation of divi-
sion gives arbitrarily many decimal places and there is simply no way to guess how many places the user
wants. In this case only, the user must specify a scale to get any decimal places at all.
The scale of remainder was chosen to make it possible to recreate the dividend from the quotient and
remainder. This is easy to implement; no digits are thrown away.

References
[1] L. L. Cherry, R. Morris, BC − An Arbitrary Precision Desk-Calculator Language,
[2] K. C. Knowlton, A Fast Storage Allocator, Comm. ACM 8, pp. 623-625 (Oct. 1965)
 BC − An Arbitrary Precision Desk-Calculator Language

Lorinda Cherry
Robert Morris

ABSTRACT

BC is a language and a compiler for doing arbitrary precision arithmetic on the

PDP-11 under the UNIX time-sharing system. The output of the compiler is interpreted
and executed by a collection of routines which can input, output, and do arithmetic on
indefinitely large integers and on scaled fixed-point numbers.
These routines are themselves based on a dynamic storage allocator. Overflow
does not occur until all available core storage is exhausted.
The language has a complete control structure as well as immediate-mode opera-
tion. Functions can be defined and saved for later execution.
Two five hundred-digit numbers can be multiplied to give a thousand digit result in
about ten seconds.
A small collection of library functions is also available, including sin, cos, arctan,
log, exponential, and Bessel functions of integer order.
Some of the uses of this compiler are
− to do computation with large integers,
− to do computation accurate to many decimal places,
− conversion of numbers from one base to another base.

January 6, 1998
 BC − An Arbitrary Precision Desk-Calculator Language

Lorinda Cherry
Robert Morris

Introduction
BC is a language and a compiler for doing arbitrary precision arithmetic on the UNIX time-sharing
system [1]. The compiler was written to make conveniently available a collection of routines (called DC
[6]) which are capable of doing arithmetic on integers of arbitrary size. The compiler is by no means
intended to provide a complete programming language. It is a minimal language facility.
There is a scaling provision that permits the use of decimal point notation. Provision is made for
input and output in bases other than decimal. Numbers can be converted from decimal to octal by simply
setting the output base to equal 8.
The actual limit on the number of digits that can be handled depends on the amount of storage avail-
able on the machine. Manipulation of numbers with many hundreds of digits is possible even on the small-
est versions of UNIX.
The syntax of BC has been deliberately selected to agree substantially with the C language [2,3].
Those who are familiar with C will find few surprises in this language.

Simple Computations with Integers

The simplest kind of statement is an arithmetic expression on a line by itself. For instance, if you
type in the line:
142857 + 285714
the program responds immediately with the line
428571
The operators −, *, /, %, and ˆ can also be used; they indicate subtraction, multiplication, division, remain-
dering, and exponentiation, respectively. Division of integers produces an integer result truncated toward
zero. Division by zero produces an error comment.
Any term in an expression may be prefixed by a minus sign to indicate that it is to be negated (the
‘unary’ minus sign). The expression
7+−3
is interpreted to mean that −3 is to be added to 7.
More complex expressions with several operators and with parentheses are interpreted just as in For-
tran, with ˆ having the greatest binding power, then * and % and /, and finally + and −. Contents of paren-
theses are evaluated before material outside the parentheses. Exponentiations are performed from right to
left and the other operators from left to right. The two expressions
aˆbˆc and aˆ(bˆc)
are equivalent, as are the two expressions
a*b*c and (a*b)*c
BC shares with Fortran and C the undesirable convention that
 -2-

a/b*c is equivalent to (a/b)*c

Internal storage registers to hold numbers have single lower-case letter names. The value of an
expression can be assigned to a register in the usual way. The statement
x=x+3
has the effect of increasing by three the value of the contents of the register named x. When, as in this case,
the outermost operator is an =, the assignment is performed but the result is not printed. Only 26 of these
named storage registers are available.
There is a built-in square root function whose result is truncated to an integer (but see scaling below).
The lines
x = sqrt(191)
x
produce the printed result
13

Bases
There are special internal quantities, called ‘ibase’ and ‘obase’. The contents of ‘ibase’, initially set
to 10, determines the base used for interpreting numbers read in. For example, the lines
ibase = 8
11
will produce the output line
9
and you are all set up to do octal to decimal conversions. Beware, however of trying to change the input
base back to decimal by typing
ibase = 10
Because the number 10 is interpreted as octal, this statement will have no effect. For those who deal in
hexadecimal notation, the characters A−F are permitted in numbers (no matter what base is in effect) and
are interpreted as digits having values 10−15 respectively. The statement
ibase = A
will change you back to decimal input base no matter what the current input base is. Negative and large
positive input bases are permitted but useless. No mechanism has been provided for the input of arbitrary
numbers in bases less than 1 and greater than 16.
The contents of ‘obase’, initially set to 10, are used as the base for output numbers. The lines
obase = 16
1000
will produce the output line
3E8
which is to be interpreted as a 3-digit hexadecimal number. Very large output bases are permitted, and they
are sometimes useful. For example, large numbers can be output in groups of five digits by setting ‘obase’
to 100000. Strange (i.e. 1, 0, or negative) output bases are handled appropriately.
Very large numbers are split across lines with 70 characters per line. Lines which are continued end
with \. Decimal output conversion is practically instantaneous, but output of very large numbers (i.e., more
than 100 digits) with other bases is rather slow. Non-decimal output conversion of a one hundred digit
number takes about three seconds.
 -3-

It is best to remember that ‘ibase’ and ‘obase’ have no effect whatever on the course of internal com-
putation or on the evaluation of expressions, but only affect input and output conversion, respectively.

Scaling
A third special internal quantity called ‘scale’ is used to determine the scale of calculated quantities.
Numbers may have up to 99 decimal digits after the decimal point. This fractional part is retained in fur-
ther computations. We refer to the number of digits after the decimal point of a number as its scale.
When two scaled numbers are combined by means of one of the arithmetic operations, the result has
a scale determined by the following rules. For addition and subtraction, the scale of the result is the larger
of the scales of the two operands. In this case, there is never any truncation of the result. For multiplica-
tions, the scale of the result is never less than the maximum of the two scales of the operands, never more
than the sum of the scales of the operands and, subject to those two restrictions, the scale of the result is set
equal to the contents of the internal quantity ‘scale’. The scale of a quotient is the contents of the internal
quantity ‘scale’. The scale of a remainder is the sum of the scales of the quotient and the divisor. The
result of an exponentiation is scaled as if the implied multiplications were performed. An exponent must be
an integer. The scale of a square root is set to the maximum of the scale of the argument and the contents
of ‘scale’.
All of the internal operations are actually carried out in terms of integers, with digits being discarded
when necessary. In every case where digits are discarded, truncation and not rounding is performed.
The contents of ‘scale’ must be no greater than 99 and no less than 0. It is initially set to 0. In case
you need more than 99 fraction digits, you may arrange your own scaling.
The internal quantities ‘scale’, ‘ibase’, and ‘obase’ can be used in expressions just like other vari-
ables. The line
scale = scale + 1
increases the value of ‘scale’ by one, and the line
scale
causes the current value of ‘scale’ to be printed.
The value of ‘scale’ retains its meaning as a number of decimal digits to be retained in internal com-
putation even when ‘ibase’ or ‘obase’ are not equal to 10. The internal computations (which are still con-
ducted in decimal, regardless of the bases) are performed to the specified number of decimal digits, never
hexadecimal or octal or any other kind of digits.

Functions
The name of a function is a single lower-case letter. Function names are permitted to collide with
simple variable names. Twenty-six different defined functions are permitted in addition to the twenty-six
variable names. The line
define a(x){
begins the definition of a function with one argument. This line must be followed by one or more state-
ments, which make up the body of the function, ending with a right brace }. Return of control from a func-
tion occurs when a return statement is executed or when the end of the function is reached. The return
statement can take either of the two forms
return
return(x)
In the first case, the value of the function is 0, and in the second, the value of the expression in parentheses.
Variables used in the function can be declared as automatic by a statement of the form
auto x,y,z
There can be only one ‘auto’ statement in a function and it must be the first statement in the definition.
 -4-

These automatic variables are allocated space and initialized to zero on entry to the function and thrown
away on return. The values of any variables with the same names outside the function are not disturbed.
Functions may be called recursively and the automatic variables at each level of call are protected. The
parameters named in a function definition are treated in the same way as the automatic variables of that
function with the single exception that they are given a value on entry to the function. An example of a
function definition is
define a(x,y){
auto z
z = x*y
return(z)
}
The value of this function, when called, will be the product of its two arguments.
A function is called by the appearance of its name followed by a string of arguments enclosed in
parentheses and separated by commas. The result is unpredictable if the wrong number of arguments is
used.
Functions with no arguments are defined and called using parentheses with nothing between them:
b().
If the function a above has been defined, then the line
a(7,3.14)
would cause the result 21.98 to be printed and the line
x = a(a(3,4),5)
would cause the value of x to become 60.

Subscripted Variables
A single lower-case letter variable name followed by an expression in brackets is called a subscripted
variable (an array element). The variable name is called the array name and the expression in brackets is
called the subscript. Only one-dimensional arrays are permitted. The names of arrays are permitted to col-
lide with the names of simple variables and function names. Any fractional part of a subscript is discarded
before use. Subscripts must be greater than or equal to zero and less than or equal to 2047.
Subscripted variables may be freely used in expressions, in function calls, and in return statements.
An array name may be used as an argument to a function, or may be declared as automatic in a func-
tion definition by the use of empty brackets:
f(a[ ])
define f(a[ ])
auto a[ ]
When an array name is so used, the whole contents of the array are copied for the use of the function, and
thrown away on exit from the function. Array names which refer to whole arrays cannot be used in any
other contexts.

Control Statements
The ‘if’, the ‘while’, and the ‘for’ statements may be used to alter the flow within programs or to
cause iteration. The range of each of them is a statement or a compound statement consisting of a collec-
tion of statements enclosed in braces. They are written in the following way
if(relation) statement
while(relation) statement
for(expression1; relation; expression2) statement
or
 -5-

if(relation) {statements}
while(relation) {statements}
for(expression1; relation; expression2) {statements}

A relation in one of the control statements is an expression of the form

x>y
where two expressions are related by one of the six relational operators <, >, <=, >=, ==, or !=. The rela-
tion == stands for ‘equal to’ and != stands for ‘not equal to’. The meaning of the remaining relational oper-
ators is clear.
BEWARE of using = instead of == in a relational. Unfortunately, both of them are legal, so you will
not get a diagnostic message, but = really will not do a comparison.
The ‘if’ statement causes execution of its range if and only if the relation is true. Then control passes
to the next statement in sequence.
The ‘while’ statement causes execution of its range repeatedly as long as the relation is true. The
relation is tested before each execution of its range and if the relation is false, control passes to the next
statement beyond the range of the while.
The ‘for’ statement begins by executing ‘expression1’. Then the relation is tested and, if true, the
statements in the range of the ‘for’ are executed. Then ‘expression2’ is executed. The relation is tested,
and so on. The typical use of the ‘for’ statement is for a controlled iteration, as in the statement
for(i=1; i<=10; i=i+1) i
which will print the integers from 1 to 10. Here are some examples of the use of the control statements.
define f(n){
auto i, x
x=1
for(i=1; i<=n; i=i+1) x=x*i
return(x)
}
The line
f(a)
will print a factorial if a is a positive integer. Here is the definition of a function which will compute values
of the binomial coefficient (m and n are assumed to be positive integers).
define b(n,m){
auto x, j
x=1
for(j=1; j<=m; j=j+1) x=x*(n−j+1)/j
return(x)
}
The following function computes values of the exponential function by summing the appropriate series
without regard for possible truncation errors:
 -6-

scale = 20
define e(x){
auto a, b, c, d, n
a=1
b=1
c=1
d=0
n=1
while(1==1){
a = a*x
b = b*n
c = c + a/b
n=n+1
if(c==d) return(c)
d=c
}
}

Some Details
There are some language features that every user should know about even if he will not use them.
Normally statements are typed one to a line. It is also permissible to type several statements on a line
separated by semicolons.
If an assignment statement is parenthesized, it then has a value and it can be used anywhere that an
expression can. For example, the line
(x=y+17)
not only makes the indicated assignment, but also prints the resulting value.
Here is an example of a use of the value of an assignment statement even when it is not parenthe-
sized.
x = a[i=i+1]
causes a value to be assigned to x and also increments i before it is used as a subscript.
The following constructs work in BC in exactly the same manner as they do in the C language. Con-
sult the appendix or the C manuals [2,3] for their exact workings.
x=y=z is the same as x=(y=z)
x =+ y x = x+y
x =− y x = x−y
x =* y x = x*y
x =/ y x = x/y
x =% y x = x%y
x =ˆ y x = xˆy
x++ (x=x+1)−1
x−− (x=x−1)+1
++x x = x+1
−−x x = x−1
Even if you don’t intend to use the constructs, if you type one inadvertently, something correct but unex-
pected may happen.
WARNING! In some of these constructions, spaces are significant. There is a real difference
between x=−y and x= −y. The first replaces x by x−y and the second by −y.
 -7-

Three Important Things

1. To exit a BC program, type ‘quit’.
2. There is a comment convention identical to that of C and of PL/I. Comments begin with ‘/*’ and
end with ‘*/’.
3. There is a library of math functions which may be obtained by typing at command level
bc −l
This command will load a set of library functions which, at the time of writing, consists of sine (named ‘s’),
cosine (‘c’), arctangent (‘a’), natural logarithm (‘l’), exponential (‘e’) and Bessel functions of integer order
(‘j(n,x)’). Doubtless more functions will be added in time. The library sets the scale to 20. You can reset it
to something else if you like. The design of these mathematical library routines is discussed elsewhere [4].
If you type
bc file ...
BC will read and execute the named file or files before accepting commands from the keyboard. In this
way, you may load your favorite programs and function definitions.

Acknowledgement
The compiler is written in YACC [5]; its original version was written by S. C. Johnson.

References
[1] K. Thompson and D. M. Ritchie, UNIX Programmer’s Manual, Fifth Edition (1974)
[2] D. M. Ritchie, C Reference Manual,
[3] B. W. Kernighan, Programming in C: A Tutorial,
[4] Robert Morris, A Library of Reference Standard Mathematical Subroutines,
[5] S. C. Johnson, YACC, Yet Another Compiler-Compiler,
[6] R. Morris and L. L. Cherry, DC − An Interactive Desk Calculator,
 The Portable C Library (on UNIX) *

M. E. Lesk

1. INTRODUCTION
The C language [1] now exists on three operating systems. A set of library routines common to PDP
11 UNIX, Honeywell 6000 GCOS, and IBM 370 OS has been provided to improve program portability. This
memorandum describes the UNIX implementation of the portable routines.
The programs defined here were chosen to follow the standard routines available on UNIX, with alter-
ations to improve transferability to other computer systems. It is expected that future C implementations
will try to support the basic library outlined in this document. It provides character stream input and output
on multiple files; simple accessing of files by name; and some elementary formatting and translating rou-
tines. The remainder of this memorandum lists the portable and non-portable library routines and explains
some of the programming aids available.
The I/O routines in the C library fall into several classes. Files are addressed through intermediate
numbers called file-descriptors which are described in section 2. Several default file-descriptors are pro-
vided by the system; other aspects of the system environment are explained in section 3.
Basic character-stream input and output involves the reading or writing of files considered as streams
of characters. The C library includes facilities for this, discussed in section 4. Higher-level character
stream operations permit translation of internal binary representations of numbers to and from character
representations, and formatting or unpacking of character data. These operations are performed with the
subprograms in section 5. Binary input and output routines permit data transmission without the cost of
translation to or from readable ASCII character representations. Such data transmission should only be
directed to files or tapes, and not to printers or terminals. As is usual with such routines, the only simple
guarantee that can be made to the programmer seeking portability is that data written by a particular
sequence of binary writes, if read by the exactly matching sequence of binary reads, will restore the previ-
ous contents of memory. Other reads or writes have system-dependent effects. See section 6 for a discus-
sion of binary input and output. Section 7 describes some further routines in the portable library. These
include a storage allocator and some other control and conversion functions.

2. FILE DESCRIPTORS
Except for the standard input and output files, all files must be explicitly opened before any I/O is
performed on them. When files are opened for writing, they are created if not already present. They must
be closed when finished, although the normal cexit routine will take care of that. When opened a disc file
or device is associated with a file descriptor, an integer between 0 and 9. This file descriptor is used for fur-
ther I/O to the file.
Initially you are given three file descriptors by the system: 0, 1, and 2. File 0 is the standard input; it
is normally the teletype in time-sharing or input data cards in batch. File 1 is the standard output; it is nor-
mally the teletype in time-sharing or the line printer in batch. File 2 is the error file; it is an output file, nor-
mally the same as file 1, except that when file 1 is diverted via a command line ’>’ operator, file 2 remains
attached to the original destination, usually the terminal. It is used for error message output. These popular
UNIX conventions are considered part of the C library specification. By closing 0 or 1, the default input or
output may be re-directed; this can also be done on the command line by >file for output or <file for input.

* This document is an abbreviated form of ‘‘The Portable C Library’’, by M. E. Lesk, describing only the UNIX section of
the library.
 -2-

Associated with the portable library are two external integers, named cin and cout. These are respec-
tively the numbers of the standard input unit and standard output unit. Initially 0 and 1 are used, but you
may redefine them at any time. These cells are used by the routines getchar, putchar, gets, and puts to
select their I-O unit number.

3. THE C ENVIRONMENT
The C language is almost exactly the same on all machines, except for essential machine differences
such as word length and number of characters per word. On UNIX ASCII character code is used. Characters
range from −128 to +127 in numeric value, there is sign extension when characters are assigned to integers,
and right shifts are arithmetic. The ‘‘first’’ character in a word is stored in the right half word.
More serious problems of compatibility are caused by the loaders on the different operating systems.
UNIX permits external names to be in upper and lower case, up to seven characters long. There may
be multiple external definitions (uninitialized) of the same name.
The C alphabet for identifier names includes the upper and lower case letters, the digits, and the
underline. It is not possible for C programs to communicate with FORTRAN programs.

4. BASIC CHARACTER STREAM ROUTINES

These routines transfer streams of characters in and out of C programs. Interpretation of the charac-
ters is left to the user. Facilities for interpreting numerical strings are presented in section 5; and routines to
transfer binary data to and from files or devices are discussed in section 6. In the following routine descrip-
tions, the optional argument fd represents a file-descriptor; if not present, it is taken to be 0 for input and 1
for output. When your program starts, remember that these are associated with the ‘‘standard’’ input and
output files.

COPEN (filename, type)

Copen initiates activity on a file; if necessary it will create the file too. Up to 10 files may be open at one
time. When called as described here, copen returns a filedescriptor for a character stream file. Values less
than zero returned by copen indicate an error trying to open the file. Other calls to copen are described in
sections 6 and 7.
Arguments :
Filename: a string representing a file name, according to the local operating system conventions. All accept
a string of letters and digits as a legal file name, although leading digits are not recommended on GCOS.
Type: a character ‘r’, ‘w’, or ‘a’ meaning read, write, or append. Note that the type is a single character,
whereas the file name must be a string.

CGETC ( fd )
Cgetc returns the next character from the input unit associated with fd. On end of file cgetc returns ‘\0’. To
signal end of file from the teletype, type the special symbol appropriate to UNIX: EOT (control-D)

CPUTC (ch , fd )
Cputc writes a character onto the given output unit. Cputc returns as its value the character written.
Output for disk files is buffered in 512 character units, irrespective of newlines; teletype output goes
character by character

CCLOSE (fd)
Activity on file fd is terminated and any output buffers are emptied. You usually don’t have to call cclose;
cexit will do it for you on all open files. However, to write some data on a file and then read it back in, the
correct sequence is:
 -3-

fd = copen (‘‘file’’, ‘w’);

write on fd ...
cclose (fd);
fd = copen(‘‘file’’, ‘r’);
read from fd ...

CFLUSH (fn)
To get buffer flushing, but retain the ability to write more on the file, you may call this routine.
Normally, output intended for the teletype is not buffered and this call is not needed.

CEXIT ([errcode])
Cexit closes all files and then terminates execution. If a non-zero argument is given, this is assumed to be
an error indication or other returned value to be signalled to the operating system.
Cexit must be called explicitly; a return from the main program is not adequate.

CEOF (fd)
Ceof returns nonzero when end of file has been reached on input unit fd.

GETCHAR ()
Getchar is a special case of cgetc; it reads one character from the standard input unit. Getchar ( ) is defined
as cgetc (cin); it should not have an argument.

PUTCHAR (ch)
Putchar (ch) is the same as cputc (ch, cout); it writes one character on the standard output.

GETS (s)
Gets reads everything up to the next newline into the string pointed to by s. If the last character read from
this input unit was newline, then gets reads the next line, which on GCOS and IBM corresponds exactly to a
logical record. The terminating newline is replaced by ‘\0’. The value of gets is s, or 0 if end of file.

PUTS (s)
Copies the string s onto the standard output unit. The terminating ‘\0’ is replaced by a newline character.
The value of puts is s.

UNGETC (ch , fd)

Ungetc pushes back its character argument to the unit fd, which must be open for input. After ungetc (‘a’,
fd); ungetc (‘b’, fd); the next two characters to be read from fd will be ‘b’ and then ‘a’. Up to 100 charac-
ters may be pushed back on each file. This subroutine permits a program to read past the end of its input,
and then restore it for the next routine to read. It is impossible to change an external file with ungetc; its
purpose is only for internal communications, most particularly scanf, which is described in section 5. Note
that scanf actually requires only one character of ‘‘unget’’ capability; thus it is possible that future imple-
mentors may change the specification of the ungetc routine.

5. HIGH-LEVEL CHARACTER STREAM ROUTINES

These two routines, printf for output and scanf for input, permit simple translation to and from char-
acter representations of numerical quantities. They also allow generation or interpretation of formatted
lines.
 -4-

PRINTF ( [fd, ] control-string, arg1, arg2, ...)

PRINTF ( [−1, output-string, ] control-string, arg1, arg2, ...)
Printf converts, formats, and prints its arguments under control of the control string. The control string
contains two types of objects: plain characters, which are simply copied to the output stream, and conver-
sion specifications, each of which causes conversion and printing of the next successive argument to printf.

Each conversion specification is introduced by the character ‘%’. Following the ‘%’, there may be:
— an optional minus sign ‘−’ which specifies left adjustment of the converted argument in the
indicated field;
— an optional digit string specifying a minimum field width; if the converted argument has
fewer characters than the field width it will be padded on the left (or right, if the left adjustment
indicator has been given) to make up the field width; the padding character is blank normally
and zero if the field width was specified with a leading zero (note that this does not imply an
octal field width);
— an optional period ‘.’ which serves to separate the field width from the next digit string;
— an optional digit string (the precision) which specifies the maximum number of characters
to be printed from a string, or the number of digits to be printed to the right of the decimal
point of a floating or double number.
— an optional length modifier ′l′ which indicates that the corresponding data item is a long
rather than an int.
— a character which indicates the type of conversion to be applied.

The conversion characters and their meanings are:

d The argument is converted to decimal notation.
o The argument is converted to octal notation.
x The argument is converted to hexadecimal notation.
u The argument is converted to unsigned decimal notation. This is only implemented (or use-
ful) on UNIX.
c The argument is taken to be a single character.
s The argument is taken to be a string and characters from the string are printed until a null
character is reached or until the number of characters indicated by the precision specification is
exhausted.
e The argument is taken to be a float or double and converted to decimal notation of the form
[−] m.nnnnnnE [−] xx where the length of the string of n’s is specified by the precision. The
default precision is 6 and the maximum is 22.
f The argument is taken to be a float or double and converted to decimal notation of the form
[−]mmm.nnnnn where the length of the string of n’s is specified by the precision. The default
precision is 6 and the maximum is 22. Note that the precision does not determine the number
of significant digits printed in f format.
If no recognizable conversion character appears after the ‘%’, that character is printed; thus ‘%’ may be
printed by use of the string ‘‘%%’’.
As an example of printf, the following program fragment
int i, j; float x; char *s;
i = 35; j=2; x= 1.732; s = ‘‘ritchie’’;
printf (‘‘%d %f %s\n’’, i, x, s);
printf (‘‘%o, %4d or %−4d%5.5s\n’’, i, j, j, s);
would print
 -5-

35 1 . 732000 r i t ch i e
043 , 2 or 2 r i t ch

If fd is not specified, output is to unit cout. It is possible to direct output to a string instead of to a
file. This is indicated by −1 as the first argument. The second argument should be a pointer to the string.
Printf will put a terminating ‘\0’ onto the string.

SCANF ( [fd, ] control-string, arg1, arg2, ....)

SCANF ( [−1, input-string, ] control-string, arg1, arg2, ....)
Scanf reads characters, interprets them according to a format, and stores the results in its arguments. It
expects as arguments:
1. An optional file-descriptor or input-string, indicating the source of the input characters; if
omitted, file cin is used;
2. A control string, described below;
3. A set of arguments, each of which must be a pointer, indicating where the converted input
should be stored.
The control string usually contains conversion specifications, which are used to direct interpretation of
input sequences. The control string may contain:
1. Blanks, tabs or newlines, which are ignored.
2. Ordinary characters (not %) which are expected to match the next non-space character of the
input stream (where space characters are defined as blank, tab or newline).
3. Conversion specifications, consisting of the character %, an optional assignment suppressing
character *, an optional numerical maximum field width, and a conversion character.
A conversion specification is used to direct the conversion of the next input field; the result is placed in the
variable pointed to by the corresponding argument, unless assignment suppression was indicated by the *
character. An input field is defined as a string of non-space characters; it extends either to the next space
character or until the field width, if specified, is exhausted.
The conversion character indicates the interpretation of the input field; the corresponding pointer
argument must usually be of a restricted type. Pointers, rather than variable names, are required by the
‘‘call-by-value’’ semantics of the C language. The following conversion characters are legal:
% indicates that a single % character is expected in the input stream at this point; no assign-
ment is done.
d indicates that a decimal integer is expected in the input stream; the corresponding argument
should be an integer pointer.
o indicates that an octal integer is expected in the input stream; the corresponding argument
should be a integer pointer.
x indicates that a hexadecimal integer is expected in the input stream; the corresponding argu-
ment should be an integer pointer.
s indicates that a character string is expected; the corresponding argument should be a charac-
ter pointer pointing to an array of characters large enough to accept the string and a terminating
‘\0’, which will be added. The input field is terminated by a space character or a newline.
c indicates that a single character is expected; the corresponding argument should be a charac-
ter pointer; the next input character is placed at the indicated spot. The normal skip over space
characters is suppressed in this case; to read the next non-space character, try %1s.
e or f indicates that a floating point number is expected in the input stream; the next field is con-
verted accordingly and stored through the corresponding argument, which should be a pointer
to a float. The input format for floats is a string of numbers possibly containing a decimal
point, followed by an optional exponent field containing an E or e followed by a possibly
signed integer.
[ indicates a string not to be delimited by space characters. The left bracket is followed by a
set of characters and a right bracket; the characters between the brackets define a set of charac-
ters making up the string. If the first character is not circumflex ( ˆ ), the input field is all
 -6-

characters until the first character not in the set between the brackets; if the first character after
the left bracket is ˆ, the input field is all characters until the first character which is in the
remaining set of characters between the brackets. The corresponding argument must point to a
character array.
The conversion characters d, o and x may be preceded by l to indicate that a pointer to long rather
than int is expected. Similarly, the conversion characters e or f may be preceded by l to indicate that a
pointer to double rather than float is in the argument list. The character h will function similarly in the
future to indicate short data items.
For example, the call
int i; float x; char name[50];
scanf ( ‘‘%d%f%s’’, &i, &x, name);
with the input line
25 54.32E−1 thompson
will assign to i the value 25, x the value 5.432, and name will contain ‘‘thompson\0’’. Or,
int i; float x; char name[50];
scanf (‘‘%2d%f%*d%[1234567890]’’, &i, &x, name);
with input
56789 0123 56a72
will assign 56 to i, 789.0 to x, skip ‘‘0123’’, and place the string ‘‘56\0’’ in name. The next call to cgetc
will return ‘a’.
Scanf returns as its value the number of successfully matched and assigned input items. This can be
used to decide how many input items were found. On end of file, −1 is returned; note that this is different
from 0, which means that the next input character does not match what you called for in the control string.
Scanf, if given a first argument of −1, will scan a string in memory given as the second argument. For
example, if you want to read up to four numbers from an input line and find out how many there were, you
could try
int a[4], amax;
char line[100];
amax = scanf (−1, gets(line), ‘‘%d%d%d%d’’, &a[0], &a[1], &a[2], &a[3]);

6. BINARY STREAM ROUTINES

These routines write binary data, not translated to printable characters. They are normally efficient
but do not produce files that can be printed or easily interpreted. No special information is added to the
records and thus they can be handled by other programming systems if you make the departure from porta-
bility required to tell the other system how big a C item (integer, float, structure, etc.) really is in machine
units.

COPEN (name, direction, ‘‘i’’)

When copen is called with a third argument as above, a binary stream filedescriptor is returned. Such a file
descriptor is required for the remaining subroutines in this section, and may not be used with the routines in
the preceding two sections. The first two arguments operate exactly as described in section 3; further
details are given in section 7. An ordinary file descriptor may be used for binary I-O, but binary and char-
acter I-O may not be mixed unless cflush is called at each switch to binary I-O. The third argument to
copen is ignored.

CWRITE ( ptr, sizeof(*ptr), nitems, fd )

Cwrite writes nitems of data beginning at ptr on file fd. Cwrite writes blocks of binary information, not
translated to printable form, on a file. It is intended for machine-oriented bulk storage of intermediate data.
Any kind of data may be written with this command, but only the corresponding cread should be expected
to make any sense of it on return. The first argument is a pointer to the beginning of a vector of any kind of
data. The second argument tells cwrite how big the items are. The third argument specifies the number of
the items to be written; the fourth indicates where.
 -7-

CREAD ( ptr, sizeof(*ptr), nitems, fd )

Cread reads up to nitems of data from file fd into a buffer beginning at ptr. Cread returns the number of
items read.
The returned number of items will be equal to the number requested by nitems except for reading certain
devices (e.g. the teletype or magnetic tape) or reading the final bytes of a disk file.
Again, the second argument indicates the size of the data items being read.

CCLOSE (fd)
The same description applies as for character-stream files.

7. OTHER PORTABLE ROUTINES

REW (fd)
Rewinds unit fd. Buffers are emptied properly and the file is left open.

SYSTEM (string)
The given string is executed as if it were typed at the terminal.

NARGS ( )
A subroutine can call this function to try to find out how many arguments it was called with. Normally,
nargs() returns the number of arguments plus 3 for every float or double argument and plus one for every
long argument. If the new UNIX feature of separated instruction and data space areas is used, nargs()
doesn’t work at all.

CALLOC (n, sizeof(object))

Calloc returns a pointer to new storage, allocated in space obtained from the operating system. The space
obtained is well enough aligned for any use, i.e. for a double-precision number. Enough space to store n
objects of the size indicated by the second argument is provided. The sizeof is executed at compile time; it
is not in the library. A returned value of −1 indicates failure to obtain space.

CFREE (ptr, n, sizeof(*ptr))

Cfree returns to the operating system memory starting at ptr and extending for n units of the size given by
the third argument. The space should have been obtained through calloc. On UNIX you can only return the
exact amount of space obtained by calloc; the second and third arguments are ignored.

FTOA (floating-number, char-string, precision, format )

Ftoa (floating to ASCII conversion) converts floating point numbers to character strings. The format argu-
ment should be either ‘f’ or ‘e’; ‘e’ is default. See the explanation of printf in section 5 for a description of
the result.

ATOF (char-string)
Returns a floating value equal to the value of the ASCII character string argument, interpreted as a decimal
floating point number.

TMPNAM (str)
This routine places in the character array expected as its argument a string which is legal to use as a file
name and which is guaranteed to be unique among all jobs executing on the computer at the same time. It
is thus appropriate for use as a temporary file name, although the user may wish to move it into an appro-
priate directory. The value of the function is the address of the string.
 -8-

ABORT (code)
Causes your program to terminate abnormally, which typically results in a dump by the operating system.

INTSS ()
This routine tells you whether you are running in foreground or background. The definition of ‘‘fore-
ground’’ is that the standard input is the terminal.

WDLENG ( )
This returns 16 on UNIX. C users should be aware that the preprocessor normally provides a defined sym-
bol suitable for distinguishing the local system; thus on UNIX the symbol unix is defined before starting to
compile your program.
 UNIX Summary (DRAFT)

A. Hardware

UNIX runs on a DEC PDP11/40*, 11/45 or 11/70 with at least the following equipment:
48K to 124K words of managed memory: parity not used,
disk: RP03, RP04, RK05(preferably 2) or equivalent,
console typewriter,
clock: KW11-L or KW11-P,
extended instruction set KE11-E, on 11/40 only.
The system is normally distributed on 9-track tape or RK05 packs.
The following equipment is strongly recommended:
communications controllers such as DL11, DC11 or DH11,
full duplex 96-character ASCII terminals,
9-track tape, or extra disk for system backup.
The minimum memory and disk space specified is enough to run and maintain UNIX. More will be needed to keep
all source on line, or to handle a large number of users, big data bases, diversified complements of devices, or large
programs. UNIX does swapping and reentrant sharing of user code to minimize main memory requirements. The
resident code of UNIX occupies 20-22K depending on configuration.

An 11/40 is not advisable for heavy floating point work, as UNIX on this hardware uses interpreted 11/45 floating
point.

B. Software

All the programs available as UNIX commands are listed. Every command, including all options, is issued as just
one line, unless specifically noted as ‘‘interactive’’. Interactive programs can be made to run from a prepared script
simply by redirecting input.

File processing commands that go from standard input to standard output are noted as usable as filters. The piping
facility of the Shell may be used to connect filters directly to the input or output of other programs.

Commercially distributed UNIX normally excludes software listed in Section 5, ‘‘Typesetting.’’ Source code is
included except as noted.

1 Basic Software

This package includes time-sharing operating system with utilities, machine language assembler and the compiler
for the programming language C—enough software to write and run new applications and to maintain or modify
UNIX itself.

1.1 Operating System

UNIX The basic resident code on which everything else depends. Supports the system calls, and main-
tains the file system. A general description of UNIX design philosophy and system facilities
appeared in the Communications of the ACM, July, 1974. Further capabilities include:
• Automatically supported reentrant code.
• Separate instruction and data spaces on 11/45 and 11/70.
• ‘‘Group’’ access permissions allow cooperative projects, with overlapping memberships.
• Timer-interrupt sampling and interprocess monitoring for debugging and measurement.

Manual Printed manuals for UNIX and all its software, except where other manuals exist.
• UNIX Programmer’s Manual.
• The UNIX Time-Sharing System, reprint setting forth design principles.
• UNIX for Beginners.

1
 UNIX Summary

• The UNIX I-O System.

• On the Security of UNIX.

(DEV) All I/O is logically synchronous. Normally, invisible buffering makes all physical record structure
transparent and exploits the hardware’s ability to do overlapped I/O. Unbuffered physical record
I/O is available for unusual applications. Drivers for these devices are available; others can be eas-
ily written:
• Asynchronous interfaces: DC11, DH11, DL11. Support for most common ASCII terminals.
• Synchronous interface: DP11.
• Automatic calling unit interface: DN11.
• Line printer: LP11.
• Magnetic tape: TU10 and TU16.
• DECtape*: TC11.
• Paper tape: PC11.
• Fixed head disk: RS11, RS03 and RS04.
• Pack type disk: RP03 and RP04, one or more logical devices per physical device, minimum-
latency seek scheduling.
• Cartridge-type disk: RK05, one or more physical devices per logical device.
• Null device.
• Physical memory of PDP11, or mapped memory in resident system.
• Phototypesetter: Graphic Systems System/1 through DR11C.
• Voice synthesizer: VOTRAX* through DC11. Includes TOUCH-TONE® input.

BOOT Procedures to get UNIX up on a naked machine.

Manual Setting up UNIX.

MKCONF Tailor device-dependent system code to hardware configuration. Other changes, such as optimal
assignment of directories to devices, inclusion of floating point simulator, or installation of device
names in file system can then be made at leisure. (As distributed, UNIX can be brought up directly
on any acceptable CPU with any acceptable disk, any sufficient amount of core and either clock.)

Manual Printed manual on setting up UNIX.

1.2 User Access Control

LOGIN Sign on as a new user.

• Verify password and establish user’s individual and group (project) identity.
• Adapt to characteristics of terminal.
• Establish working directory.
• Announce presence of mail (from MAIL).
• Publish message of the day.
• Start command interpreter or other initial program.

PASSWD Change a password.

• User can change his own password.
• Passwords are kept encrypted for better security.

NEWGRP Change working group (project). Protects against changes to unauthorized projects.

1.3 File Manipulation

CAT Concatenate one or more files onto standard output. Particularly used for unadorned printing, for
inserting data into a pipeline, and for buffering output that comes in dribs and drabs.
• Usable as filter.

2
 UNIX Summary

CP Copy one file to another. Works on any file without distinction as to contents.

PR Print files with title, date, and page number on every page.
• Multicolumn output.
• Parallel column merge of several files.
• Usable as a filter.

OPR Off line print. Spools arbitrary files to the line printer.
• Usable as a filter.

SPLIT Split a large file into more manageable pieces. Is occasionally necessary for editing (ED).

ED Interactive context editor. Can work on single lines, blocks of lines, or all pattern-selected lines in
a given range.
• Finds lines by number or pattern.
• Random access to lines.
• Add, delete, change, copy or move lines.
• Permute or split contents of a line.
• Replace one or all instances of a pattern within a line.
• Combine or split files.
• Escape to Shell (UNIX command language) during editing.
• Patterns may include:
specified characters,
don’t care characters,
choices among characters,
repetitions of above,
beginning of line,
end of line.
• All operations may be done globally on every pattern-selected line in a given range.

Manual Introductory manual for ED.

DD Physical file format translator, for exchanging data with foreign systems, especially OS/360.

STTY Sets up options for optimal control of a terminal. In so far as they are deducible from the input,
these options are set automatically by LOGIN.
• Half vs. full duplex.
• Carriage return+line feed vs. newline.
• Interpretation of tabs.
• Parity.
• Mapping of upper case to lower.
• Raw vs. edited input.
• Delays for tabs, newlines and carriage returns.

1.4 Manipulation of Directories and File Names

RM Remove a file. Only the name goes away if any other names are linked to the file.

LN ‘‘Link’’ another name (alias) to an existing file.

MV Move a file.
• Used for renaming files or directories.

CHMOD Change permissions on one or more files. Executable by files’ owner.

3
 UNIX Summary

CHOWN Change owner of one or more files.

CHGRP Change group (project) to which a file belongs.

MKDIR Make a new directory.

RMDIR Remove a directory.

CHDIR Change working directory.

FIND Prowl the directory hierarchy finding every file that meets specified criteria.
• These criteria are understood:
spelling of name matches a given pattern,
creation date in given range,
date of last use in given range,
permissions,
owner,
characteristics of device files,
boolean combinations of above.
• Any directory may be considered to be the root.
• Specified commands may be performed on every file found.

DSW Interactively step through a directory, deleting or keeping files.

1.5 Running of Programs

SH The Shell, or command language interpreter. Provides ‘‘background’’ and macro capability when
run with a file of commands as input.
• Any executable object file is automatically a command.
• Redirect standard input or standard output.
• Operators to compose compound commands:
‘;’ for sequential execution,
‘|’ for functional composition with output of one command taken directly as input to another
running simultaneously,
‘&’ for asynchronous operation,
parentheses for grouping.
• Substitutable arguments.
• Construction of argument lists from all file names satisfying specified patterns.
• Collects command usage statistics.

IF A conditional statement for Shell programs.

• String comparison.
• Querying file accessibility.

GOTO A ‘‘go-to’’ statement for Shell programs.

WAIT Wait for termination of asynchronously running processes.

EXIT Terminate a Shell program. Useful with IF.

ECHO Print remainder of command line. Useful for diagnostic or prompting data in Shell programs, or
for inserting data into a pipeline.

SLEEP Suspend execution for a specified time.

4
 UNIX Summary

NOHUP Run a command immune to hanging up the terminal.

NICE Run a command in low (or high) priority.

KILL Terminate named processes.

CRON A table of actions to be taken at specified times.

• Actions are arbitrary Shell (SH) scripts.
• Times are conjunctions of month, day of month, day of week, hour and minute. Ranges are
specifiable for each.

TEE Pass data between processes and divert a copy into a file. Used as a filter.

1.6 Status Inquiries

LS List the names of one, several, or all files in one or more directories.
• Alphabetic or temporal sorting, up or down.
• Optional information: size, owner, group, date last modified, date last accessed, permissions, i-
node number.

FILE Tries to determine what kind of information is in a file by consulting the file system index and by
reading the file itself.

DATE System date routine. Has considerable knowledge of calendric and horological peculiarities.
• Print present date, day of week, local time.
• May set UNIX’s idea of date and time.

DF Report amount of free space on file system devices.

DU Print a summary of total space occupied by all files in a hierarchy.

WHO Tell who’s on the system.

• List of presently logged in users, ports and times on.
• Optional history of all logins and logouts.

PS Report on all active processes attached to a terminal.

• Gives all commands being executed.
• Can also report on other terminals.
• Extended status information available:
state and scheduling info,
priority,
attached terminal,
what it’s waiting for,
size.

TTY Find name of your terminal.

PWD Print name of your working directory.

PFE Print type of last floating exception.

1.7 Backup and Maintenance

MOUNT Attach a device containing a file system to the tree of directories. Protects against nonsense
arrangements.

5
 UNIX Summary

UMOUNT Remove the file system contained on a device from the tree of directories. Protects against remov-
ing a busy device.

MKFS Make a new file system on a device.

MKNOD Make an i-node (file system entry) for a special file. Special files are physical devices, virtual
devices, physical memory, etc.

TP Manage file archives on magnetic tape or DEC tape.

• Collect files into an archive.
• Update DECtape archive by date.
• Replace or delete DECtape files.
• Table of contents.
• Retrieve from archive.

DUMP Dump the file system stored on a specified device, selectively by date, or indiscriminately.

RESTOR Restore a dumped file system, or selectively retrieve parts thereof.

SU Temporarily become the super user with all the rights and privileges thereof. Requires a password.

DCHECK Check consistency of file system.

ICHECK • Gross statistics:
NCHECK number of files,
number of directories,
number of special files,
spaced used,
space free.
• Report of duplicate use of space.
• Retrieval of lost space.
• Report of inaccessible files.
• Check consistency of directories.
• List names of all files.

CLRI Peremptorily expunge a file and its space from a file system. Used in putting damaged file systems
together again.

SYNC Force all outstanding I/O on the system to completion. Used to shut down gracefully.

1.8 Accounting

These routines use floating point. The timing information on which the reports are based can be manually cleared or
shut off completely.

AC Publish cumulative connect time report.

• Connect time by user or by day.
• For all users or for selected users.

SA Publish Shell accounting report. Gives usage information on each command executed.
• Number of times used.
• Total system time, user time and elapsed time.
• Optional averages and percentages.
• Sorting on various fields.

6
 UNIX Summary

1.9 Inter-user Communication

MAIL Mail a message to one or more users. Also used to read and dispose of incoming mail. The pres-
ence of mail is announced by LOGIN.

WRITE Establish direct terminal communication with another user.

WALL Write to all users.

MESG Inhibit receipt of messages from WRITE and WALL.

1.10 Basic Program Development Package

A kit of fundamental programming tools. Some of these utilities are used as integral parts of the higher level lan-
guages described below.

AR Archive and library maintainer. Combines several files into one for housekeeping efficiency.
Archive files are used by the link editor LD as libraries.
• Create new archive.
• Update archive by date.
• Replace or delete files.
• Table of contents.
• Retrieve from archive.

AS Assembler. Similar to PAL-11, but different in detail.

• Creates object program consisting of
code, possibly read-only,
initialized data or read-write code,
uninitialized data.
• Relocatable object code is directly executable without further transformation.
• Object code normally includes a symbol table.
• Combines source files.
• Local labels.
• Conditional assembly.
• ‘‘Conditional jump’’ instructions become branches or branches plus jumps depending on dis-
tance.

Manual Printed manual for the assembly language.

Library The basic run-time library. These routines are used freely by all system software.
• Formatted writing on standard output.
• Time conversions.
• Convert integer and floating numbers to ASCII and vice versa.
• Elementary functions: sin, cos, log, exp, atan, sqrt, gamma.
• Password encryption.
• Quicksort.
• Buffered character-by-character I/O.
• Random number generator.
• Floating point interpreter for 11/40’s and non-floating point machines.

(LIBP) An elaborated I/O library.

• Formatted input and output.
• Ability to put characters back into input streams.

7
 UNIX Summary

Manual Printed manual for LIBP.

DB Interactive post-mortem debugger. Works on core dump files, such as are produced by all program
aborts, on object files, or on any arbitrary file.
• Symbolic addressing of files that have symbol tables.
• Octal, decimal or ASCII output.
• Symbolic disassembly.
• Octal or decimal patching.

OD Dump any file.

• Output options include:
octal or decimal by words,
octal by bytes,
ASCII,
opcodes,
hexadecimal,
any combination of the above.
• Range of dumping is controllable.

LD Link edit. Combine relocatable object files. Insert required routines from specified libraries.
• Resulting code may be sharable.
• Resulting code may have separate instruction and data spaces.

NM Print the namelist (symbol table) of an object program. Provides control over the style and order of
names that are printed.

SIZE Report the core requirements of one or more object files.

STRIP Remove the relocation and symbol table information from an object file to save space.

TIME Run a command and report timing information on it.

PROF Construct a profile of time spent per routine from statistics gathered by time-sampling the execu-
tion of a program. Uses floating point.
• Subroutine call frequency and average times for C programs.

1.11 The Programming Language ‘‘C’’

CC Compile and/or link edits programs in the C language. The UNIX operating system, most of the
subsystems and C itself are written in C.
• Full general purpose language designed for structured programming.
• Data types:
character,
integer,
float,
double,
pointers to all types,
arrays of all types,
structures of all types,
functions returning all types.
• Operations intended to give access to full machine facility, including to-memory operations and
data-sensitive pointer arithmetic.
• Macro preprocessor for parameterized code and inclusion of standard files.
• All procedures recursive, with parameters by value.
• Natural coercions.

8
 UNIX Summary

• True compiled object code capitalizing on addressing capability of the PDP11.

• Runtime library gives access to all system facilities.

Manuals Printed manual and tutorial for the C language.

CDB An interactive debugger tailored for use with C.

• Usable in real time or post-mortem.
• The debugger is a completely separate process from the debuggee. No debugging code is loaded
with debuggee.
• Prints all kinds of data in natural notation:
character,
integer (octal and decimal),
float,
double,
machine instructions (disassembled).
• Stack trace and fault identification.
• Breakpoint tracing.

2 Other Languages

2.1 FORTRAN

FC Compile and/or link-edit FORTRAN IV programs. Object code is ‘‘threaded’’. Relies heavily on
floating point.
• Idiosyncracies:
free form, lower-case source code,
no arithmetic statement functions,
unformatted I/O requires record lengths agree,
no BACKSPACE,
no P FORMAT control on input.
• Handles mixed-mode arithmetic, general subscripts and general DO limits.
• 32-bit integer arithmetic.
• Free format numeric input.
• Understands these nonstandard specifications:
LOGICAL*1, *2, *4,
INTEGER*2, *4,
REAL*4, *8,
COMPLEX*8, *16,
IMPLICIT.

RC ‘‘Ratfor’’, a preprocessor that adds rational control structure à la C to FORTRAN.

• Else, for, while, repeat...until statements.
• Symbolic constants.
• File insertion.
• Compound statements.
• Can produce genuine FORTRAN to carry away.

Manual Printed manual for Ratfor.

2.2 Other Algorithmic Languages

BAS An interpreter, similar in style to BASIC, that allows immediate execution of unnumbered state-
ments, or deferred execution of numbered statements.
• Statements include:
comment,

9
 UNIX Summary

dump,
for...next,
goto,
if...else...fi,
list,
print,
prompt,
return,
run,
save.
• All calculations double precision.
• Recursive function defining and calling.
• Builtin functions include log, exp, sin, cos, atn, int, sqr, abs, rnd.
• Escape to ED for complex program editing.
• Usable as a filter.

DC Programmable reverse Polish desk calculator. Has named storage locations as well as conventional
stack for holding integers or programs.
• Unlimited precision decimal arithmetic.
• Appropriate treatment of decimal fractions.
• Arbitrary input and output radices, in particular binary, octal, decimal and hexadecimal.
• Operators include:
+−*/
remainder, power, square root,
load, store, duplicate, clear,
print, enter program text, execute.
• Usable as a filter.

BC A C-like interface to the desk calculator DC.

• All the capabilities of DC with a high-level syntax.
• Arrays and recursive functions.
• Immediate evaluation of expressions and evaluation of functions upon call.
• Arbitrary precision elementary functions: exp, sin, cos, atan, Jn.
• Go-to-less programming.
• Usable as a filter.

Manual Printed manual for BC.

SNO An interpreter very similar to SNOBOL 3.

• Limitations:
function definitions are static,
pattern matches are always anchored,
no built-in functions.
• Usable as a filter.

Manual Reprint of basic article.

2.3 Macroprocessing

M6 A general purpose macroprocessor.

• Stream-oriented, recognizes macros anywhere in text.
• Integer arithmetic.
• Usable as a filter.

10
 UNIX Summary

Manual Printed manual for M6.

2.4 Compiler-compilers

TMG A classical top-down compiler-compiler language. Provides a formalism for syntax-directed trans-
lation. Produces driving tables to be loaded with a standard interpreter.
• Resulting compilers can have arbitrary tables kept in paged secondary store.
• Integer arithmetic capability.
• Syntactic function capability (similar to ALGOL 68 metaproductions).

Manual Printed manual for the TMG compiler-writing system.

YACC An LR(1)-based compiler writing system. During execution of resulting parsers, arbitrary C-lan-
guage, Ratfor or FORTRAN functions may be called to do code generation or semantic actions.
• BNF syntax specifications.
• Handles precedence relations.
• Accepts formally ambiguous grammars with non-BNF resolution rules.
• Optimizes space taken by driving tables.

Manual Printed manual for the YACC compiler-writing system.

3 Word Processing

ROFF A typesetting program for terminals. Easy for nontechnical people to learn, and good for most
ordinary kinds of documents. Input consists of data lines intermixed with control lines, such as
.sp 2 insert two lines of space
.ce center the next line
• Justification of either or both margins.
• Automatic hyphenation.
• Generalized running heads and feet, with even-odd page capability, numbering, etc.
• Definable macros for frequently used control sequences (no substitutable arguments).
• All 4 margins and page size dynamically adjustable.
• Hanging indents and one-line indents.
• Absolute and relative parameter settings.
• Optional legal-style numbering of output lines.
• Multiple file capability.

CREF Make cross-reference listings of a collection of files. Each symbol is listed together with file, line
number, and text of each line in which it occurs.
• Assembler or C language.
• Gathering or suppressing references to selected symbols.
• Last symbol defined may replace line number.
• Various ways to sort output available.
• Selective print of uniquely occurring symbols.

INDEX Make cross-reference indexes of English text.

• Handles lists of specific index terms or excluded terms.
• Handles words hyphenated across lines.
• Understands TROFF and NROFF output, so can gather references according to final pagination.
• Output capabilities like CREF.
• Frequency counts.

FORM Form letter generator. Remembers any number of forms and stock phrases such as names and
addresses. Output usually intended to be ROFFed.
• Anything that is typed in can be remembered for later use.

11
 UNIX Summary

• Runs interactively, querying only for those items that are not in its memory.
• Any item may call for the inclusion of other items. For example, full name, address, first name,
title, etc., may be separately retrieved from one name key.

FED Editor for the memory used by FORM. Extract any item, turn it over to context editor ED for edit-
ing, and put it back when done.
• List names of selected items.
• Print contents of selected item.

SORT Sort or merge ASCII files line-by-line.

• Sort up or down.
• Sort lexicographically or on numeric key.
• Multiple keys located by delimiters or by character position.
• May sort upper case together with lower into dictionary order.
• Usable as a filter.

UNIQ Collapse successive duplicate lines in a file into one line.

• Publishes lines that were originally unique, duplicated, or both.
• May give redundancy count for each line.
• Usable as a filter.

TR Do one-to-one character translation according to an arbitrary code.

• May coalesce selected repeated characters.
• May delete selected characters.
• Usable as a filter.

DIFF Report line changes, additions and deletions necessary to bring two files into agreement.
• May produce an editor script to convert one file into another.

COMM Identify common lines in two sorted files. Output in up to 3 columns shows lines present in first
file only, present in both, and/or present in second only.

CMP Compare two files and report disagreeing bytes.

GREP Print all lines in a file that satisfy a pattern of the kind used in the editor ED.
• May print all lines that fail to match.
• May print count of hits.
• Usable as a filter.

WC Count the lines and ‘‘words’’ (blank-separated strings) in a file.

• Usable as a filter.

TYPO Find typographical errors. Statistically analyzes all the words in a text, weeds out several thousand
familiar ones, and publishes the rest sorted so that the most improbably spelled ones tend to come
to the top of the list.

GSI Simulate Model 37 Teletype facilities on GSI-300, DASI and other Diablo-mechanism terminals.
• Gives half-line and reverse platen motions.
• Approximates Greek letters and other special characters by overstriking.
• Usable as a filter.

COL Canonicalize files with reverse line feeds for one-pass printing.
• Usable as a filter.

12
 UNIX Summary

4 Novelties

Source code for game-playing programs is not distributed.

SPEAK Driver for Vocal Interface’s VOTRAX speech synthesizer. Reads input text and utters it.
• Associative memory allows pronunciation rules for whole words or word fragments to be added,
changed, deleted or queried.
• Can use different memories for different languages.
• Usable as a filter to make the output of any other program audible.

CHESS This chess-playing program scored 1-2-1 and 3-0-1 in the 1973 and 1974 Computer Chess Cham-
pionships.

BJ A blackjack dealer.

CUBIC An accomplished player of 4×4×4 tic-tac-toe.

MOO A fascinating number-guessing game.

CAL Prints a calendar of specified month and year.

UNITS Converts amounts between different scales of measurement. Knows hundreds of units. For exam-
ple, how many km/sec is a parsec/megayear?

TTT A tic-tac-toe program that learns. It never makes the same mistake twice.

QUIZ Tests your knowledge of Shakespeare, Presidents, capitals, etc.

WUMP Hunt the wumpus, thrilling search in a dangerous cave.

5 Typesetting

This software is distributed separately as an enhancement to UNIX.

5.1 Formatters

High programming skill is required to exploit the formatting capabilities of these programs, although unskilled per-
sonnel can easily be trained to enter documents according to canned formats. Terminal-oriented and typesetter-ori-
ented formatters are sufficiently compatible that it is usually possible to define interchangeable formats.

NROFF Advanced typesetting for terminals. Style similar to ROFF, but capable of much more elaborate
feats of formatting, at a price in ease of use.
• All ROFF capabilities available or definable.
• Completely definable page format keyed to dynamically planted ‘‘interrupts’’ at specified lines.
• Maintains several separately definable typesetting environments (e.g. one for body text, one for
footnotes, and one for unusually elaborate headings).
• Arbitrary number of output pools can be combined at will.
• Macros with substitutable arguments, and macros invocable in mid-line.
• Computation and printing of numerical quantities.
• Conditional execution of macros.
• Tabular layout facility.
• Multicolumn output on terminals capable of reverse line feed, or through the postprocessor COL.
• Usable as a filter

13
 UNIX Summary

Manual Printed manual for NROFF.

TROFF Advanced phototypesetting for the Graphic Systems System/1. Provides facilities like NROFF,
augmented as follows. This Summary was typeset by TROFF.
• Vocabulary of several 102-character fonts (4 simultaneously) in 15 sizes.
• Positions expressible in inches, centimeters, ems, points, machine units or arithmetic combina-
tions thereof.
• Access to character-width computation for unusually difficult layout problems.
• Overstrikes, built-up brackets, horizontal and vertical line drawing.
• Dynamic relative or absolute positioning and size selection, globally or at the character level.
• Terminal output for rough sampling of the product, usually needs a wide platen. Not a substitute
for NROFF.
• Usable as a filter.

Manuals Printed manual and tutorial for TROFF.

EQN A mathematical typesetting preprocessor for TROFF. Translates easily readable formulas, either
in-line or displayed, into detailed typesetting instructions. Formulas are written in a style like this:
sigma sup 2 ˜=˜ 1 over N sum from i=1 to N ( x sub i - x bar ) sup 2
which produces this:
σ2 = N Σ
1 i =N 1 (xi − x ) 2
• Automatic calculation of size changes for subscripts, sub-subscripts, etc.
• Full vocabulary of Greek letters, such as ‘gamma’, ‘GAMMA’.
• Automatic calculation of large bracket sizes.
• Vertical ‘‘piling’’ of formulae for matrices, conditional alternatives, etc.
• Integrals, sums, etc. with arbitrarily complex limits.
• Diacriticals: dots, double dots, hats, bars.
• Easily learned by nonprogrammers and mathematical typists.
• Usable as a filter.

Manual Printed manual for EQN.

NEQN A mathematical typesetting preprocessor for NROFF. Prepares formulas for display on Model 37
Teletypes with half-line functions and 128-character font.
• For Diablo-mechanism terminals, filter output through GSI.
• Same facilities as EQN within graphical capability of terminal.

TBL A preprocessor for NROFF/TROFF that translates simple descriptions of table layouts and contents
into detailed typesetting instructions.
• Computes column widths.
• Handles left- and right-justified columns, centered columns and decimal-point alignment.
• Places column titles.

MS A standardized manuscript layout for use with NROFF/TROFF.

• Page numbers and draft dates.
• Cover sheet and title page.
• Automatically numbered subheads.
• Footnotes.
• Single or double column.
• Paragraphing, display and indentation.
• Numbered equations.

14
 UNIX Summary

5.2 UNIX Programmer’s Manual

MAN Print specified manual section on your terminal.

Manual Machine-readable version of the UNIX Programmer’s Manual.

• System overview.
• All commands.
• All system calls.
• All subroutines in assembler, C and FORTRAN libraries.
• All devices and other special files.
• Formats of file system and kinds of files known to system software.
• Boot procedures.

May, 1975

* DEC, PDP and DECtape are registered trademarks of Digital Equipment Corporation. VOTRAX is a registered
trademark of Vocal Interface Division, Federal Screw Works.

15