path: root/contrib/binutils/etc/configure.texi

\input texinfo    @c -*-texinfo-*-
@setfilename configure.info
@settitle Cygnus configure

@synindex ky cp

@setchapternewpage odd

@ifinfo
@format
START-INFO-DIR-ENTRY
* configure: (configure).        Cygnus configure.
END-INFO-DIR-ENTRY
@end format
@end ifinfo

@ifinfo
This document describes the Cygnus Support version of @code{configure}.

Copyright (C) 1991, 1992, 1993 Cygnus Support
Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.

@ignore
Permission is granted to process this file through TeX and print the
results, provided the printed document carries copying permission
notice identical to this one except for the removal of this paragraph
(this paragraph not being relevant to the printed manual).
@end ignore

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.

Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation approved
by Cygnus Support.
@end ifinfo

@c We should not distribute texinfo files with smallbook enabled.
@c @smallbook
@finalout
@titlepage
@title Cygnus configure
@author K. Richard Pixley
@author Cygnus Support
@page
@cindex copyleft

@vskip 0pt plus 1filll
Edited January, 1993, by Jeffrey Osier, Cygnus Support.

Copyright @copyright{} 1991, 1992, 1993 Cygnus Support

Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.

Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation approved
by Cygnus Support.
@end titlepage

@c ---------------------------------------------------------------------
@ifinfo
@node Top
@top Cygnus configure

This file documents the configuration system used and distributed by
Cygnus Support.

@menu
* What configure does::    What configure does
* Invoking configure::     Invoking configure---basic usage
* Using configure::        More than you ever wanted to know
* Porting::                How to use configure with new programs
* Variables Index::
* Concept Index::
@end menu
@end ifinfo

@c ---------------------------------------------------------------------
@node What configure does
@chapter What @code{configure} does
@cindex Introduction
@cindex Overview
@cindex What @code{configure} does
@kindex Cygnus Support Developer's Kit

This manual documents Cygnus @code{configure}, a program which helps to
automate much of the setup activity associated with building large suites of
programs, such the Cygnus Support Developer's Kit.  This manual is therefore
geared toward readers who are likely to face the problem of configuring
software in source form before compiling and installing it.  We assume you are
an experienced programmer or system administrator.
@ifinfo
For further background on this topic, see @ref{Some Basic Terms, , Apologia
Configure, cfg-paper, On Configuring Development Tools}, by K. Richard
Pixley.
@end ifinfo
@iftex
For further background on this topic, see @cite{On Configuring Development
Tools} by K. Richard Pixley.
@end iftex

When @code{configure} runs, it does the following things:

@table @emph
@item @bullet{} creates build directories
@vindex srcdir
@cindex @code{srcdir}
@cindex Build directories
When you run @code{configure} with the @samp{--srcdir} option, it uses the
current directory as the @dfn{build directory}, creating under it a directory
tree that parallels the directory structure of the source directory.  If you
don't specify a @samp{srcdir}, @code{configure} first assumes that the source
code you wish to configure is in your current directory; if it finds no
@file{configure.in} input file there, it searches in the directory
@code{configure} itself lies in.  (For details, see @ref{Build directories, ,
Build directories}.)

@item @bullet{} generates @file{Makefile}
@cindex @code{Makefile} generation
A @file{Makefile} template from the source directory, usually called
@file{Makefile.in}, is copied to an output file in the build directory which is
most often named @file{Makefile}.  @code{configure} places definitions for a
number of standard @file{Makefile} macros at the beginning of the output file.
If @w{@samp{--prefix=@var{dir}}} or @w{@samp{--exec_prefix=@var{dir}}} are
specified on the @code{configure} command line, corresponding @file{Makefile}
variables are set accordingly.  If host, target, or site-specific
@file{Makefile} fragments exist, these are inserted into the output file.  (For
details, see @ref{Makefile generation, , @code{Makefile} generation}.)

@item @bullet{} generates @file{.gdbinit}
@cindex @code{.gdbinit}
If the source directory contains a @file{.gdbinit} file and the build directory
is not the same as the source directory, a @file{.gdbinit} file is created in
the build directory.  This @file{.gdbinit} file contains commands which allow
the source directory to be read when debugging with the @sc{gnu} debugger,
@code{gdb}.  (@xref{Command Files, , Command Files, gdb, Debugging With GDB}.)

@item @bullet{} makes symbolic links
@cindex Symbolic links
Most build directories require that some symbolic links with generic names are
built pointing to specific files in the source directory.  If the system where
@code{configure} runs cannot support symbolic links, hard links are used
instead.  (For details, see @ref{configure.in, , The @code{configure.in} input
file}.)

@item @bullet{} generates @file{config.status}
@cindex @code{config.status}
@code{configure} creates a shell script named @file{config.status} in the build
directory.  This shell script, when run from the build directory (usually from
within a @file{Makefile}), will reconfigure the build directory (but not its
subdirectories).  This is most often used to have a @file{Makefile} update
itself automatically if a new source directory is available.

@item @bullet{} calls itself recursively
@cindex Recursion
If the source directory has subdirectories that should also be configured,
@code{configure} is called for each.  
@end table

@c ---------------------------------------------------------------------
@node Invoking configure
@chapter Invoking @code{configure}
@cindex Invoking @code{configure}
@cindex Usage

Cygnus @code{configure} is a shell script which resides in a source tree.  The
usual way to invoke @code{configure} is from the shell, as follows:

@cindex Example session
@example
eg$ ./configure @var{hosttype}
@end example

@noindent
This prepares the source in the current directory (@file{.}) to be
compiled for a @var{hosttype} environment.  It assumes that you wish to
build programs and files in the default @dfn{build directory} (also the
current directory, @file{.}).  If you do not specify a value for
@var{hosttype}, Cygnus @code{configure} will attempt to discover this
information by itself (@pxref{config.guess, , Determining system
information}).  For information on @var{hosttype} environments,
@xref{Host, , Host}.

All @sc{gnu} software is packaged with one or more @code{configure} script(s)
(@pxref{Configuration, , How Configuration Should Work, standards, GNU Coding
Standards}).  By using @code{configure} you prepare the source for your
specific environment by selecting and using @file{Makefile} fragments and
fragments of shell scripts, which are prepared in advance and stored with the
source.

@code{configure}'s command-line options also allow you to specify other aspects
of the source configuration:

@smallexample
  configure @var{hosttype}  [--target=@var{target}] [--srcdir=@var{dir}] [--rm]
            [--site=@var{site}] [--prefix=@var{dir}] [--exec-prefix=@var{dir}]
            [--program-prefix=@var{string}] [--tmpdir=@var{dir}]
            [--with-@var{package}[=@var{yes/no}]] [--without-@var{package}]
            [--enable-@var{feature}[=@var{yes/no}]] [--disable-@var{feature}]
            [--norecursion] [--nfp] [-s] [-v] [-V | --version] [--help]
@end smallexample

@table @code
@item --target=@var{target}
@cindex @code{--target}
@cindex @code{target} option
@vindex target
Requests that the sources be configured to target the @var{target} machine.  If
no target is specified explicitly, the target is assumed to be the same as the
host (i.e., a @dfn{native} configuration).  @xref{Host, , Host}, and
@ref{Target, , Target}, for
discussions of each.

@item --srcdir=@var{dir}
@cindex @code{--srcdir}
@cindex @code{srcdir} option
@vindex srcdir
Direct each generated @file{Makefile} to use the sources located in directory
@var{dir}.  Use this option whenever you wish the object code to reside in a
different place from the source code.  The @dfn{build directory} is always
assumed to be the directory you call @code{configure} from.  See @ref{Build
directories, , Build directories}, for an example.  If the source directory is
not specified, @code{configure} assumes that the source is in your current
directory.  If @code{configure} finds no @file{configure.in} there, it searches
in the same directory that the @code{configure} script itself lies in.
Pathnames specified (Values for @var{dir}) can be either absolute relative to
the @emph{build} directory.

@item --rm
@cindex @code{--rm}
@cindex @code{rm} option
@vindex rm
@emph{Remove} the configuration specified by @var{hosttype} and the other
command-line options, rather than create it.

@c FIXME: check @ref
@quotation
@emph{Note:} We recommend that you use @samp{make distclean} rather than
use this option; see @ref{Invoking make,,Invoking @code{make},make,GNU
Make}, for details on @samp{make distclean}.
@end quotation

@item --site=@var{site}
@cindex @code{--site}
@cindex @code{site} option
@vindex site
Generate the @file{Makefile} using site-specific @file{Makefile} fragments for
@var{site}.  @xref{Makefile fragments, , Adding information about local
conventions}.

@item --prefix=@var{dir}
@cindex @code{--prefix}
@cindex @code{prefix} option
@vindex prefix
Configure the source to install programs and files under directory @var{dir}.

This option sets the variable @samp{prefix}.  Each generated @file{Makefile}
will have its @samp{prefix} variables set to this value.  (@xref{What configure
really does, , What @code{configure} really does}.)

@item --exec-prefix=@var{dir}
@cindex @code{--exec-prefix}
@cindex @code{exec-prefix} option
@vindex exec-prefix
Configure the source to install @dfn{host dependent} files in @var{dir}.

This option sets the variable @samp{exec_prefix}.  Each generated
@file{Makefile} will have its @samp{exec_prefix} variables set to this value.
(@xref{What configure really does, , What @code{configure} really does}.)

@item --program-prefix=@var{string}
@cindex @code{--program-prefix}
@cindex @code{program-prefix} option
@vindex program-prefix
Configure the source to install certain programs using @var{string} as a
prefix.  This applies to programs which might be used for cross-compilation,
such as the compiler and the binary utilities, and also to programs which have
the same names as common Unix programs, such as @code{make}.

This option sets the variable @samp{program_prefix}.  Each generated
@file{Makefile} will have its @samp{program_prefix} variables set to this
value.  (@xref{What configure really does, , What @code{configure} really
does}.)

@item --tmpdir=@var{tmpdir}
@cindex @code{--tmpdir}
@cindex @code{tmpdir} option
@vindex tmpdir
Use the directory @var{tmpdir} for @code{configure}'s temporary files.  The
default is the value of the environment variable @w{@code{TMPDIR}}, or
@file{/tmp} if the environment variable is not set.

@item --with-@var{package}[=@var{yes/no}]
@itemx --without-@var{package}
@cindex @code{--with-@var{package}}
@cindex @code{with-@var{package}} option
@vindex with-@var{package}
@cindex @code{--without-@var{package}}
@cindex @code{without-@var{package}} option
@vindex without-@var{package}
Indicate that @var{package} is present, or not present, depending on
@var{yes/no}.  If @var{yes/no} is nonexistent, its value is assumed to be
@code{yes}.  @samp{--without-@var{package}} is equivalent to
@samp{--with-@var{package}=no}.

For example, if you wish to configure the program @code{gcc} for a Sun
SPARCstation running SunOS 4.x, and you want @code{gcc} to use the
@sc{gnu} linker @code{ld}, you can configure @code{gcc} using

@cindex Example session
@smallexample
eg$ configure --with-gnu-ld sun4
@end smallexample

@noindent
@xref{What configure really does, , What @code{configure} really does}, for
details.  See the installation or release notes for your particular package for
details on which other @var{package} options are recognized.
@c FIXME - need to include info about --with-* in other dox!

@item --enable-@var{feature}[=@var{yes/no}]
@itemx --disable-@var{feature}
@cindex @code{--enable-@var{feature}}
@cindex @code{enable-@var{feature}} option
@vindex enable-@var{feature}
@cindex @code{--disable-@var{feature}}
@cindex @code{disable-@var{feature}} option
@vindex disable-@var{feature}
Include @var{feature}, or not, depending on @var{yes/no}.  If @var{yes/no} is
nonexistent, its value is assumed to be @code{yes}.
@samp{--disable-@var{feature}} is equivalent to
@samp{--enable-@var{feature}=no}.

@noindent
@xref{What configure really does, , What @code{configure} really does}, for
details.  See the installation or release notes for your particular package for
details on which other @var{feature} options are recognized.
@c FIXME - need to include info about --enable-* in other dox!

@item --norecursion
@cindex @code{--norecursion}
@cindex @code{norecursion} option
@vindex norecursion
Configure only this directory; ignore any subdirectories.  This is used by the
executable shell script @file{config.status} to reconfigure only the current
directory; it is most often used non-interactively, when @code{make} is
invoked.  (@xref{config.status, , @code{config.status}}.)

@item --nfp
@cindex @code{--nfp}
@cindex @code{nfp} option
@vindex nfp
Assume that the intended @var{hosttype} has no floating point unit.

@item -s
@cindex @code{-s}
@cindex @code{s} option
Suppress status output.  This option is used internally by
@code{configure} when calling itself recursively in subdirectories.  You
can override this option with the @code{--verbose} option.

@item -v
@itemx --verbose
@cindex @code{-v}
@cindex @code{--verbose}
@cindex @code{v} option
@cindex @code{verbose} option
@cindex Verbose Output
@vindex verbose
Print status lines for each directory configured.  Normally, only the
status lines for the initial working directory are printed.

@item --version
@itemx -V
@cindex version
@cindex @code{--version}
@cindex version
Print the @code{configure} version number.

@item --help
@cindex Usage
@cindex @code{--help}
@cindex @code{help} option
Print a short summary of how to invoke @code{configure}.
@end table

@cindex Abbreviating option names
@cindex Truncating option names
@cartouche
@emph{Note:} You may introduce options with a single dash, @samp{-}, rather
than two dashes, @samp{--}.  However, you may not be able to truncate long
option names when using a single dash.  When using two dashes, options may be
abbreviated as long as each option can be uniquely identified.  For example,
@smallexample
eg$ configure --s=/u/me/src @var{hosttype}
@end smallexample
@noindent
is ambiguous, as @w{@samp{--s}} could refer to either @w{@samp{--site}} or
@w{@samp{--srcdir}}.  However,
@smallexample
eg$ configure --src=/u/me/src @var{hosttype}
@end smallexample
@noindent
is a valid abbreviation.
@end cartouche


@c ========================================================================
@node Using configure
@chapter Using @code{configure}
@cindex Using @code{configure}
@cindex Detailed usage
@cindex Usage: detailed

@code{configure} prepares source directories for building programs in
them.  ``Configuring'' is the process of preparing software to compile
correctly on a given @dfn{host}, for a given @dfn{target}.

@code{configure} subsequently writes a configured @file{Makefile} from a
pre-built template; @code{configure} uses variables that have been set in the
configuring process to determine the values of some variables in the
@file{Makefile}.  Because of this we will refer to both @code{configure}
variables and @file{Makefile} variables.  This convention allows us to
determine where the variable should be set initially, in either
@file{configure.in} or @file{Makefile.in}.

@menu
* What configure really does:: What configure really does
* configure.in::               The configure.in input file
* Install locations::          Where to install things once they are built
* Host::                       Telling configure what will source will be built
* Target::                     Telling configure what the source will target
* Makefile fragments::         Adding information about local conventions
* Makefile extensions::        Extensions to the GNU coding standards
@end menu

@c ---------------------------------------------------------------------
@node What configure really does
@section What @code{configure} really does
@cindex What @code{configure} really does
@cindex Behind the scenes
@cindex @code{configure} back end
@cindex @code{configure} details

Cygnus @code{configure} is a shell script that sets up an environment in
which your programs will compile correctly for your machine and
operating system, and will install in proper places.  @code{configure}
accomplishes this task by doing the following:

@itemize @bullet
@item 
it generates a @file{Makefile} from a custom template called
@file{Makefile.in} in each relevant source directory;

@item 
it customizes the build process to your specifications; you set certain
variables for @code{configure}, either on the command line or in the
file @file{configure.in}, which subsequently sets variables in each
generated @file{Makefile} to be used by @code{make} when actually
building the software;

@item 
it creates @dfn{build directories}, places for your code to be compiled
in before being installed;

@item 
it generates a @file{.gdbinit} in the build directory, if needed, to
communicate to @code{gdb} where to find the program's source code;

@item 
it generates a shell script called @file{config.status}
which is used most often by the @file{Makefile} to reconfigure itself;

@item 
it recurses in subdirectories, setting up entire trees so that they build
correctly; if @code{configure} finds another @code{configure} script
further down in a given source tree, it knows to use this script and not
recur.
@end itemize

For the sake of safety (i.e., in order to prevent broken installations), the
@sc{gnu} coding standards call for software to be @dfn{configured} in such a
way that an end user trying to build a given package will be able to do so by
affecting a finite number of variables.  All @sc{gnu} software comes with an
executable @code{configure} shell script which sets up an environment within a
build directory which will correctly compile your new package for your host
(or, alternatively, whatever host you specify to @code{configure}).
@ifinfo
For further background on this topic, see @ref{Some Basic Terms, , Apologia
Configure, cfg-paper, On Configuring Development Tools}, by K. Richard
Pixley.
@end ifinfo
@iftex
For further background on this topic, see @cite{On Configuring Development
Tools} by K. Richard Pixley.
@end iftex

Use @code{configure} to set for the build process:

@itemize @bullet
@item 
correct values for certain variables;

@item 
which type of host you wish to configure a given package for
(@pxref{Host, , Host});

@item 
where you want to install this package (by using @samp{prefix},
@samp{exec-prefix} and @samp{program-prefix}; @pxref{Install details, ,
Full descriptions of all installation directories});

@item 
optionally, which type of machine you wish to @dfn{target} this
package's output to (@pxref{Target, , Target});

@item 
which other @sc{gnu} packages are already installed and available to
this particular build (by using the @samp{--with-@var{package}} option;
@pxref{Invoking configure, , Invoking @code{configure}});

@item 
where to place temporary files (by using the @samp{--tmpdir=@var{dir}}
option; @pxref{Invoking configure, , Invoking @code{configure}});

@item whether to recur in subdirectories (changeable through the
@w{@samp{--norecursion}} option; @pxref{Invoking configure, , Invoking
@code{configure}}).
@end itemize

@code{configure} uses a few other files to complete its tasks.  These are
discussed in detail where noted.

@table @code
@cindex Other files
@item configure.in
@cindex @code{configure.in} definition
Input file for @code{configure}.  Shell script fragments reside here.
@xref{configure.in, , The @code{configure.in} input file}.

@item Makefile.in
@cindex @code{Makefile.in} definition
Template which @code{configure} uses to build a file called @file{Makefile} in
the @dfn{build directory}.  @xref{Makefile generation, , @code{Makefile}
generation}.

@item config.sub
@cindex @code{config.sub} definition
Shell script used by @code{configure} to expand referents to the
@var{hosttype} argument into a single specification of the form
@w{@var{cpu-vendor-os}}.  For instance, on the command line you can
specify

@cindex Example session
@example
eg$ ./configure sun4
@end example

@noindent
to configure for a Sun SPARCstation running SunOS 4.x.  @code{configure}
consults @code{config.sub} to find that the three-part specification for this
is

@example
sparc-sun-sunos4.1.1
@end example

@noindent
which notes the @var{cpu} as @samp{sparc}, the @var{manufacturer} as @samp{sun}
(Sun Microsystems), and the @var{os} (operating system) as @samp{sunos4.1.1},
the SunOS 4.1.1 release.  @xref{configure variables, , Variables available to @code{configure}}.

@item config.guess
@cindex @code{config.guess} definition
If you do not put the @var{hosttype} argument on the command line,
@code{configure} uses the @code{config.guess} shell script to make an
analysis of your machine (it assumes that you wish to configure your
software for the type of machine on which you are running).  The output
of @code{config.guess} is a three-part identifier as described above.

@item config.status
@cindex @code{config.status} definition
The final step in configuring a directory is to create a shell script,
@code{config.status}.  The main purpose of this file is to allow the
@file{Makefile} for the current directory to rebuild itself, if
necessary.  @xref{config.status, , @code{config.status}}.

@item config/*
@cindex @code{config/} subdirectory
@code{configure} uses three types of @file{Makefile} @dfn{fragments}, which
reside in the directory @file{@var{srcdir}/config/}.  @xref{Makefile fragments,
, Adding information about local conventions}.
@end table

@menu
* Build variables::         Variable-spaghetti made simple 
* Build directories::       Build directories described well
* Makefile generation::     To build a Makefile
* config.guess::            Be vewwy quiet, I'm hunting system information
* config.status::           To rebuild a Makefile
@end menu

@c ---------------------------------------------------------------------
@node Build variables
@subsection Build variables
@cindex Build variables
@cindex Cygnus Support Developer's Kit
@cindex Variables

There are several variables in the build process which you can control through
build programs such as @code{make}.  These include machine definitions, local
conventions, installation locations, locations for temporary files, etc.  This
data is accessible through certain variables which are configurable in the
build process; we refer to them as @dfn{build variables}.

For lists of build variables which you can affect by using @code{configure},
see @ref{configure variables, , Variables available to @code{configure.in}},
and @ref{Install details, , Full descriptions of all installation directories}.

Generally, build variables, which are used by the @file{Makefile} to
determine various aspects of the build and installation processes, are
changeable with command-line options to @code{configure}.  In most large
suites of programs, like the Cygnus Support Developer's Kit, the
individual programs reside in several subdirectories of a single source
code ``tree''.  All of these subdirectories need to be configured with
information relative to the @dfn{build directory}, which is not known
until @code{configure} is run.  Unless specified otherwise,
@code{configure} recursively configures every subdirectory in the source
tree.

Build variables are passed from @code{configure} directly into the
@file{Makefile}, and use the same names (except that dashes are
transformed into underbars; for example, when you specify the option
@samp{--exec-prefix} on the command line, the @file{Makefile} variable
@samp{exec_prefix} is set).  In other words, if you specify

@cindex Example session
@example
eg$ ./configure --prefix=/usr/gnu/local @dots{} @var{hosttype}
@end example

@noindent
on the command line, @code{configure} sets an variable called @samp{prefix} to
@samp{/usr/gnu/local}, and passes this into the @file{Makefile} in the same
manner.  After this command, each @file{Makefile} generated by @code{configure}
will contain a line that reads:

@example
prefix = /usr/gnu/local
@end example

For a list of the @file{Makefile} variables @code{configure} can change, and
instructions on how to change them, see @ref{configure variables, , Variables
available to @code{configure.in}}, and @ref{Invoking configure, , Invoking
@code{configure}}.

@c ---------------------------------------------------------------------
@node Build directories
@subsection Build directories
@cindex Build directories
@cindex Object directories
@cindex Building for multiple hosts
@cindex Building for multiple targets

By default, @code{configure} builds a @file{Makefile} and symbolic links in the
same directory as the source files.  This default works for many cases, but it
has limitations.  For instance, using this approach, you can only build object
code for one host at a time.

We refer to each directory where @code{configure} builds a @file{Makefile} as
a @dfn{build directory}.

The build directory for any given build is always the directory from which you
call @code{configure}, or @file{.} relative to your prompt.  The default
@dfn{source directory}, the place @code{configure} looks to find source code,
is also @file{.}.  For instance, if we have a directory @file{/gnu-stuff/src/}
that is the top branch of a tree of @sc{gnu} source code we wish to configure,
then the program we will use to configure this code is
@file{/gnu-stuff/src/configure}, as follows.  (Assume for the sake of argument
that our machine is a sun4.)

@cindex Example session
@smallexample
@group
eg$ cd /gnu-stuff/src
eg$ ./configure sun4
Created "Makefile" in /gnu-stuff/src
eg$
@end group
@end smallexample

We just configured the code in @file{/gnu-stuff/src} to run on a Sun
SPARCstation using SunOS 4.x by creating a @file{Makefile} in
@file{/gnu-stuff/src}.  By default, we also specified that when this code is
built, the object code should reside in the same directory,
@file{/gnu-stuff/src}.

However, if we wanted to build this code for more than one host, we would be in
trouble, because the new configuration would write over the old one, destroying
it in the process.  What we can do is to make a new @dfn{build directory} and
configure from there.  Running @code{configure} from the new directory will
place a correct @file{Makefile} and a @file{config.status} in this new file.
That is all @code{configure} does; we must run @code{make} to generate any
object code.  

The new @file{Makefile} in @file{/gnu-stuff/sun4-obj}, created from the
template file @file{/gnu-stuff/src/Makefile.in}, contains all the information
needed to build the program.

@cindex Example session
@smallexample
@group
eg$ mkdir /gnu-stuff/sun4-obj
eg$ cd /gnu-stuff/sun4-obj
eg$ ../src/configure --srcdir=../src sun4
Created "Makefile" in /gnu-stuff/sun4-obj
eg$ ls
Makefile       config.status
eg$ make all info install install-info clean
@var{compilation messages@dots{}}
eg$ mkdir /gnu-stuff/solaris2
eg$ cd /gnu-stuff/solaris2
eg$ ../src/configure --srcdir=../src sol2
Created "Makefile" in /gnu-stuff/solaris2
eg$ ls
Makefile       config.status
eg$ make all info install install-info clean
@var{compilation messages@dots{}}
@end group
@end smallexample

We can repeat this for other configurations of the same software simply
by making a new build directory and reconfiguring from inside it.  If
you do not specify the @var{hosttype} argument, @code{configure}
will attempt to figure out what kind of machine and operating system you
happen to be using.  @xref{config.guess, , Determining system
information}.  Of course, this may not always be the configuration you
wish to build.

@emph{Caution:} If you build more than one configuration for a single program,
remember that you must also specify a different @samp{--prefix} for each
configuration at configure-time.  Otherwise, both configurations will be
installed in the same default location (@file{/usr/local}); the configuration
to be installed last would overwrite previously installed configurations.

@c ---------------------------------------------------------------------
@node Makefile generation
@subsection @code{Makefile} generation
@cindex @code{Makefile} generation

Cygnus @code{configure} creates a file called @file{Makefile} in the build
directory which can be used with @code{make} to automatically build a given
program or package.  @code{configure} also builds a @file{Makefile} for each
relevant subdirectory for a given program or package (irrelevant subdirectories
would be those which contain no code which needs configuring, and which
therefore have no @code{configure} input file @file{configure.in} and no
@file{Makefile} template @file{Makefile.in}).  @xref{Running, @code{make}
Invocation, How to Run @code{make}, make, GNU Make}, for details on using
@code{make} to compile your source code.

Each @file{Makefile} contains variables which have been configured for a
specific build.  These build variables are determined when @code{configure} is
run.  All build variables have defaults.  By default, @code{configure}
generates a @file{Makefile} which specifies: 

@cindex Default configuration
@itemize @bullet
@item a @dfn{native} build, which is to occur 

@item in the current directory, and which will be installed 

@item in the default installation directory (@file{/usr/local}) when the code
is compiled with @code{make}.  
@end itemize

@noindent
Variables are changeable through command-line options to @code{configure}
(@pxref{Invoking configure, , Invoking @code{configure}}).

If you are porting a new program and intend to use @code{configure}, see
@ref{Porting, , Porting with @code{configure}}, as well as @ref{Makefiles, ,
Writing Makefiles, make, GNU Make}, and @ref{Makefiles, , Makefile Conventions,
standards, GNU Coding Standards}.

@c ---------------------------------------------------------------------
@node config.guess
@subsection Determining system information
@cindex @code{config.guess}

The shell script @code{config.guess} is called when you do not specify a
@var{hosttype} on the command line to @code{configure}.  @code{config.guess}
acquires available system information from your local machine through the shell
command @code{uname}.  It compares this information to a database and attempts
to determine a usable three-part system identifier (known as a @dfn{triple}) to
use as your @var{hosttype}.  @xref{What configure really does, , What
@code{configure} really does}, to see how this information is used.

@emph{Note:}  If you do not specify a @var{hosttype} on the command line,
@code{configure} will attempt to configure your software to run on the machine
you happen to be using.  This may not be the configuration you desire.

@c ---------------------------------------------------------------------
@node config.status
@subsection @code{config.status}
@cindex @code{config.status}

The final step in configuring a directory is to create an executable shell
script, @file{config.status}.  The main purpose of this file is to allow the
@file{Makefile} for the current directory to rebuild itself, if necessary.  It
is usually run from within the @file{Makefile}.  @xref{Makefile extensions, ,
Extensions to the @sc{gnu} coding standards}.

@file{config.status} also contains a record of the @code{configure} session
which created it.  

@c ---------------------------------------------------------------------
@node configure.in
@section The @code{configure.in} input file
@cindex @code{configure.in}

A @file{configure.in} file for Cygnus @code{configure} consists of a
@dfn{per-invocation} section, followed by a @dfn{per-host} section, followed by
a @dfn{per-target} section, optionally followed by a @dfn{post-target} section.
Each section is a shell script fragment, which is executed by the
@code{configure} shell script at an appropriate time.  Values are passed among
@code{configure} and the shell fragments through a set of shell variables.
When each section is being interpreted by the shell, the shell's current
directory is the build directory, and any files created by the section (or
referred to by the section) will be relative to the build directory.  To
reference files in other places (such as the source directory), prepend a shell
variable such as @samp{$(srcdir)/} to the desired file name.

@cindex @i{per-invocation} section
The beginning of the @file{configure.in} file begins the @dfn{per-invocation}
section.

@cindex @i{per-host} section
A line beginning with @samp{# per-host:} begins the @dfn{per-host} section.

@cindex @i{per-target} section
A line beginning with @samp{# per-target:} begins the @dfn{per-target} section.

@cindex @i{post-target} section
If it exists, the @dfn{post-target} section begins with @samp{# post-target:}.

@menu
* configure variables::    Variables available to configure.in
* Minimal::                A minimal configure.in
* Declarations::           For each invocation
* per-host::               Host-specific instructions
* per-target::             Target-specific instructions
* post-target::            Instructions to be executed after target info
* Example::                An example configure.in
@end menu

@c ---------------------------------------------------------------------
@node configure variables
@subsection Variables available to @code{configure.in}
@cindex @file{configure.in} interface
@cindex configure variables

The following variables pass information between the standard parts of
@code{configure} and the shell-script fragments in @file{configure.in}:

@table @code
@item srctrigger
@cindex @code{srctrigger}
@vindex srctrigger
Contains the name of a source file that is expected to live in the source
directory.  You must usually set this in the @dfn{per-invocation} section of
@file{configure.in}.  @code{configure} tests to see that this file exists.  If
the file does not exist, @code{configure} prints an error message.  This is
used as a sanity check that @file{configure.in} matches the source directory.

@item srcname
@cindex @code{srcname}
@vindex srcname
Contains the name of the source collection contained in the source directory.
You must usually set this in the @dfn{per-invocation} section of
@file{configure.in}.  If the file named in @samp{srctrigger} does not exist,
@code{configure} uses the value of @samp{srcname} when it prints the error
message.

@item configdirs
@cindex @code{configdirs}
@vindex configdirs
Contains the names of any subdirectories in which @code{configure} should
recurse.  You must usually set this in the @dfn{per-invocation} section of
@file{configure.in}.
If @file{Makefile.in} contains a line starting with @samp{SUBDIRS =},
then it will be replaced with an assignment to @samp{SUBDIRS} using
the value of @samp{configdirs} (if @samp{subdirs} is empty).  This can
be used to determine which directories to configure and build depending
on the host and target configurations.
@c Most other matching makefile/config vars use the same name.  Why not
@c this? (FIXME).
@c Can we get rid of SUBDIRS-substitution?  It doesn't work well with subdirs.
Use @samp{configdirs} (instead of the @samp{subdirs} variable
described below) if you want to be able to partition the
subdirectories, or use independent @file{Makefile} fragments.
Each subdirectory can be independent, and independently reconfigured.

@item subdirs
@cindex @code{subdirs}
@vindex subdirs
Contains the names of any subdirectories where @code{configure} should create a
@file{Makefile} (in addition to the current directory), @emph{without}
recursively running @code{configure}.  Use @samp{subdirs} (instead of the
@samp{configdirs} variable described above) if you want to configure all of the
directories as a unit.  Since there is a single invocation of @code{configure}
that configures many directories, all the directories can use the same
@file{Makefile} fragments, and the same @code{configure.in}.

@item host
@cindex @code{host}
@cindex Canonical ``triple''
@vindex host
Contains the full configuration name for the host (generated by the script
@file{config.sub} from the name that you entered).  This is a three-part
name (commonly referred to as a @dfn{triple}) of the form
@var{cpu}-@var{vendor}-@var{os}.

There are separate variables @samp{host_cpu}, @samp{host_vendor}, and
@samp{host_os} that you can use to test each of the three parts; this variable
is useful, however, for error messages, and for testing combinations of the
three components.

@item host_cpu
@vindex host_cpu
Contains the first element of the canonical triple representing the host
as returned by @file{config.sub}.  This is occasionally used to
distinguish between minor variations of a particular vendor's operating
system and sometimes to determine variations in binary format between
the host and the target.

@item host_vendor
@vindex host_vendor
Contains the second element of the canonical triple representing the host as
returned by @file{config.sub}.  This is usually used to distinguish among the
numerous variations of @emph{common} operating systems.
@c "@emph{common} OS" doesn't convey much to me.  Is this meant to cover 
@c cases like Unix, widespread but with many variations?

@item host_os
@vindex host_os
Contains the the third element of the canonical triple representing the
host as returned by @file{config.sub}.

@item target
@cindex @code{target}
@cindex Canonical ``triple''
@vindex target
Contains the full configuration name (generated by the script @file{config.sub}
from the name that you entered) for the target.  Like the host, this is a
three-part name of the form @var{cpu}-@var{vendor}-@var{os}.

There are separate variables @samp{target_cpu}, @samp{target_vendor}, and
@samp{target_os} that you can use to test each of the three parts; this
variable is useful, however, for error messages, and for testing combinations
of the three components.

@item target_cpu
@vindex target_cpu
Contains the first element of the canonical triple representing the target as
returned by @file{config.sub}.  This variable is used heavily by programs which
are involved in building other programs, like the compiler, assembler, linker,
etc.  Most programs will not need the @samp{target} variables at all, but this
one could conceivably be used to build a program, for instance, that operated
on binary data files whose byte order or alignment differ from the system where
the program is running.

@item target_vendor
@vindex target_vendor
Contains the second element of the canonical triple representing the target as
returned by @file{config.sub}.  This is usually used to distinguish among the
numerous variations of @emph{common} operating systems or object file
formats.  It is sometimes used to switch between different flavors of user
interfaces.
@c above query re "@emph{common} OS" applies here too

@item target_os
@vindex target_os
Contains the the third element of the canonical triple representing the
target as returned by @file{config.sub}.  This variable is used by
development tools to distinguish between subtle variations in object
file formats that some vendors use across operating system releases.  It
might also be use to decide which libraries to build or what user
interface the tool should provide.

@item floating_point
@cindex @code{floating_point}
@cindex @code{nfp} option
@vindex floating_point
Set to @samp{no} if you invoked @code{configure} with the @samp{--nfp}
command-line option, otherwise it is empty.  This is a request to target
machines with @dfn{no floating point} unit, even if the targets ordinarily have
floating point units available.

@item gas
@cindex @code{with-gnu-as} option
@vindex gas
Set to @samp{true} if you invoked @code{configure} with the
@w{@samp{--with-gnu-as}} command line option, otherwise it is empty.  This is a
request to assume that the specified @var{hosttype} machine has @sc{gnu} @code{as}
available even if it ordinarily does not.

@item srcdir
@cindex @code{srcdir}
@vindex srcdir
Set to the name of the directory containing the source for this program.
This will be different from @file{.} if you have specified the
@samp{--srcdir=@var{dir}} option.  @samp{srcdir} can indicate either an
absolute path or a path relative to the build directory.

@item package_makefile_frag
@vindex package_makefile_frag
If set in @file{configure.in}, this variable should be the name a file relative
to @samp{srcdir} to be included in the resulting @file{Makefile}.  If the named
file does not exist, @code{configure} will print a warning message.  This
variable is not set by @code{configure}.

@item host_makefile_frag
@vindex host_makefile_frag
If set in @file{configure.in}, this variable should be the name a file relative
to @samp{srcdir} to be included in the resulting @file{Makefile}.  If the named
file does not exist, @code{configure} will print a warning message.  This
variable is not set by @code{configure}.

@item target_makefile_frag
@vindex target_makefile_frag
If set in @file{configure.in}, this variable should be the name of a file,
relative to @samp{srcdir}, to be included in the resulting @file{Makefile}.  If
the named file does not exist, @code{configure} will print a warning message.
This variable is not set by @code{configure}.

@item site_makefile_frag
@vindex site_makefile_frag
Set to a file name representing to the default @file{Makefile} fragment for
this host.  It may be set in @file{configure.in} to override this default.
Normally @samp{site_makefile_frag} is empty, but will have a value if you
specify @samp{--site=@var{site}} on the command line.
@ignore -- this doesn't fit
It is probably not a good idea to override this variable from
@file{configure.in}, since that may defeat the @code{configure} user's
intentions.
@end ignore

@item Makefile
@vindex Makefile
Set to the name of the generated @file{Makefile}.  Normally this value is
precisely @file{Makefile}, but some programs may want something else.

@item removing
@cindex @code{rm} option
@vindex removing
Normally empty but will be set to some non-null value if you specified
@samp{--rm} on the command line.  That is, if @samp{removing} is not empty,
then @code{configure} is @emph{removing} a configuration rather than creating
one.

@item files
@cindex Symbolic links
@vindex files
If this variable is not empty following the @dfn{per-target} section,
then each word in its value will be the target of a symbolic link named
in the corresponding word from the @samp{links} variable.

@item links
@cindex Symbolic links
@vindex links
If the @samp{files} variable is not empty following the @dfn{per-target}
section, then @code{configure} creates symbolic links with the first word of
@samp{links} pointing to the first word of @samp{files}, the second word of
@samp{links} pointing to the second word of @samp{files}, and so on.
@end table

@c ---------------------------------------------------------------------
@node Minimal
@subsection A minimal @code{configure.in}
@cindex Minimal @file{configure.in} example

A minimal @file{configure.in} consists of four lines.

@example
srctrigger=foo.c
srcname="source for the foo program"
# per-host:
# per-target:
@end example

The @samp{# per-host:} and @samp{# per-target:} lines divide the file into the
three required sections.  The @samp{srctrigger} line names a file.
@code{configure} checks to see that this file exists in the source directory
before configuring.  If the @samp{srctrigger} file does not exist,
@code{configure} uses the value of @samp{srcname} to print an error message
about not finding the source.

This particular example uses no links, and only the default host,
target, and site-specific @file{Makefile} fragments if they exist.

@c ---------------------------------------------------------------------
@node Declarations
@subsection For each invocation
@cindex For each invocation
@cindex Declarations section
@cindex @i{per-invocation} section

@code{configure} invokes the entire shell script fragment from the start of
@file{configure.in} up to a line beginning with @w{@samp{# per-host:}}
immediately after parsing command line arguments.  The variables
@samp{srctrigger} and @samp{srcname} @emph{must} be set here.

You might also want to set the variables @samp{configdirs} and
@samp{package_makefile_frag} here.

@c ---------------------------------------------------------------------
@node per-host
@subsection Host-specific instructions
@cindex Host-specific instructions
@cindex @i{host} shell-script fragment
@cindex @i{per-host} section

The @dfn{per-host} section of @file{configure.in} starts with the line that
begins with @w{@samp{# per-host:}} and ends before a line beginning with
@w{@samp{# per-target:}}.  @code{configure} invokes the commands in the
@dfn{per-host} section when determining host-specific information.

This section usually contains a big @code{case} statement using the variable
@samp{host} to determine appropriate values for @samp{host_makefile_frag} and
@samp{files}, although @samp{files} is not usually set here.  Usually, it is
set at the end of the @dfn{per-target} section after determining the names of
the target specific configuration files.

@c ---------------------------------------------------------------------
@node per-target
@subsection Target-specific instructions
@cindex Target-specific instructions
@cindex target shell-script fragment
@cindex @i{per-target} section

The @dfn{per-target} section of @file{configure.in} starts with the line that
begins with @w{@samp{# per-target:}} and ends before the line that begins with
@w{@samp{# post-target:}}, if there is such a line.  Otherwise the
@dfn{per-target} section extends to the end of the file.  @code{configure}
invokes the commands in the @dfn{per-target} section when determining
target-specific information, and before building any files, directories, or
links.

This section usually contains a big @code{case} statement using the variable
@samp{target} to determine appropriate values for @samp{target_makefile_frag}
and @samp{files}.  The last lines in the @dfn{per-target} section normally set
the variables @samp{files} and @samp{links}.

@c ---------------------------------------------------------------------
@node post-target
@subsection Instructions to be executed after target info
@cindex Post-target shell-script fragment
@cindex @i{post-target} section

The @dfn{post-target} section is optional.  If it exists, the
@samp{post-target} section starts with a line beginning with @w{@samp{#
Post-target:}} and extends to the end of the file.  If it exists,
@code{configure} invokes this section once for each target after
building all files, directories, or links.

This section is seldom needed, but you can use it to edit the @file{Makefile}
generated by @code{configure}.

@c ---------------------------------------------------------------------
@node Example
@subsection An example @code{configure.in}
@cindex Example @file{configure.in}
@cindex Sample @file{configure.in}
@c @cindex @code{bison} @file{configure.in}
@c this won't be the bison configure.in for long.. need better example

Here is a small example of a @file{configure.in} file.

@cartouche
@example
@group
# This file is a collection of shell script fragments
# used to tailor a template configure script as
# appropriate for this directory.  For more information,
# see configure.texi.

configdirs=
srctrigger=warshall.c
srcname="bison"

# per-host:
case "$@{host@}" in
m88k-motorola-*)
        host_makefile_frag=config/mh-delta88
        ;;
esac

# per-target:
files="bison_in.hairy"
links="bison.hairy"

# post-target:
@end group
@end example
@end cartouche

@c ---------------------------------------------------------------------
@node Install locations
@section Install locations
@cindex Where to install
@cindex Install locations

Using the default configuration, @samp{make install} creates a single tree of
files, some of which are programs.  The location of this tree is determined by
the value of the variable @samp{prefix}.  The default value of @samp{prefix} is
@samp{/usr/local}.  This is often correct for native tools installed on only
one host.

@menu
* prefix::            Changing the default install directory
* exec_prefix::       How to separate host independent files
                                         from host dependent files when
                                         installing for multiple hosts
* Install details::   Full descriptions of all installation subdirectories
@end menu

@c ---------------------------------------------------------------------
@node prefix
@subsection Changing the default install directory
@cindex Changing the install directory
@cindex @code{prefix} option
@vindex prefix

In the default configuration, all files are installed in subdirectories
of @file{/usr/local}.  The location is determined by the value of
the @code{configure} variable @samp{prefix}; in turn, this determines the
value of the @file{Makefile} variable of the same name (@samp{prefix}).

You can also set the value of the @file{Makefile} variable @samp{prefix}
explicitly each time you invoke @code{make} if you are so inclined.  However,
because many programs have this location compiled in, you must specify the
@samp{prefix} value consistently on each invocation of @code{make}, or you will
end up with a broken installation.

To make this easier, the value of the @code{configure} variable
@samp{prefix} can be set on the command line to @code{configure}
using the option @samp{--prefix=}.  

@c ---------------------------------------------------------------------
@node exec_prefix
@subsection Installing for multiple hosts
@cindex Configuring for multiple hosts
@cindex Sharing host-independent files
@cindex Installing host-independent files
@cindex The @code{exec_prefix} directory
@vindex exec_prefix

By default, host dependent files are installed in subdirectories of
@file{$(exec_prefix)}.  The location is determined by the value of the
@code{configure} variable @samp{exec_prefix}, which determines the value of the
@file{Makefile} variable @samp{exec_prefix}.  This makes it easier to install
for a single host, and simplifies changing the default location for the install
tree.  The default doesn't allow for multiple hosts to effectively share
host independent files, however.

To configure so that multiple hosts can share common files, use something like:

@cindex Example session
@smallexample
configure @var{host1} -prefix=/usr/gnu -exec_prefix=/usr/gnu/H-host1
make all info install install-info clean

configure @var{host2} -prefix=/usr/gnu -exec_prefix=/usr/gnu/H-host2
make all info install install-info
@end smallexample

The first line configures the source for @var{host1} to place host-specific
programs in subdirectories of @file{/usr/gnu/H-@var{host1}}.

The second line builds and installs all programs for @var{host1},
including both host-independent and host-specific files, as well as removing
the host-specific object files from of the build directory.

The third line reconfigures the source for @var{host2} to place host
specific programs in subdirectories of @file{/usr/gnu/H-@var{host2}}.

The fourth line builds and installs all programs for @var{host2}.  Host
specific files are installed in new directories, but the host
independent files are installed @emph{on top of} the host
independent files installed for @var{host1}.  This results in a single
copy of the host independent files, suitable for use by both hosts.

@xref{Makefile extensions, , Extensions to the @sc{gnu} coding standards}, for
more information.

@c ---------------------------------------------------------------------
@node Install details
@subsection Full descriptions of all installation subdirectories
@cindex Install details
@cindex Installation subdirectories
@cindex Subdirectories

During any install, a number of standard directories are created.  Their names
are determined by @file{Makefile} variables.  Some of the defaults for
@file{Makefile} variables can be changed at configuration time using command
line options to @code{configure}.  For more information on the standard
directories or the @file{Makefile} variables, please refer to @ref{Makefiles, ,
Makefile Conventions, standards, GNU Coding Standards}.  See also @ref{Makefile
extensions, , Extensions to the @sc{gnu} coding standards}.

Note that @code{configure} does not create the directory indicated by the
variable @samp{srcdir} at any time.  @code{$(srcdir)} is not an installation
directory.

You can override all @file{Makefile} variables on the command line to
@code{make}.  (@xref{Overriding, , Overriding Variables, make, GNU Make}.)  If
you do so, you will need to specify the value precisely the same way for each
invocation of @code{make}, or you risk ending up with a broken installation.
This is because many programs have the locations of other programs or files
compiled into them.  If you find yourself overriding any of the variables
frequently, you should consider site dependent @file{Makefile} fragments.  See
also @ref{Sites, , Adding site info}.

During @samp{make install}, a number of standard directories are created and
populated.  The following @file{Makefile} variables define them.  Those whose
defaults are set by corresponding @code{configure} variables are marked
``@code{Makefile} and @code{configure}''.

@table @code
@item prefix (@code{Makefile} and @code{configure})
@cindex @code{prefix}
@vindex prefix
The root of the installation tree.  You can set its @file{Makefile} default
with the @samp{--prefix=} command line option to @code{configure}
(@pxref{Invoking configure, , Invoking @code{configure}}).  The default value
for @samp{prefix} is @samp{/usr/local}.

@item bindir
@cindex @code{bindir}
@vindex bindir
A directory for binary programs that users can run.  The default value for
@samp{bindir} depends on @samp{prefix}; @samp{bindir} is normally changed only
indirectly through @samp{prefix}.  The default value for @samp{bindir} is
@samp{$(prefix)/bin}.

@item exec_prefix (@code{Makefile} and @code{configure})
@cindex @code{exec_prefix}
@vindex exec_prefix
A directory for host dependent files.  You can specify the @file{Makefile}
default value by using the @samp{--exec_prefix=} option to @code{configure}.
(@xref{Invoking configure, , Invoking @code{configure}}.)  The default value
for @samp{exec_prefix} is @samp{$(prefix)}.

@item libdir
@cindex @code{libdir}
@vindex libdir
A directory for libraries and support programs.  The default value for
@samp{libdir} depends on @samp{prefix}; @samp{libdir} is normally changed only
indirectly through @samp{prefix}.  The default value for @samp{libdir} is
@samp{$(prefix)/lib}.

@item mandir
@cindex @code{mandir}
@vindex mandir
A directory for @code{man} format documentation (``man pages'').  The default
value for @samp{mandir} depends on @samp{prefix}; @samp{mandir} is normally
changed only indirectly through @samp{prefix}.  The default value for
@samp{mandir} is @samp{$(prefix)/man}.

@item man@var{N}dir
@cindex @code{man@var{N}dir}
@vindex man@var{N}dir
These are eight variables named @samp{man1dir}, @samp{man2dir}, etc.  They name
the specific directories for each man page section.  For example,
@samp{man1dir} by default holds the filename @file{$(mandir)/man1}; this
directory contains @file{emacs.1} (the man page for @sc{gnu} Emacs).
Similarly, @samp{man5dir} contains the value @file{$(mandir)/man5}, indicating
the directory which holds @file{rcsfile.5} (the man page describing the
@code{rcs} data file format).  The default value for any of the
@samp{man@var{N}dir} variables depends indirectly on @samp{prefix}, and is
normally changed only through @samp{prefix}.  The default value for
@samp{man@var{N}dir} is @samp{$(mandir)/man@var{N}}.

@item man@var{N}ext
@cindex @code{man@var{N}ext}
@vindex man@var{N}ext
@emph{Not supported by Cygnus @code{configure}}.  The @cite{@sc{gnu} Coding
Standards} do not call for @samp{man1ext}, @samp{man2ext}, so the intended use
for @code{manext} is apparently not parallel to @samp{mandir}.  Its use is not
clear.  (See also @ref{Makefile extensions, , Extensions to the @sc{gnu} coding
standards}.)

@item infodir
@cindex @code{infodir}
@vindex infodir
A directory for @code{info} format documentation.  The default value for
@samp{infodir} depends indirectly on @samp{prefix}; @samp{infodir} is
normally changed only through @samp{prefix}.  The default value for
@samp{infodir} is @samp{$(prefix)/info}.

@item docdir
@cindex @code{docdir}
@vindex docdir
A directory for any documentation that is in a format other than those used by
@code{info} or @code{man}.  The default value for @samp{docdir} depends
indirectly on @samp{prefix}; @samp{docdir} is normally changed only through
@samp{prefix}.  The default value for @samp{docdir} is @samp{$(datadir)/doc}.
@emph{This variable is an extension to the @sc{gnu} coding standards}.  (See
also @ref{Makefile extensions, , Extensions to the @sc{gnu} coding standards}.)

@item includedir
@cindex @code{includedir}
@vindex includedir
A directory for the header files accompanying the libraries installed in
@samp{libdir}.  The default value for @samp{includedir} depends on
@samp{prefix}; @samp{includedir} is normally changed only indirectly
through @samp{prefix}.  The default value for @samp{includedir} is
@samp{$(prefix)/include}.
@end table

@c ---------------------------------------------------------------------
@node Host
@section Host
@cindex Host

The arguments to @code{configure} are @dfn{hosttypes}.  By
@dfn{hosttype} we mean the @dfn{environment} in which the source will be
compiled.  This need not necessarily be the same as the physical machine
involved, although it usually is.

For example, if some obscure machine had the @sc{gnu} @code{POSIX} emulation
libraries available, it would be possible to configure most @sc{gnu} source for
a @code{POSIX} system and build it on the obscure host.

For more on this topic, see @ref{Host Environments, On Configuring Development
Tools, Host Environments, cfg-paper, On Configuring Development Tools}.

@c ---------------------------------------------------------------------
@node Target
@section Target
@cindex Target

For building native development tools, or most of the other @sc{gnu}
tools, you need not worry about the target.  The @dfn{target} of a
configuration defaults to the same as the @dfn{host}.

For building cross development tools, please see @ref{Building Development
Environments, On Configuring Development Tools, Building Development
Environments, cfg-paper, On Configuring Development Tools}.

@c ---------------------------------------------------------------------
@node Makefile fragments
@section Adding information about local conventions
@cindex @code{Makefile} fragments
@cindex Local conventions
@cindex Adding local info
@cindex Adding site info

If you find that a tool does not get configured to your liking, or if
@code{configure}'s conventions differ from your local conventions, you should
probably consider @dfn{site-specific @file{Makefile} fragments}.  See also
@ref{Sites, , Adding site info}.

These are probably not the right choice for options that can be set from
the @code{configure} command line or for differences that are host or
target dependent.

Cygnus @code{configure} uses three types of @file{Makefile} fragments.  In a
generated @file{Makefile} they appear in the order: @dfn{target fragment},
@dfn{host fragment}, and @dfn{site fragment}.  This allows host fragments to
override target fragments, and site fragments to override both.

Host-specific @file{Makefile} fragments conventionally reside in the
@file{./config/} subdirectory with names of the form @file{mh-@var{hosttype}}.
They are used for hosts that require odd options to the standard compiler and
for compile time options based on the host configuration.

Target-specific @file{Makefile} fragments conventionally reside in the
@file{./config/} subdirectory with names of the form @file{mt-@var{target}}.
They are used for target dependent compile time options.

Site specific @file{Makefile} fragments conventionally reside in the
@file{./config/} subdirectory with names of the form @file{ms-@var{site}}.
They are used to override host- and target-independent compile time options.
Note that you can also override these options on the @code{make} invocation
line.

@c ---------------------------------------------------------------------
@node Makefile extensions
@section Extensions to the @sc{gnu} coding standards
@cindex @code{Makefile} extensions
@cindex Cygnus extensions
@cindex Coding standards extensions

The following additions to the @sc{gnu} coding standards are required for
Cygnus @code{configure} to work properly.

@itemize @bullet
@item
The @file{Makefile} must contain exactly one line starting with @samp{####}.
This line should follow any default macro definitions but precede any rules.
Host, target, and site-specific @file{Makefile} fragments will be inserted
immediately after this line.  If the line is missing, the fragments will not be
inserted.

@item
Cygnus adds the following targets to each @file{Makefile}.  Their existence is
not required for Cygnus @code{configure}, but they are documented here for
completeness.

@table @code
@kindex info
@item info
Build all info files from texinfo source.

@kindex install-info
@item install-info
Install all info files.

@kindex clean-info
@item clean-info
Remove all info files and any intermediate files that can be generated
from texinfo source.

@kindex Makefile
@item Makefile
Calls @code{./config.status} to rebuild the @file{Makefile} in this directory.
@end table

@item
The following @file{Makefile} targets have revised semantics:

@table @code
@kindex install
@item install
Should @emph{not} depend on the target @samp{all}.  If the program is not
already built, @samp{make install} should fail.  This allows you to install
programs even when @code{make} would otherwise determine them to be out of
date.  This can happen, for example, when the result of a @samp{make all} is
transported via tape to another machine for installation.

@kindex clean
@item clean
Should remove any file that can be regenerated by the @file{Makefile},
excepting only the @file{Makefile} itself, and any links created by
@code{configure}.  That is, @code{make all clean} should return all directories
to their original condition.  If this is not done, then the command sequence

@cindex Example session
@example
configure @var{host1} ; make all install clean ; 
configure @var{host2} ; make all install
@end example

@noindent
will fail because of intermediate files intended for @var{host1}.
@end table

@item
Cygnus adds the following macros to all @file{Makefile.in} files, but
you are not required to use them to run Cygnus @code{configure}.

@table @code
@kindex docdir
@item docdir
The directory in which to install any documentation that is not either a
@code{man} page or an @code{info} file.  For @code{man} pages, see
@samp{mandir}; for @code{info}, see @samp{infodir}.

@kindex includedir
@item includedir
The directory in which to install any header files that should be made
available to users.  This is distinct from the @code{gcc} include directory,
which is intended for @code{gcc} only.  Files in @samp{includedir} may be used
by @code{cc} as well.
@end table

@item
The following macros have revised semantics.  Most of them describe
installation directories; see also @ref{Install details, , Full description of
all installation subdirectories}.

@table @code
@kindex datadir
@item datadir
is used for host independent data files.

@kindex mandir
@item mandir
The default path for @samp{mandir} depends on @samp{prefix}.

@kindex infodir
@item infodir
The default path for @samp{infodir} depends on @samp{prefix}.

@kindex BISON
@item BISON
is assumed to have a @code{yacc} calling convention.  To use @sc{gnu}
@code{bison}, use @samp{BISON=bison -y}.
@end table

@item
Each Cygnus @file{Makefile} also conforms to one additional restriction:

When libraries are installed, the line containing the call to
@samp{INSTALL_DATA} should always be followed by a line containing a call to
@samp{RANLIB} on the installed library.  This is to accommodate systems that
use @code{ranlib}.  Systems that do not use @code{ranlib} can set @samp{RANLIB}
to ``@code{echo}'' in a host specific @file{Makefile} fragment.
@end itemize

@c ========================================================================
@node Porting
@chapter Porting with @code{configure}
@cindex Porting with @code{configure}

This section explains how to add programs, host and target configuration
names, and site-specific information to Cygnus @code{configure}.

@menu
* Programs::               Adding configure to new programs
* Hosts and targets::      Adding hosts and targets
* Sites::                  Adding site info
@end menu

@c ---------------------------------------------------------------------
@node Programs
@section Adding @code{configure} to new programs
@cindex Adding @code{configure} to new programs

If you are writing a new program, you probably shouldn't worry about porting or
configuration issues until it is running reasonably on some host.  Then refer
back to this section.

If your program currently has a @code{configure} script that meets the @sc{gnu}
standards (@pxref{Configuration, , How Configuration Should Work, standards,
GNU Coding Standards}, please do not add Cygnus @code{configure}.  It should be
possible to add this program without change to a Cygnus @code{configure} style
source tree.

@cindex @code{autoconf}
If the program is not target dependent, please consider using @code{autoconf}
instead of Cygnus @code{configure}.  @code{autoconf} is available from the Free
Software Foundation; it is a program which generates an executable shell script
called @file{configure} by automatically finding information on the system to
be configured on and embedding this information in the shell script.
@file{configure} scripts generated by @code{autoconf} require no arguments, and
accept the same options as Cygnus @code{configure}.  For detailed instructions
on using @code{autoconf}, see @ref{Making configure Scripts, , How to organize
and produce Autoconf scripts, autoconf, Autoconf}.


To add Cygnus @code{configure} to an existing program, do the following:

@table @bullet
@item Make sure the @file{Makefile} conforms to the @sc{gnu} standard
The coding standard for writing a @sc{gnu} @file{Makefile} is described in
@ref{Makefiles, , Makefile Conventions, standards, GNU Coding Standards}.  For
technical information on writing a @file{Makefile}, see @ref{Makefiles, ,
Writing Makefiles, make, GNU Make}.

@item Add Cygnus extensions to the @file{Makefile}
These are described in @ref{Makefile extensions, , Extensions to the @sc{gnu}
coding standards}.

@item Collect package specific definitions in a single file
Many packages are best configured using a common @file{Makefile} fragment which
is included by all of the makefiles in the different directories of the
package.  In order to accomplish this, set the variable
@samp{package_makefile_fragment} to the name of the file.  It will be inserted
into the final @file{Makefile} before the target-specific fragment.

@item Move host support from @file{Makefile} to fragments
This usually involves finding sections of the @file{Makefile} that say things
like ``uncomment these lines for host @var{hosttype}'' and moving them to a new
file called @file{./config/mh-@var{hosttype}}. For more information, see @ref{Hosts
and targets, , Adding hosts and targets}.

@item Choose defaults
If the program has compile-time options that determine the way the program
should behave, choose reasonable defaults and make these @file{Makefile}
variables.  Be sure the variables are assigned their default values before the
@samp{####} line so that site-specific @file{Makefile} fragments can override
them (@pxref{Makefile extensions, , Extensions to the @sc{gnu} coding
standards}).

@item Locate configuration files
If there is configuration information in header files or source files, separate
it in such a way that the files have generic names.  Then move the specific
instances of those files into the @file{./config/} subdirectory.

@item Separate host and target information
Some programs already have this information separated.  If yours does not, you
will need to separate these two kinds of configuration information.  @dfn{Host
specific} information is the information needed to compile the program.
@dfn{Target specific} information is information on the format of data files
that the program will read or write.  This information should live in separate
files in the @file{./config/} subdirectory with names that reflect the
configuration for which they are intended.

At this point you might skip this step and simply move on.  If you do, you
should end up with a program that can be configured only to build @dfn{native}
tools, that is, tools for which the host system is also the target system.
Later, you could attempt to build a cross tool and separate out the
target-specific information by figuring out what went wrong.  This is often
simpler than combing through all of the source code.

@item Write @code{configure.in}
Usually this involves writing shell script fragments to map from canonical
configuration names into the names of the configuration files.  These files
will then be linked at configure time from the specific instances of those
files in @file{./config} to files in the build directory with more generic
names.  (See also @ref{Build directories, , Build directories}.)  The format of
@file{configure.in} is described in @ref{configure.in, , The
@code{configure.in} input file}.

@item Rename @file{Makefile} to @file{Makefile.in}
@end table

At this point you should have a program that can be configured using
Cygnus @code{configure}.

@c ---------------------------------------------------------------------
@node Hosts and targets
@section Adding hosts and targets
@cindex Adding hosts and targets
@cindex Hosts and targets

To add a host or target to a program that already uses Cygnus @code{configure},
do the following.

@itemize @bullet

@item
Make sure the new configuration name is represented in @file{config.sub}.  If
not, add it.  For more details, see the comments in the shell script
@file{config.sub}.

@item
If you are adding a host configuration, look in @file{configure.in}, in the
@dfn{per-host} section.  Make sure that your configuration name is represented
in the mapping from host configuration names to configuration files.  If not,
add it.  Also see @ref{configure.in, , The @code{configure.in} input file}.

@item
If you are adding a target configuration, look in @file{configure.in}, in the
@dfn{per-target} section.  Make sure that your configuration name is
represented in the mapping from target configuration names to configuration
files.  If not, add it.  Also see @ref{configure.in, , The @code{configure.in}
input file}.

@item
Look in @file{configure.in} for the variables @samp{files}, @samp{links},
@samp{host_makefile_frag}, and @samp{target_makefile_frag}.  The values
assigned to these variables are the names of the configuration files, (relative
to @samp{srcdir}) that the program uses.  Make sure that copies of the files
exist for your host.  If not, create them.  See also @ref{configure variables,
, Variables available to @code{configure.in}}.
@end itemize

This should be enough to @code{configure} for a new host or target
configuration name.  Getting the program to compile and run properly represents
the hardest work of any port.

@c ---------------------------------------------------------------------
@node Sites
@section Adding site info
@cindex Sites
@cindex Adding site info

If some of the @file{Makefile} defaults are not right for your site, you can
build site-specific @file{Makefile} fragments.  To do this, do the following.

@itemize @bullet

@item
Choose a name for your site.  It must currently be less than eleven characters.

@item
If the program source does not have a @file{./config/} subdirectory, create it.

@item
Create a file called @file{./config/ms-@var{site}} where @var{site} is the name
of your site.  In it, set whatever @file{Makefile} variables you need to
override to match your site's conventions.

@item
Configure the program with:

@cindex Example session
@example
configure @dots{} --site=@var{site}
@end example

@end itemize

@c ---------------------------------------------------------------------
@node Variables Index
@unnumbered Variable Index

@printindex vr

@page
@c ---------------------------------------------------------------------
@node Concept Index
@unnumbered Concept Index

@printindex cp
@contents
@bye

@c Local Variables:
@c fill-column: 79
@c outline-regexp: "@chap"
@c End:
@c (setq outline-regexp "@chapt\\\|@unnum\\\|@setf\\\|@conte\\\|@sectio\\\|@subsect\\\|@itemize\\\|@defvar{")