/[svn]/qsampler/trunk/INSTALL

Diff of /qsampler/trunk/INSTALL

Parent Directory | Revision Log | View Patch Patch

-revision 2319 by capela,
Thu Oct 30 22:26:46 2003 UTC
+revision 2320 by capela,
Mon Feb 20 20:27:07 2012 UTC
 Line 1
+ Installation Instructions
+ *************************
+ Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004, 2005,
+, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
+    Copying and distribution of this file, with or without modification,
+ are permitted in any medium without royalty provided the copyright
+ notice and this notice are preserved.  This file is offered as-is,
+ without warranty of any kind.
  Basic Installation
  ==================
-    These are generic installation instructions.
+    Briefly, the shell commands `./configure; make; make install' should
+ configure, build, and install this package.  The following
+ more-detailed instructions are generic; see the `README' file for
+ instructions specific to this package.  Some packages provide this
+ `INSTALL' file but do not implement all of the features documented
+ below.  The lack of an optional feature in a given package is not
+ necessarily a bug.  More recommendations for GNU packages can be found
+ in *note Makefile Conventions: (standards)Makefile Conventions.
     The `configure' shell script attempts to guess correct values for
  various system-dependent variables used during compilation.  It uses
  those values to create a `Makefile' in each directory of the package.
  It may also create one or more `.h' files containing system-dependent
  definitions.  Finally, it creates a shell script `config.status' that
- you can run in the future to recreate the current configuration, a file
+ you can run in the future to recreate the current configuration, and a
- `config.cache' that saves the results of its tests to speed up
+ file `config.log' containing compiler output (useful mainly for
- reconfiguring, and a file `config.log' containing compiler output
+ debugging `configure').
- (useful mainly for debugging `configure').
+    It can also use an optional file (typically called `config.cache'
+ and enabled with `--cache-file=config.cache' or simply `-C') that saves
+ the results of its tests to speed up reconfiguring.  Caching is
+ disabled by default to prevent problems with accidental use of stale
+ cache files.
     If you need to do unusual things to compile the package, please try
  to figure out how `configure' could check whether to do them, and mail
  diffs or instructions to the address given in the `README' so they can
- be considered for the next release.  If at some point `config.cache'
+ be considered for the next release.  If you are using the cache, and at
- contains results you don't want to keep, you may remove or edit it.
+ some point `config.cache' contains results you don't want to keep, you
+ may remove or edit it.
-    The file `configure.in' is used to create `configure' by a program
- called `autoconf'.  You only need `configure.in' if you want to change
+    The file `configure.ac' (or `configure.in') is used to create
- it or regenerate `configure' using a newer version of `autoconf'.
+ `configure' by a program called `autoconf'.  You need `configure.ac' if
+ you want to change it or regenerate `configure' using a newer version
+ of `autoconf'.
- The simplest way to compile this package is:
+    The simplest way to compile this package is:
 . `cd' to the directory containing the package's source code and type
-      `./configure' to configure the package for your system.  If you're
+      `./configure' to configure the package for your system.
-      using `csh' on an old version of System V, you might need to type
-      `sh ./configure' instead to prevent `csh' from trying to execute
-      `configure' itself.
-      Running `configure' takes a while.  While running, it prints some
+      Running `configure' might take a while.  While running, it prints
-      messages telling which features it is checking for.
+      some messages telling which features it is checking for.
 . Type `make' to compile the package.
-. Type `make install' to install the programs and any data files and
+. Optionally, type `make check' to run any self-tests that come with
-      documentation.
+      the package, generally using the just-built uninstalled binaries.
-. You can remove the program binaries and object files from the
+. Type `make install' to install the programs and any data files and
-      source code directory by typing `make clean'.
+      documentation.  When installing into a prefix owned by root, it is
+      recommended that the package be configured and built as a regular
+      user, and only the `make install' phase executed with root
+      privileges.
+. Optionally, type `make installcheck' to repeat any self-tests, but
+      this time using the binaries in their final installed location.
+      This target does not install anything.  Running this target as a
+      regular user, particularly if the prior `make install' required
+      root privileges, verifies that the installation completed
+      correctly.
+. You can remove the program binaries and object files from the
+      source code directory by typing `make clean'.  To also remove the
+      files that `configure' created (so you can compile the package for
+      a different kind of computer), type `make distclean'.  There is
+      also a `make maintainer-clean' target, but that is intended mainly
+      for the package's developers.  If you use it, you may have to get
+      all sorts of other programs in order to regenerate files that came
+      with the distribution.
+. Often, you can also type `make uninstall' to remove the installed
+      files again.  In practice, not all packages have tested that
+      uninstallation works correctly, even though it is required by the
+      GNU Coding Standards.
+. Some packages, particularly those that use Automake, provide `make
+      distcheck', which can by used by developers to test that all other
+      targets like `make install' and `make uninstall' work correctly.
+      This target is generally not run by end users.
  Compilers and Options
  =====================
     Some systems require unusual options for compilation or linking that
- the `configure' script does not know about.  You can give `configure'
+ the `configure' script does not know about.  Run `./configure --help'
- initial values for variables by setting them in the environment.  Using
+ for details on some of the pertinent environment variables.
- a Bourne-compatible shell, you can do that on the command line like
- this:
+    You can give `configure' initial values for configuration parameters
-      CC=c89 CFLAGS=-O2 LIBS=-lposix ./configure
+ by setting variables in the command line or in the environment.  Here
+ is an example:
+      ./configure CC=c99 CFLAGS=-g LIBS=-lposix
- Or on systems that have the `env' program, you can do it like this:
+    *Note Defining Variables::, for more details.
-      env CPPFLAGS=-I/usr/local/include LDFLAGS=-s ./configure
  Compiling For Multiple Architectures
  ====================================
     You can compile the package for more than one kind of computer at the
  same time, by placing the object files for each architecture in their
- own directory.  To do this, you must use a version of `make' that
+ own directory.  To do this, you can use GNU `make'.  `cd' to the
- supports the `VPATH' variable, such as GNU `make'.  `cd' to the
  directory where you want the object files and executables to go and run
  the `configure' script.  `configure' automatically checks for the
- source code in the directory that `configure' is in and in `..'.
+ source code in the directory that `configure' is in and in `..'.  This
+ is known as a "VPATH" build.
-    If you have to use a `make' that does not supports the `VPATH'
+    With a non-GNU `make', it is safer to compile the package for one
- variable, you have to compile the package for one architecture at a time
+ architecture at a time in the source code directory.  After you have
- in the source code directory.  After you have installed the package for
+ installed the package for one architecture, use `make distclean' before
- one architecture, use `make distclean' before reconfiguring for another
+ reconfiguring for another architecture.
- architecture.
+    On MacOS X 10.5 and later systems, you can create libraries and
+ executables that work on multiple system types--known as "fat" or
+ "universal" binaries--by specifying multiple `-arch' options to the
+ compiler but only a single `-arch' option to the preprocessor.  Like
+ this:
+      ./configure CC="gcc -arch i386 -arch x86_64 -arch ppc -arch ppc64" \
+                  CXX="g++ -arch i386 -arch x86_64 -arch ppc -arch ppc64" \
+                  CPP="gcc -E" CXXCPP="g++ -E"
+    This is not guaranteed to produce working output in all cases, you
+ may have to build one architecture at a time and combine the results
+ using the `lipo' tool if you have problems.
  Installation Names
  ==================
-    By default, `make install' will install the package's files in
+    By default, `make install' installs the package's commands under
- `/usr/local/bin', `/usr/local/man', etc.  You can specify an
+ `/usr/local/bin', include files under `/usr/local/include', etc.  You
- installation prefix other than `/usr/local' by giving `configure' the
+ can specify an installation prefix other than `/usr/local' by giving
- option `--prefix=PATH'.
+ `configure' the option `--prefix=PREFIX', where PREFIX must be an
+ absolute file name.
     You can specify separate installation prefixes for
  architecture-specific files and architecture-independent files.  If you
- give `configure' the option `--exec-prefix=PATH', the package will use
+ pass the option `--exec-prefix=PREFIX' to `configure', the package uses
- PATH as the prefix for installing programs and libraries.
+ PREFIX as the prefix for installing programs and libraries.
- Documentation and other data files will still use the regular prefix.
+ Documentation and other data files still use the regular prefix.
+    In addition, if you use an unusual directory layout you can give
+ options like `--bindir=DIR' to specify different values for particular
+ kinds of files.  Run `configure --help' for a list of the directories
+ you can set and what kinds of files go in them.  In general, the
+ default for these options is expressed in terms of `${prefix}', so that
+ specifying just `--prefix' will affect all of the other directory
+ specifications that were not explicitly provided.
+    The most portable way to affect installation locations is to pass the
+ correct locations to `configure'; however, many packages provide one or
+ both of the following shortcuts of passing variable assignments to the
+ `make install' command line to change installation locations without
+ having to reconfigure or recompile.
+    The first method involves providing an override variable for each
+ affected directory.  For example, `make install
+ prefix=/alternate/directory' will choose an alternate location for all
+ directory configuration variables that were expressed in terms of
+ `${prefix}'.  Any directories that were specified during `configure',
+ but not in terms of `${prefix}', must each be overridden at install
+ time for the entire installation to be relocated.  The approach of
+ makefile variable overrides for each directory variable is required by
+ the GNU Coding Standards, and ideally causes no recompilation.
+ However, some platforms have known limitations with the semantics of
+ shared libraries that end up requiring recompilation when using this
+ method, particularly noticeable in packages that use GNU Libtool.
+    The second method involves providing the `DESTDIR' variable.  For
+ example, `make install DESTDIR=/alternate/directory' will prepend
+ `/alternate/directory' before all installation names.  The approach of
+ `DESTDIR' overrides is not required by the GNU Coding Standards, and
+ does not work on platforms that have drive letters.  On the other hand,
+ it does better at avoiding recompilation issues, and works well even
+ when some directory options were not specified in terms of `${prefix}'
+ at `configure' time.
+ Optional Features
+ =================
     If the package supports it, you can cause programs to be installed
  with an extra prefix or suffix on their names by giving `configure' the
  option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'.
- Optional Features
- =================
     Some packages pay attention to `--enable-FEATURE' options to
  `configure', where FEATURE indicates an optional part of the package.
  They may also pay attention to `--with-PACKAGE' options, where PACKAGE
-Line 105 
 find the X include and library files aut
+Line 208 
 find the X include and library files aut
  you can use the `configure' options `--x-includes=DIR' and
  `--x-libraries=DIR' to specify their locations.
+    Some packages offer the ability to configure how verbose the
+ execution of `make' will be.  For these packages, running `./configure
+ --enable-silent-rules' sets the default to minimal output, which can be
+ overridden with `make V=1'; while running `./configure
+ --disable-silent-rules' sets the default to verbose, which can be
+ overridden with `make V=0'.
+ Particular systems
+ ==================
+    On HP-UX, the default C compiler is not ANSI C compatible.  If GNU
+ CC is not installed, it is recommended to use the following options in
+ order to use an ANSI C compiler:
+      ./configure CC="cc -Ae -D_XOPEN_SOURCE=500"
+ and if that doesn't work, install pre-built binaries of GCC for HP-UX.
+    On OSF/1 a.k.a. Tru64, some versions of the default C compiler cannot
+ parse its `<wchar.h>' header file.  The option `-nodtk' can be used as
+ a workaround.  If GNU CC is not installed, it is therefore recommended
+ to try
+      ./configure CC="cc"
+ and if that doesn't work, try
+      ./configure CC="cc -nodtk"
+    On Solaris, don't put `/usr/ucb' early in your `PATH'.  This
+ directory contains several dysfunctional programs; working variants of
+ these programs are available in `/usr/bin'.  So, if you need `/usr/ucb'
+ in your `PATH', put it _after_ `/usr/bin'.
+    On Haiku, software installed for all users goes in `/boot/common',
+ not `/usr/local'.  It is recommended to use the following options:
+      ./configure --prefix=/boot/common
  Specifying the System Type
  ==========================
-    There may be some features `configure' can not figure out
+    There may be some features `configure' cannot figure out
- automatically, but needs to determine by the type of host the package
+ automatically, but needs to determine by the type of machine the package
- will run on.  Usually `configure' can figure that out, but if it prints
+ will run on.  Usually, assuming the package is built to be run on the
- a message saying it can not guess the host type, give it the
+ _same_ architectures, `configure' can figure that out, but if it prints
- `--host=TYPE' option.  TYPE can either be a short name for the system
+ a message saying it cannot guess the machine type, give it the
- type, such as `sun4', or a canonical name with three fields:
+ `--build=TYPE' option.  TYPE can either be a short name for the system
+ type, such as `sun4', or a canonical name which has the form:
       CPU-COMPANY-SYSTEM
- See the file `config.sub' for the possible values of each field.  If
+ where SYSTEM can have one of these forms:
+      OS
+      KERNEL-OS
+    See the file `config.sub' for the possible values of each field.  If
  `config.sub' isn't included in this package, then this package doesn't
- need to know the host type.
+ need to know the machine type.
-    If you are building compiler tools for cross-compiling, you can also
+    If you are _building_ compiler tools for cross-compiling, you should
- use the `--target=TYPE' option to select the type of system they will
+ use the option `--target=TYPE' to select the type of system they will
- produce code for and the `--build=TYPE' option to select the type of
+ produce code for.
- system on which you are compiling the package.
+    If you want to _use_ a cross compiler, that generates code for a
+ platform different from the build platform, you should specify the
+ "host" platform (i.e., that on which the generated programs will
+ eventually be run) with `--host=TYPE'.
  Sharing Defaults
  ================
-Line 136 
 default values for variables like `CC',
+Line 289 
 default values for variables like `CC',
  `CONFIG_SITE' environment variable to the location of the site script.
  A warning: not all `configure' scripts look for a site script.
- Operation Controls
+ Defining Variables
  ==================
+    Variables not defined in a site shell script can be set in the
+ environment passed to `configure'.  However, some packages may run
+ configure again during the build, and the customized values of these
+ variables may be lost.  In order to avoid this problem, you should set
+ them in the `configure' command line, using `VAR=value'.  For example:
+      ./configure CC=/usr/local2/bin/gcc
+ causes the specified `gcc' to be used as the C compiler (unless it is
+ overridden in the site shell script).
+ Unfortunately, this technique does not work for `CONFIG_SHELL' due to
+ an Autoconf bug.  Until the bug is fixed you can use this workaround:
+      CONFIG_SHELL=/bin/bash /bin/bash ./configure CONFIG_SHELL=/bin/bash
+ `configure' Invocation
+ ======================
     `configure' recognizes the following options to control how it
  operates.
- `--cache-file=FILE'
-      Use and save the results of the tests in FILE instead of
-      `./config.cache'.  Set FILE to `/dev/null' to disable caching, for
-      debugging `configure'.
  `--help'
-      Print a summary of the options to `configure', and exit.
+ `-h'
+      Print a summary of all of the options to `configure', and exit.
+ `--help=short'
+ `--help=recursive'
+      Print a summary of the options unique to this package's
+      `configure', and exit.  The `short' variant lists options used
+      only in the top level, while the `recursive' variant lists options
+      also present in any nested packages.
+ `--version'
+ `-V'
+      Print the version of Autoconf used to generate the `configure'
+      script, and exit.
+ `--cache-file=FILE'
+      Enable the cache: use and save the results of the tests in FILE,
+      traditionally `config.cache'.  FILE defaults to `/dev/null' to
+      disable caching.
+ `--config-cache'
+ `-C'
+      Alias for `--cache-file=config.cache'.
  `--quiet'
  `--silent'
  `-q'
-      Do not print messages saying which checks are being made.
+      Do not print messages saying which checks are being made.  To
+      suppress all normal output, redirect it to `/dev/null' (any error
+      messages will still be shown).
  `--srcdir=DIR'
       Look for the package's source code in directory DIR.  Usually
       `configure' can determine that directory automatically.
- `--version'
+ `--prefix=DIR'
-      Print the version of Autoconf used to generate the `configure'
+      Use DIR as the installation prefix.  *note Installation Names::
-      script, and exit.
+      for more details, including other options available for fine-tuning
+      the installation locations.
+ `--no-create'
+ `-n'
+      Run the configure checks, but stop before creating any output
+      files.
- `configure' also accepts some other, not widely useful, options.
+ `configure' also accepts some other, not widely useful, options.  Run
+ `configure --help' for more details.

 Legend:



Removed from v.2319
 


changed lines


 
Added in v.2320
 Legend:



Removed from v.2319
 


changed lines


 
Added in v.2320
-Removed from v.2319
+Added in v.2320

	ViewVC Help
Powered by ViewVC