Diff for /gforth/Attic/gforth.ds between versions 1.1 and 1.20

version 1.1, 1994/10/24 19:15:57 version 1.20, 1995/10/29 21:35:11
Line 1 Line 1
 \input texinfo   @c -*-texinfo-*-  \input texinfo   @c -*-texinfo-*-
 @comment The source is gforth.ds, from which gforth.texi is generated  @comment The source is gforth.ds, from which gforth.texi is generated
 @comment %**start of header (This is for running Texinfo on a region.)  @comment %**start of header (This is for running Texinfo on a region.)
 @setfilename gforth-info  @setfilename gforth.info
 @settitle GNU Forth Manual  @settitle Gforth Manual
 @setchapternewpage odd  @comment @setchapternewpage odd
 @comment %**end of header (This is for running Texinfo on a region.)  @comment %**end of header (This is for running Texinfo on a region.)
   
 @ifinfo  @ifinfo
 This file documents GNU Forth 0.0  This file documents Gforth 0.1
   
 Copyright @copyright{} 1994 GNU Forth Development Group  Copyright @copyright{} 1994 Gforth Development Group
   
      Permission is granted to make and distribute verbatim copies of       Permission is granted to make and distribute verbatim copies of
      this manual provided the copyright notice and this permission notice       this manual provided the copyright notice and this permission notice
      are preserved on all copies.       are preserved on all copies.
             
      @ignore  @ignore
      Permission is granted to process this file through TeX and print the       Permission is granted to process this file through TeX and print the
      results, provided the printed document carries a copying permission       results, provided the printed document carries a copying permission
      notice identical to this one except for the removal of this paragraph       notice identical to this one except for the removal of this paragraph
      (this paragraph not being relevant to the printed manual).       (this paragraph not being relevant to the printed manual).
             
      @end ignore  @end ignore
      Permission is granted to copy and distribute modified versions of this       Permission is granted to copy and distribute modified versions of this
      manual under the conditions for verbatim copying, provided also that the       manual under the conditions for verbatim copying, provided also that the
      sections entitled "Distribution" and "General Public License" are       sections entitled "Distribution" and "General Public License" are
Line 38  Copyright @copyright{} 1994 GNU Forth De Line 38  Copyright @copyright{} 1994 GNU Forth De
   
 @titlepage  @titlepage
 @sp 10  @sp 10
 @center @titlefont{GNU Forth Manual}  @center @titlefont{Gforth Manual}
 @sp 2  @sp 2
 @center for version 0.0  @center for version 0.1
 @sp 2  @sp 2
 @center Anton Ertl  @center Anton Ertl
   @sp 3
   @center This manual is under construction
   
 @comment  The following two commands start the copyright page.  @comment  The following two commands start the copyright page.
 @page  @page
 @vskip 0pt plus 1filll  @vskip 0pt plus 1filll
 Copyright @copyright{} 1994 GNU Forth Development Group  Copyright @copyright{} 1994 Gforth Development Group
   
 @comment !! Published by ... or You can get a copy of this manual ...  @comment !! Published by ... or You can get a copy of this manual ...
   
Line 72  Copyright @copyright{} 1994 GNU Forth De Line 74  Copyright @copyright{} 1994 GNU Forth De
   
 @node Top, License, (dir), (dir)  @node Top, License, (dir), (dir)
 @ifinfo  @ifinfo
 GNU Forth is a free implementation of ANS Forth available on many  Gforth is a free implementation of ANS Forth available on many
 personal machines. This manual corresponds to version 0.0.  personal machines. This manual corresponds to version 0.0.
 @end ifinfo  @end ifinfo
   
 @menu  @menu
 * License::               * License::                     
 * Goals::               About the GNU Forth Project  * Goals::                       About the Gforth Project
 * Other Books::         Things you might want to read  * Other Books::                 Things you might want to read
 * Invocation::          Starting GNU Forth  * Invocation::                  Starting Gforth
 * Words::               Forth words available in GNU Forth  * Words::                       Forth words available in Gforth
 * ANS conformance::     Implementation-defined options etc.  * ANS conformance::             Implementation-defined options etc.
 * Model::               The abstract machine of GNU Forth  * Model::                       The abstract machine of Gforth
 * Emacs and GForth::    The GForth Mode  * Emacs and Gforth::            The Gforth Mode
 * Internals::           Implementation details  * Internals::                   Implementation details
 * Bugs::                How to report them  * Bugs::                        How to report them
 * Pedigree::            Ancestors of GNU Forth  * Pedigree::                    Ancestors of Gforth
 * Word Index::          An item for each Forth word  * Word Index::                  An item for each Forth word
 * Node Index::          An item for each node  * Node Index::                  An item for each node
 @end menu  @end menu
   
 @node License, Goals, Top, Top  @node License, Goals, Top, Top
 @unnumbered License  @unnumbered GNU GENERAL PUBLIC LICENSE
 !! Insert GPL here  @center Version 2, June 1991
   
   @display
   Copyright @copyright{} 1989, 1991 Free Software Foundation, Inc.
   675 Mass Ave, Cambridge, MA 02139, USA
   
   Everyone is permitted to copy and distribute verbatim copies
   of this license document, but changing it is not allowed.
   @end display
   
   @unnumberedsec Preamble
   
     The licenses for most software are designed to take away your
   freedom to share and change it.  By contrast, the GNU General Public
   License is intended to guarantee your freedom to share and change free
   software---to make sure the software is free for all its users.  This
   General Public License applies to most of the Free Software
   Foundation's software and to any other program whose authors commit to
   using it.  (Some other Free Software Foundation software is covered by
   the GNU Library General Public License instead.)  You can apply it to
   your programs, too.
   
     When we speak of free software, we are referring to freedom, not
   price.  Our General Public Licenses are designed to make sure that you
   have the freedom to distribute copies of free software (and charge for
   this service if you wish), that you receive source code or can get it
   if you want it, that you can change the software or use pieces of it
   in new free programs; and that you know you can do these things.
   
     To protect your rights, we need to make restrictions that forbid
   anyone to deny you these rights or to ask you to surrender the rights.
   These restrictions translate to certain responsibilities for you if you
   distribute copies of the software, or if you modify it.
   
     For example, if you distribute copies of such a program, whether
   gratis or for a fee, you must give the recipients all the rights that
   you have.  You must make sure that they, too, receive or can get the
   source code.  And you must show them these terms so they know their
   rights.
   
     We protect your rights with two steps: (1) copyright the software, and
   (2) offer you this license which gives you legal permission to copy,
   distribute and/or modify the software.
   
     Also, for each author's protection and ours, we want to make certain
   that everyone understands that there is no warranty for this free
   software.  If the software is modified by someone else and passed on, we
   want its recipients to know that what they have is not the original, so
   that any problems introduced by others will not reflect on the original
   authors' reputations.
   
     Finally, any free program is threatened constantly by software
   patents.  We wish to avoid the danger that redistributors of a free
   program will individually obtain patent licenses, in effect making the
   program proprietary.  To prevent this, we have made it clear that any
   patent must be licensed for everyone's free use or not licensed at all.
   
     The precise terms and conditions for copying, distribution and
   modification follow.
   
   @iftex
   @unnumberedsec TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
   @end iftex
   @ifinfo
   @center TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
   @end ifinfo
   
   @enumerate 0
   @item
   This License applies to any program or other work which contains
   a notice placed by the copyright holder saying it may be distributed
   under the terms of this General Public License.  The ``Program'', below,
   refers to any such program or work, and a ``work based on the Program''
   means either the Program or any derivative work under copyright law:
   that is to say, a work containing the Program or a portion of it,
   either verbatim or with modifications and/or translated into another
   language.  (Hereinafter, translation is included without limitation in
   the term ``modification''.)  Each licensee is addressed as ``you''.
   
   Activities other than copying, distribution and modification are not
   covered by this License; they are outside its scope.  The act of
   running the Program is not restricted, and the output from the Program
   is covered only if its contents constitute a work based on the
   Program (independent of having been made by running the Program).
   Whether that is true depends on what the Program does.
   
   @item
   You may copy and distribute verbatim copies of the Program's
   source code as you receive it, in any medium, provided that you
   conspicuously and appropriately publish on each copy an appropriate
   copyright notice and disclaimer of warranty; keep intact all the
   notices that refer to this License and to the absence of any warranty;
   and give any other recipients of the Program a copy of this License
   along with the Program.
   
   You may charge a fee for the physical act of transferring a copy, and
   you may at your option offer warranty protection in exchange for a fee.
   
   @item
   You may modify your copy or copies of the Program or any portion
   of it, thus forming a work based on the Program, and copy and
   distribute such modifications or work under the terms of Section 1
   above, provided that you also meet all of these conditions:
   
   @enumerate a
   @item
   You must cause the modified files to carry prominent notices
   stating that you changed the files and the date of any change.
   
   @item
   You must cause any work that you distribute or publish, that in
   whole or in part contains or is derived from the Program or any
   part thereof, to be licensed as a whole at no charge to all third
   parties under the terms of this License.
   
   @item
   If the modified program normally reads commands interactively
   when run, you must cause it, when started running for such
   interactive use in the most ordinary way, to print or display an
   announcement including an appropriate copyright notice and a
   notice that there is no warranty (or else, saying that you provide
   a warranty) and that users may redistribute the program under
   these conditions, and telling the user how to view a copy of this
   License.  (Exception: if the Program itself is interactive but
   does not normally print such an announcement, your work based on
   the Program is not required to print an announcement.)
   @end enumerate
   
   These requirements apply to the modified work as a whole.  If
   identifiable sections of that work are not derived from the Program,
   and can be reasonably considered independent and separate works in
   themselves, then this License, and its terms, do not apply to those
   sections when you distribute them as separate works.  But when you
   distribute the same sections as part of a whole which is a work based
   on the Program, the distribution of the whole must be on the terms of
   this License, whose permissions for other licensees extend to the
   entire whole, and thus to each and every part regardless of who wrote it.
   
   Thus, it is not the intent of this section to claim rights or contest
   your rights to work written entirely by you; rather, the intent is to
   exercise the right to control the distribution of derivative or
   collective works based on the Program.
   
   In addition, mere aggregation of another work not based on the Program
   with the Program (or with a work based on the Program) on a volume of
   a storage or distribution medium does not bring the other work under
   the scope of this License.
   
   @item
   You may copy and distribute the Program (or a work based on it,
   under Section 2) in object code or executable form under the terms of
   Sections 1 and 2 above provided that you also do one of the following:
   
   @enumerate a
   @item
   Accompany it with the complete corresponding machine-readable
   source code, which must be distributed under the terms of Sections
   1 and 2 above on a medium customarily used for software interchange; or,
   
   @item
   Accompany it with a written offer, valid for at least three
   years, to give any third party, for a charge no more than your
   cost of physically performing source distribution, a complete
   machine-readable copy of the corresponding source code, to be
   distributed under the terms of Sections 1 and 2 above on a medium
   customarily used for software interchange; or,
   
   @item
   Accompany it with the information you received as to the offer
   to distribute corresponding source code.  (This alternative is
   allowed only for noncommercial distribution and only if you
   received the program in object code or executable form with such
   an offer, in accord with Subsection b above.)
   @end enumerate
   
   The source code for a work means the preferred form of the work for
   making modifications to it.  For an executable work, complete source
   code means all the source code for all modules it contains, plus any
   associated interface definition files, plus the scripts used to
   control compilation and installation of the executable.  However, as a
   special exception, the source code distributed need not include
   anything that is normally distributed (in either source or binary
   form) with the major components (compiler, kernel, and so on) of the
   operating system on which the executable runs, unless that component
   itself accompanies the executable.
   
   If distribution of executable or object code is made by offering
   access to copy from a designated place, then offering equivalent
   access to copy the source code from the same place counts as
   distribution of the source code, even though third parties are not
   compelled to copy the source along with the object code.
   
   @item
   You may not copy, modify, sublicense, or distribute the Program
   except as expressly provided under this License.  Any attempt
   otherwise to copy, modify, sublicense or distribute the Program is
   void, and will automatically terminate your rights under this License.
   However, parties who have received copies, or rights, from you under
   this License will not have their licenses terminated so long as such
   parties remain in full compliance.
   
   @item
   You are not required to accept this License, since you have not
   signed it.  However, nothing else grants you permission to modify or
   distribute the Program or its derivative works.  These actions are
   prohibited by law if you do not accept this License.  Therefore, by
   modifying or distributing the Program (or any work based on the
   Program), you indicate your acceptance of this License to do so, and
   all its terms and conditions for copying, distributing or modifying
   the Program or works based on it.
   
   @item
   Each time you redistribute the Program (or any work based on the
   Program), the recipient automatically receives a license from the
   original licensor to copy, distribute or modify the Program subject to
   these terms and conditions.  You may not impose any further
   restrictions on the recipients' exercise of the rights granted herein.
   You are not responsible for enforcing compliance by third parties to
   this License.
   
   @item
   If, as a consequence of a court judgment or allegation of patent
   infringement or for any other reason (not limited to patent issues),
   conditions are imposed on you (whether by court order, agreement or
   otherwise) that contradict the conditions of this License, they do not
   excuse you from the conditions of this License.  If you cannot
   distribute so as to satisfy simultaneously your obligations under this
   License and any other pertinent obligations, then as a consequence you
   may not distribute the Program at all.  For example, if a patent
   license would not permit royalty-free redistribution of the Program by
   all those who receive copies directly or indirectly through you, then
   the only way you could satisfy both it and this License would be to
   refrain entirely from distribution of the Program.
   
   If any portion of this section is held invalid or unenforceable under
   any particular circumstance, the balance of the section is intended to
   apply and the section as a whole is intended to apply in other
   circumstances.
   
   It is not the purpose of this section to induce you to infringe any
   patents or other property right claims or to contest validity of any
   such claims; this section has the sole purpose of protecting the
   integrity of the free software distribution system, which is
   implemented by public license practices.  Many people have made
   generous contributions to the wide range of software distributed
   through that system in reliance on consistent application of that
   system; it is up to the author/donor to decide if he or she is willing
   to distribute software through any other system and a licensee cannot
   impose that choice.
   
   This section is intended to make thoroughly clear what is believed to
   be a consequence of the rest of this License.
   
   @item
   If the distribution and/or use of the Program is restricted in
   certain countries either by patents or by copyrighted interfaces, the
   original copyright holder who places the Program under this License
   may add an explicit geographical distribution limitation excluding
   those countries, so that distribution is permitted only in or among
   countries not thus excluded.  In such case, this License incorporates
   the limitation as if written in the body of this License.
   
   @item
   The Free Software Foundation may publish revised and/or new versions
   of the General Public License from time to time.  Such new versions will
   be similar in spirit to the present version, but may differ in detail to
   address new problems or concerns.
   
   Each version is given a distinguishing version number.  If the Program
   specifies a version number of this License which applies to it and ``any
   later version'', you have the option of following the terms and conditions
   either of that version or of any later version published by the Free
   Software Foundation.  If the Program does not specify a version number of
   this License, you may choose any version ever published by the Free Software
   Foundation.
   
   @item
   If you wish to incorporate parts of the Program into other free
   programs whose distribution conditions are different, write to the author
   to ask for permission.  For software which is copyrighted by the Free
   Software Foundation, write to the Free Software Foundation; we sometimes
   make exceptions for this.  Our decision will be guided by the two goals
   of preserving the free status of all derivatives of our free software and
   of promoting the sharing and reuse of software generally.
   
   @iftex
   @heading NO WARRANTY
   @end iftex
   @ifinfo
   @center NO WARRANTY
   @end ifinfo
   
   @item
   BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
   FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW.  EXCEPT WHEN
   OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
   PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
   OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
   MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.  THE ENTIRE RISK AS
   TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU.  SHOULD THE
   PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
   REPAIR OR CORRECTION.
   
   @item
   IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
   WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
   REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
   INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
   OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
   TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
   YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
   PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
   POSSIBILITY OF SUCH DAMAGES.
   @end enumerate
   
   @iftex
   @heading END OF TERMS AND CONDITIONS
   @end iftex
   @ifinfo
   @center END OF TERMS AND CONDITIONS
   @end ifinfo
   
   @page
   @unnumberedsec How to Apply These Terms to Your New Programs
   
     If you develop a new program, and you want it to be of the greatest
   possible use to the public, the best way to achieve this is to make it
   free software which everyone can redistribute and change under these terms.
   
     To do so, attach the following notices to the program.  It is safest
   to attach them to the start of each source file to most effectively
   convey the exclusion of warranty; and each file should have at least
   the ``copyright'' line and a pointer to where the full notice is found.
   
   @smallexample
   @var{one line to give the program's name and a brief idea of what it does.}
   Copyright (C) 19@var{yy}  @var{name of author}
   
   This program is free software; you can redistribute it and/or modify 
   it under the terms of the GNU General Public License as published by 
   the Free Software Foundation; either version 2 of the License, or 
   (at your option) any later version.
   
   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.
   
   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
   @end smallexample
   
   Also add information on how to contact you by electronic and paper mail.
   
   If the program is interactive, make it output a short notice like this
   when it starts in an interactive mode:
   
   @smallexample
   Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
   Gnomovision comes with ABSOLUTELY NO WARRANTY; for details
   type `show w'.  
   This is free software, and you are welcome to redistribute it 
   under certain conditions; type `show c' for details.
   @end smallexample
   
   The hypothetical commands @samp{show w} and @samp{show c} should show
   the appropriate parts of the General Public License.  Of course, the
   commands you use may be called something other than @samp{show w} and
   @samp{show c}; they could even be mouse-clicks or menu items---whatever
   suits your program.
   
   You should also get your employer (if you work as a programmer) or your
   school, if any, to sign a ``copyright disclaimer'' for the program, if
   necessary.  Here is a sample; alter the names:
   
   @smallexample
   Yoyodyne, Inc., hereby disclaims all copyright interest in the program
   `Gnomovision' (which makes passes at compilers) written by James Hacker.
   
   @var{signature of Ty Coon}, 1 April 1989
   Ty Coon, President of Vice
   @end smallexample
   
   This General Public License does not permit incorporating your program into
   proprietary programs.  If your program is a subroutine library, you may
   consider it more useful to permit linking proprietary applications with the
   library.  If this is what you want to do, use the GNU Library General
   Public License instead of this License.
   
 @iftex  @iftex
 @unnumbered Preface  @unnumbered Preface
 This manual documents GNU Forth. The reader is expected to know  This manual documents Gforth. The reader is expected to know
 Forth. This manual is primarily a reference manual. @xref{Other Books}  Forth. This manual is primarily a reference manual. @xref{Other Books}
 for introductory material.  for introductory material.
 @end iftex  @end iftex
   
 @node    Goals, Other Books, License, Top  @node    Goals, Other Books, License, Top
 @comment node-name,     next,           previous, up  @comment node-name,     next,           previous, up
 @chapter Goals of GNU Forth  @chapter Goals of Gforth
 @cindex Goals  @cindex Goals
 The goal of the GNU Forth Project is to develop a standard model for  The goal of the Gforth Project is to develop a standard model for
 ANSI Forth. This can be split into several subgoals:  ANSI Forth. This can be split into several subgoals:
   
 @itemize @bullet  @itemize @bullet
 @item  @item
 GNU Forth should conform to the ANSI Forth standard.  Gforth should conform to the ANSI Forth standard.
 @item  @item
 It should be a model, i.e. it should define all the  It should be a model, i.e. it should define all the
 implementation-dependent things.  implementation-dependent things.
Line 121  It should become standard, i.e. widely a Line 511  It should become standard, i.e. widely a
 is the most difficult one.  is the most difficult one.
 @end itemize  @end itemize
   
 To achieve these goals GNU Forth should be  To achieve these goals Gforth should be
 @itemize @bullet  @itemize @bullet
 @item  @item
 Similar to previous models (fig-Forth, F83)  Similar to previous models (fig-Forth, F83)
Line 137  Free. Line 527  Free.
 Available on many machines/easy to port.  Available on many machines/easy to port.
 @end itemize  @end itemize
   
 Have we achieved these goals? GNU Forth conforms to the ANS Forth  Have we achieved these goals? Gforth conforms to the ANS Forth
 standard; it may be considered a model, but we have not yet documented  standard. It may be considered a model, but we have not yet documented
 which parts of the model are stable and which parts we are likely to  which parts of the model are stable and which parts we are likely to
 change; it certainly has not yet become a de facto standard. It has some  change. It certainly has not yet become a de facto standard. It has some
 similarities and some differences to previous models; It has some  similarities and some differences to previous models. It has some
 powerful features, but not yet everything that we envisioned; on RISCs  powerful features, but not yet everything that we envisioned. We
 it is as fast as interpreters programmed in assembly, on  certainly have achieved our execution speed goals (@pxref{Performance}).
 register-starved machines it is not so fast, but still faster than any  It is free and available on many machines.
 other C-based interpretive implementation; it is free and available on  
 many machines.  
   
 @node Other Books, Invocation, Goals, Top  @node Other Books, Invocation, Goals, Top
 @chapter Other books on ANS Forth  @chapter Other books on ANS Forth
   
 As the standard is relatively new, there are not many books out yet. It  As the standard is relatively new, there are not many books out yet. It
 is not recommended to learn Forth by using GNU Forth and a book that is  is not recommended to learn Forth by using Gforth and a book that is
 not written for ANS Forth, as you will not know your mistakes from the  not written for ANS Forth, as you will not know your mistakes from the
 deviations of the book.  deviations of the book.
   
 There is, of course, the standard, the definite reference if you want to  There is, of course, the standard, the definite reference if you want to
 write ANS Forth programs. It will be available in printed form from  write ANS Forth programs. It is available in printed form from the
 Global Engineering Documents !! somtime in spring or summer 1994. If you  National Standards Institute Sales Department (Tel.: USA (212) 642-4900;
 are lucky, you can still get dpANS6 (the draft that was approved as  Fax.: USA (212) 302-1286) as document @cite{X3.215-1994} for about $200. You
 standard) by aftp from ftp.uu.net:/vendor/minerva/x3j14.  can also get it from Global Engineering Documents (Tel.: USA (800)
   854-7179; Fax.: (303) 843-9880) for about $300.
   
   @cite{dpANS6}, the last draft of the standard, which was then submitted to ANSI
   for publication is available electronically and for free in some MS Word
   format, and it has been converted to HTML. Some pointers to these
   versions can be found through
   http://www.complang.tuwien.ac.at/projects/forth.html.
   
 @cite{Forth: The new model} by Jack Woehr (!! Publisher) is an  @cite{Forth: The new model} by Jack Woehr (!! Publisher) is an
 introductory book based on a draft version of the standard. It does not  introductory book based on a draft version of the standard. It does not
Line 173  other languages should find it ok. Line 568  other languages should find it ok.
 @chapter Invocation  @chapter Invocation
   
 You will usually just say @code{gforth}. In many other cases the default  You will usually just say @code{gforth}. In many other cases the default
 GNU Forth image will be invoked like this:  Gforth image will be invoked like this:
   
 @example  @example
 gforth [files] [-e forth-code]  gforth [files] [-e forth-code]
Line 193  line. They are: Line 588  line. They are:
   
 @table @code  @table @code
 @item --image-file @var{file}  @item --image-file @var{file}
   @item -i @var{file}
 Loads the Forth image @var{file} instead of the default  Loads the Forth image @var{file} instead of the default
 @file{gforth.fi}.  @file{gforth.fi}.
   
 @item --path @var{path}  @item --path @var{path}
   @item -p @var{path}
 Uses @var{path} for searching the image file and Forth source code  Uses @var{path} for searching the image file and Forth source code
 files instead of the default in the environment variable  files instead of the default in the environment variable
 @code{GFORTHPATH} or the path specified at installation time (typically  @code{GFORTHPATH} or the path specified at installation time (typically
Line 253  the user initialization file @file{.gfor Line 650  the user initialization file @file{.gfor
 option @code{--no-rc} is given; this file is first searched in @file{.},  option @code{--no-rc} is given; this file is first searched in @file{.},
 then in @file{~}, then in the normal path (see above).  then in @file{~}, then in the normal path (see above).
   
 @node Words,  , Invocation, Top  @node Words, ANS conformance, Invocation, Top
 @chapter Forth Words  @chapter Forth Words
   
 @menu  @menu
 * Notation::  * Notation::                    
 * Arithmetic::  * Arithmetic::                  
 * Stack Manipulation::  * Stack Manipulation::          
 * Memory access::  * Memory access::               
 * Control Structures::  * Control Structures::          
 * Local Variables::  * Locals::                      
 * Defining Words::  * Defining Words::              
 * Vocabularies::  * Wordlists::                   
 * Files::  * Files::                       
 * Blocks::  * Blocks::                      
 * Other I/O::  * Other I/O::                   
 * Programming Tools::  * Programming Tools::           
   * Assembler and Code words::    
   * Threading Words::             
 @end menu  @end menu
   
 @node Notation, Arithmetic, Words, Words  @node Notation, Arithmetic, Words, Words
Line 277  then in @file{~}, then in the normal pat Line 676  then in @file{~}, then in the normal pat
 The Forth words are described in this section in the glossary notation  The Forth words are described in this section in the glossary notation
 that has become a de-facto standard for Forth texts, i.e.  that has become a de-facto standard for Forth texts, i.e.
   
 @quotation  @format
 @var{word}     @var{Stack effect}   @var{wordset}   @var{pronunciation}  @var{word}     @var{Stack effect}   @var{wordset}   @var{pronunciation}
   @end format
 @var{Description}  @var{Description}
 @end quotation  
   
 @table @var  @table @var
 @item word  @item word
 The name of the word. BTW, GNU Forth is case insensitive, so you can  The name of the word. BTW, Gforth is case insensitive, so you can
 type the words in in lower case.  type the words in in lower case (However, @pxref{core-idef}).
   
 @item Stack effect  @item Stack effect
 The stack effect is written in the notation @code{@var{before} --  The stack effect is written in the notation @code{@var{before} --
 @var{after}}, where @var{before} and @var{after} describe the top of  @var{after}}, where @var{before} and @var{after} describe the top of
 stack entries before and after the execution of the word. The rest of  stack entries before and after the execution of the word. The rest of
 the stack is not touched by the word. The top of stack is rightmost,  the stack is not touched by the word. The top of stack is rightmost,
 i.e., a stack sequence is written as it is typed in. Note that GNU Forth  i.e., a stack sequence is written as it is typed in. Note that Gforth
 uses a separate floating point stack, but a unified stack  uses a separate floating point stack, but a unified stack
 notation. Also, return stack effects are not shown in @var{stack  notation. Also, return stack effects are not shown in @var{stack
 effect}, but in @var{Description}. The name of a stack item describes  effect}, but in @var{Description}. The name of a stack item describes
 the type and/or the function of the item. See below for a discussion of  the type and/or the function of the item. See below for a discussion of
 the types.  the types.
   
   All words have two stack effects: A compile-time stack effect and a
   run-time stack effect. The compile-time stack-effect of most words is
   @var{ -- }. If the compile-time stack-effect of a word deviates from
   this standard behaviour, or the word does other unusual things at
   compile time, both stack effects are shown; otherwise only the run-time
   stack effect is shown.
   
 @item pronunciation  @item pronunciation
 How the word is pronounced  How the word is pronounced
   
Line 308  system need not support all of them. So, Line 714  system need not support all of them. So,
 uses the more portable it will be in theory. However, we suspect that  uses the more portable it will be in theory. However, we suspect that
 most ANS Forth systems on personal machines will feature all  most ANS Forth systems on personal machines will feature all
 wordsets. Words that are not defined in the ANS standard have  wordsets. Words that are not defined in the ANS standard have
 @code{gforth} as wordset.  @code{gforth} or @code{gforth-internal} as wordset. @code{gforth}
   describes words that will work in future releases of Gforth;
   @code{gforth-internal} words are more volatile. Environmental query
   strings are also displayed like words; you can recognize them by the
   @code{environment} in the wordset field.
   
 @item Description  @item Description
 A description of the behaviour of the word.  A description of the behaviour of the word.
 @end table  @end table
   
 The name of a stack item corresponds in the following way with its type:  The type of a stack item is specified by the character(s) the name
   starts with:
   
 @table @code  @table @code
 @item name starts with  
 Type  
 @item f  @item f
 Bool, i.e. @code{false} or @code{true}.  Bool, i.e. @code{false} or @code{true}.
 @item c  @item c
Line 353  Wordlist ID, same size as Cell Line 762  Wordlist ID, same size as Cell
 Pointer to a name structure  Pointer to a name structure
 @end table  @end table
   
 @node Arithmetic,  , Notation, Words  @node Arithmetic, Stack Manipulation, Notation, Words
 @section Arithmetic  @section Arithmetic
 Forth arithmetic is not checked, i.e., you will not hear about integer  Forth arithmetic is not checked, i.e., you will not hear about integer
 overflow on addition or multiplication, you may hear about division by  overflow on addition or multiplication, you may hear about division by
Line 363  corresponds to @code{2 1 -}. Forth offer Line 772  corresponds to @code{2 1 -}. Forth offer
 operators. If you perform division with potentially negative operands,  operators. If you perform division with potentially negative operands,
 you do not want to use @code{/} or @code{/mod} with its undefined  you do not want to use @code{/} or @code{/mod} with its undefined
 behaviour, but rather @code{fm/mod} or @code{sm/mod} (probably the  behaviour, but rather @code{fm/mod} or @code{sm/mod} (probably the
 former).  former, @pxref{Mixed precision}).
   
   @menu
   * Single precision::            
   * Bitwise operations::          
   * Mixed precision::             operations with single and double-cell integers
   * Double precision::            Double-cell integer arithmetic
   * Floating Point::              
   @end menu
   
   @node Single precision, Bitwise operations, Arithmetic, Arithmetic
 @subsection Single precision  @subsection Single precision
 doc-+  doc-+
 doc--  doc--
Line 377  doc-abs Line 795  doc-abs
 doc-min  doc-min
 doc-max  doc-max
   
   @node Bitwise operations, Mixed precision, Single precision, Arithmetic
 @subsection Bitwise operations  @subsection Bitwise operations
 doc-and  doc-and
 doc-or  doc-or
Line 385  doc-invert Line 804  doc-invert
 doc-2*  doc-2*
 doc-2/  doc-2/
   
   @node Mixed precision, Double precision, Bitwise operations, Arithmetic
 @subsection Mixed precision  @subsection Mixed precision
 doc-m+  doc-m+
 doc-*/  doc-*/
Line 396  doc-um/mod Line 816  doc-um/mod
 doc-fm/mod  doc-fm/mod
 doc-sm/rem  doc-sm/rem
   
   @node Double precision, Floating Point, Mixed precision, Arithmetic
 @subsection Double precision  @subsection Double precision
   
   The outer (aka text) interpreter converts numbers containing a dot into
   a double precision number. Note that only numbers with the dot as last
   character are standard-conforming.
   
 doc-d+  doc-d+
 doc-d-  doc-d-
 doc-dnegate  doc-dnegate
Line 404  doc-dabs Line 830  doc-dabs
 doc-dmin  doc-dmin
 doc-dmax  doc-dmax
   
 @node Stack Manipulation,,,  @node Floating Point,  , Double precision, Arithmetic
   @subsection Floating Point
   
   The format of floating point numbers recognized by the outer (aka text)
   interpreter is: a signed decimal number, possibly containing a decimal
   point (@code{.}), followed by @code{E} or @code{e}, optionally followed
   by a signed integer (the exponent). E.g., @code{1e} ist the same as
   @code{+1.0e+1}. Note that a number without @code{e}
   is not interpreted as floating-point number, but as double (if the
   number contains a @code{.}) or single precision integer. Also,
   conversions between string and floating point numbers always use base
   10, irrespective of the value of @code{BASE}. If @code{BASE} contains a
   value greater then 14, the @code{E} may be interpreted as digit and the
   number will be interpreted as integer, unless it has a signed exponent
   (both @code{+} and @code{-} are allowed as signs).
   
   Angles in floating point operations are given in radians (a full circle
   has 2 pi radians). Note, that Gforth has a separate floating point
   stack, but we use the unified notation.
   
   Floating point numbers have a number of unpleasant surprises for the
   unwary (e.g., floating point addition is not associative) and even a few
   for the wary. You should not use them unless you know what you are doing
   or you don't care that the results you get are totally bogus. If you
   want to learn about the problems of floating point numbers (and how to
   avoid them), you might start with @cite{David Goldberg, What Every
   Computer Scientist Should Know About Floating-Point Arithmetic, ACM
   Computing Surveys 23(1):5@minus{}48, March 1991}.
   
   doc-f+
   doc-f-
   doc-f*
   doc-f/
   doc-fnegate
   doc-fabs
   doc-fmax
   doc-fmin
   doc-floor
   doc-fround
   doc-f**
   doc-fsqrt
   doc-fexp
   doc-fexpm1
   doc-fln
   doc-flnp1
   doc-flog
   doc-falog
   doc-fsin
   doc-fcos
   doc-fsincos
   doc-ftan
   doc-fasin
   doc-facos
   doc-fatan
   doc-fatan2
   doc-fsinh
   doc-fcosh
   doc-ftanh
   doc-fasinh
   doc-facosh
   doc-fatanh
   
   @node Stack Manipulation, Memory access, Arithmetic, Words
 @section Stack Manipulation  @section Stack Manipulation
   
 gforth has a data stack (aka parameter stack) for characters, cells,  Gforth has a data stack (aka parameter stack) for characters, cells,
 addresses, and double cells, a floating point stack for floating point  addresses, and double cells, a floating point stack for floating point
 numbers, a return stack for storing the return addresses of colon  numbers, a return stack for storing the return addresses of colon
 definitions and other data, and a locals stack for storing local  definitions and other data, and a locals stack for storing local
Line 417  theoretically keep floating point number Line 905  theoretically keep floating point number
 additional difficulty, you don't know how many cells a floating point  additional difficulty, you don't know how many cells a floating point
 number takes. It is reportedly possible to write words in a way that  number takes. It is reportedly possible to write words in a way that
 they work also for a unified stack model, but we do not recommend trying  they work also for a unified stack model, but we do not recommend trying
 it. Also, a Forth system is allowed to keep the local variables on the  it. Instead, just say that your program has an environmental dependency
   on a separate FP stack.
   
   Also, a Forth system is allowed to keep the local variables on the
 return stack. This is reasonable, as local variables usually eliminate  return stack. This is reasonable, as local variables usually eliminate
 the need to use the return stack explicitly. So, if you want to produce  the need to use the return stack explicitly. So, if you want to produce
 a standard complying program and if you are using local variables in a  a standard complying program and if you are using local variables in a
 word, forget about return stack manipulations in that word (see the  word, forget about return stack manipulations in that word (see the
 standard document for the exact rules).  standard document for the exact rules).
   
   @menu
   * Data stack::                  
   * Floating point stack::        
   * Return stack::                
   * Locals stack::                
   * Stack pointer manipulation::  
   @end menu
   
   @node Data stack, Floating point stack, Stack Manipulation, Stack Manipulation
 @subsection Data stack  @subsection Data stack
 doc-drop  doc-drop
 doc-nip  doc-nip
Line 444  doc-2tuck Line 944  doc-2tuck
 doc-2swap  doc-2swap
 doc-2rot  doc-2rot
   
   @node Floating point stack, Return stack, Data stack, Stack Manipulation
 @subsection Floating point stack  @subsection Floating point stack
 doc-fdrop  doc-fdrop
 doc-fnip  doc-fnip
Line 453  doc-ftuck Line 954  doc-ftuck
 doc-fswap  doc-fswap
 doc-frot  doc-frot
   
   @node Return stack, Locals stack, Floating point stack, Stack Manipulation
 @subsection Return stack  @subsection Return stack
 doc->r  doc->r
 doc-r>  doc-r>
Line 463  doc-2r> Line 965  doc-2r>
 doc-2r@  doc-2r@
 doc-2rdrop  doc-2rdrop
   
   @node Locals stack, Stack pointer manipulation, Return stack, Stack Manipulation
 @subsection Locals stack  @subsection Locals stack
   
   @node Stack pointer manipulation,  , Locals stack, Stack Manipulation
 @subsection Stack pointer manipulation  @subsection Stack pointer manipulation
 doc-sp@  doc-sp@
 doc-sp!  doc-sp!
Line 475  doc-rp! Line 979  doc-rp!
 doc-lp@  doc-lp@
 doc-lp!  doc-lp!
   
 @node Memory access  @node Memory access, Control Structures, Stack Manipulation, Words
 @section Memory access  @section Memory access
   
   @menu
   * Stack-Memory transfers::      
   * Address arithmetic::          
   * Memory block access::         
   @end menu
   
   @node Stack-Memory transfers, Address arithmetic, Memory access, Memory access
 @subsection Stack-Memory transfers  @subsection Stack-Memory transfers
   
 doc-@  doc-@
Line 494  doc-sf! Line 1005  doc-sf!
 doc-df@  doc-df@
 doc-df!  doc-df!
   
   @node Address arithmetic, Memory block access, Stack-Memory transfers, Memory access
 @subsection Address arithmetic  @subsection Address arithmetic
   
 ANS Forth does not specify the sizes of the data types. Instead, it  ANS Forth does not specify the sizes of the data types. Instead, it
Line 509  must only occur at specific addresses; e Line 1021  must only occur at specific addresses; e
 accessed at addresses divisible by 4. Even if a machine allows unaligned  accessed at addresses divisible by 4. Even if a machine allows unaligned
 accesses, it can usually perform aligned accesses faster.   accesses, it can usually perform aligned accesses faster. 
   
 For the performance-concious: alignment operations are usually only  For the performance-conscious: alignment operations are usually only
 necessary during the definition of a data structure, not during the  necessary during the definition of a data structure, not during the
 (more frequent) accesses to it.  (more frequent) accesses to it.
   
Line 519  char-aligned have no use in the standard Line 1031  char-aligned have no use in the standard
 created.  created.
   
 The standard guarantees that addresses returned by @code{CREATE}d words  The standard guarantees that addresses returned by @code{CREATE}d words
 are cell-aligned; in addition, gforth guarantees that these addresses  are cell-aligned; in addition, Gforth guarantees that these addresses
 are aligned for all purposes.  are aligned for all purposes.
   
   Note that the standard defines a word @code{char}, which has nothing to
   do with address arithmetic.
   
 doc-chars  doc-chars
 doc-char+  doc-char+
 doc-cells  doc-cells
Line 540  doc-dfloats Line 1055  doc-dfloats
 doc-dfloat+  doc-dfloat+
 doc-dfalign  doc-dfalign
 doc-dfaligned  doc-dfaligned
   doc-maxalign
   doc-maxaligned
   doc-cfalign
   doc-cfaligned
 doc-address-unit-bits  doc-address-unit-bits
   
   @node Memory block access,  , Address arithmetic, Memory access
 @subsection Memory block access  @subsection Memory block access
   
 doc-move  doc-move
Line 555  doc-cmove> Line 1075  doc-cmove>
 doc-fill  doc-fill
 doc-blank  doc-blank
   
 @node Control Structures  @node Control Structures, Locals, Memory access, Words
 @section Control Structures  @section Control Structures
   
 Control structures in Forth cannot be used in interpret state, only in  Control structures in Forth cannot be used in interpret state, only in
Line 563  compile state, i.e., in a colon definiti Line 1083  compile state, i.e., in a colon definiti
 limitation, but have not seen a satisfying way around it yet, although  limitation, but have not seen a satisfying way around it yet, although
 many schemes have been proposed.  many schemes have been proposed.
   
   @menu
   * Selection::                   
   * Simple Loops::                
   * Counted Loops::               
   * Arbitrary control structures::  
   * Calls and returns::           
   * Exception Handling::          
   @end menu
   
   @node Selection, Simple Loops, Control Structures, Control Structures
 @subsection Selection  @subsection Selection
   
 @example  @example
Line 581  ELSE Line 1111  ELSE
 ENDIF  ENDIF
 @end example  @end example
   
 You can use @code{THEN} instead of {ENDIF}. Indeed, @code{THEN} is  You can use @code{THEN} instead of @code{ENDIF}. Indeed, @code{THEN} is
 standard, and @code{ENDIF} is not, although it is quite popular. We  standard, and @code{ENDIF} is not, although it is quite popular. We
 recommend using @code{ENDIF}, because it is less confusing for people  recommend using @code{ENDIF}, because it is less confusing for people
 who also know other languages (and is not prone to reinforcing negative  who also know other languages (and is not prone to reinforcing negative
Line 608  can avoid using @code{?dup}. Line 1138  can avoid using @code{?dup}.
 CASE  CASE
   @var{n1} OF @var{code1} ENDOF    @var{n1} OF @var{code1} ENDOF
   @var{n2} OF @var{code2} ENDOF    @var{n2} OF @var{code2} ENDOF
   @dots    @dots{}
 ENDCASE  ENDCASE
 @end example  @end example
   
Line 617  Executes the first @var{codei}, where th Line 1147  Executes the first @var{codei}, where th
 the last @code{ENDOF}. It may use @var{n}, which is on top of the stack,  the last @code{ENDOF}. It may use @var{n}, which is on top of the stack,
 but must not consume it.  but must not consume it.
   
   @node Simple Loops, Counted Loops, Selection, Control Structures
 @subsection Simple Loops  @subsection Simple Loops
   
 @example  @example
Line 648  AGAIN Line 1179  AGAIN
   
 This is an endless loop.  This is an endless loop.
   
   @node Counted Loops, Arbitrary control structures, Simple Loops, Control Structures
 @subsection Counted Loops  @subsection Counted Loops
   
 The basic counted loop is:  The basic counted loop is:
Line 685  There are several variations on the coun Line 1217  There are several variations on the coun
   
 @code{LEAVE} leaves the innermost counted loop immediately.  @code{LEAVE} leaves the innermost counted loop immediately.
   
   If @var{start} is greater than @var{limit}, a @code{?DO} loop is entered
   (and @code{LOOP} iterates until they become equal by wrap-around
   arithmetic). This behaviour is usually not what you want. Therefore,
   Gforth offers @code{+DO} and @code{U+DO} (as replacements for
   @code{?DO}), which do not enter the loop if @var{start} is greater than
   @var{limit}; @code{+DO} is for signed loop parameters, @code{U+DO} for
   unsigned loop parameters. These words can be implemented easily on
   standard systems, so using them does not make your programs hard to
   port; e.g.:
   @example
   : +DO ( compile-time: -- do-sys; run-time: n1 n2 -- )
       POSTPONE over POSTPONE min POSTPONE ?DO ; immediate
   @end example
   
 @code{LOOP} can be replaced with @code{@var{n} +LOOP}; this updates the  @code{LOOP} can be replaced with @code{@var{n} +LOOP}; this updates the
 index by @var{n} instead of by 1. The loop is terminated when the border  index by @var{n} instead of by 1. The loop is terminated when the border
 between @var{limit-1} and @var{limit} is crossed. E.g.:  between @var{limit-1} and @var{limit} is crossed. E.g.:
   
 4 0 ?DO  i .  2 +LOOP   prints 0 2  @code{4 0 +DO  i .  2 +LOOP}   prints @code{0 2}
   
 4 1 ?DO  i .  2 +LOOP   prints 1 3  @code{4 1 +DO  i .  2 +LOOP}   prints @code{1 3}
   
 The behaviour of @code{@var{n} +LOOP} is peculiar when @var{n} is negative:  The behaviour of @code{@var{n} +LOOP} is peculiar when @var{n} is negative:
   
 -1 0 ?DO  i .  -1 +LOOP  prints 0 -1  @code{-1 0 ?DO  i .  -1 +LOOP}  prints @code{0 -1}
   
  0 0 ?DO  i .  -1 +LOOP  prints nothing  @code{ 0 0 ?DO  i .  -1 +LOOP}  prints nothing
   
 Therefore we recommend avoiding using @code{@var{n} +LOOP} with negative  Therefore we recommend avoiding @code{@var{n} +LOOP} with negative
 @var{n}. One alternative is @code{@var{n} S+LOOP}, where the negative  @var{n}. One alternative is @code{@var{u} -LOOP}, which reduces the
 case behaves symmetrical to the positive case:  index by @var{u} each iteration. The loop is terminated when the border
   between @var{limit+1} and @var{limit} is crossed. Gforth also provides
   @code{-DO} and @code{U-DO} for down-counting loops. E.g.:
   
 -2 0 ?DO  i .  -1 +LOOP  prints 0 -1  @code{-2 0 -DO  i .  1 -LOOP}  prints @code{0 -1}
   
 -1 0 ?DO  i .  -1 +LOOP  prints 0  @code{-1 0 -DO  i .  1 -LOOP}  prints @code{0}
   
  0 0 ?DO  i .  -1 +LOOP  prints nothing  @code{ 0 0 -DO  i .  1 -LOOP}  prints nothing
   
   Another alternative is @code{@var{n} S+LOOP}, where the negative
   case behaves symmetrical to the positive case:
   
 The loop is terminated when the border between @var{limit-sgn(n)} and  @code{-2 0 -DO  i .  -1 S+LOOP}  prints @code{0 -1}
 @var{limit} is crossed. However, @code{S+LOOP} is not part of the ANS  
 Forth standard.  
   
 @code{?DO} can be replaced by @code{DO}. @code{DO} enters the loop even  The loop is terminated when the border between @var{limit@minus{}sgn(n)}
 when the start and the limit value are equal. We do not recommend using  and @var{limit} is crossed. Unfortunately, neither @code{-LOOP} nor
 @code{DO}. It will just give you maintenance troubles.  @code{S+LOOP} are part of the ANS Forth standard, and they are not easy
   to implement using standard words. If you want to write standard
   programs, just avoid counting down.
   
   @code{?DO} can also be replaced by @code{DO}. @code{DO} always enters
   the loop, independent of the loop parameters. Do not use @code{DO}, even
   if you know that the loop is entered in any case. Such knowledge tends
   to become invalid during maintenance of a program, and then the
   @code{DO} will make trouble.
   
 @code{UNLOOP} is used to prepare for an abnormal loop exit, e.g., via  @code{UNLOOP} is used to prepare for an abnormal loop exit, e.g., via
 @code{EXIT}. @code{UNLOOP} removes the loop control parameters from the  @code{EXIT}. @code{UNLOOP} removes the loop control parameters from the
Line 729  FOR Line 1286  FOR
 NEXT  NEXT
 @end example  @end example
 This is the preferred loop of native code compiler writers who are too  This is the preferred loop of native code compiler writers who are too
 lazy to optimize @code{?DO} loops properly. In GNU Forth, this loop  lazy to optimize @code{?DO} loops properly. In Gforth, this loop
 iterates @var{n+1} times; @code{i} produces values starting with @var{n}  iterates @var{n+1} times; @code{i} produces values starting with @var{n}
 and ending with 0. Other Forth systems may behave differently, even if  and ending with 0. Other Forth systems may behave differently, even if
 they support @code{FOR} loops.  they support @code{FOR} loops.
   
 @node Locals  @node Arbitrary control structures, Calls and returns, Counted Loops, Control Structures
   @subsection Arbitrary control structures
   
   ANS Forth permits and supports using control structures in a non-nested
   way. Information about incomplete control structures is stored on the
   control-flow stack. This stack may be implemented on the Forth data
   stack, and this is what we have done in Gforth.
   
   An @i{orig} entry represents an unresolved forward branch, a @i{dest}
   entry represents a backward branch target. A few words are the basis for
   building any control structure possible (except control structures that
   need storage, like calls, coroutines, and backtracking).
   
   doc-if
   doc-ahead
   doc-then
   doc-begin
   doc-until
   doc-again
   doc-cs-pick
   doc-cs-roll
   
   On many systems control-flow stack items take one word, in Gforth they
   currently take three (this may change in the future). Therefore it is a
   really good idea to manipulate the control flow stack with
   @code{cs-pick} and @code{cs-roll}, not with data stack manipulation
   words.
   
   Some standard control structure words are built from these words:
   
   doc-else
   doc-while
   doc-repeat
   
   Counted loop words constitute a separate group of words:
   
   doc-?do
   doc-+do
   doc-u+do
   doc--do
   doc-u-do
   doc-do
   doc-for
   doc-loop
   doc-s+loop
   doc-+loop
   doc--loop
   doc-next
   doc-leave
   doc-?leave
   doc-unloop
   doc-done
   
   The standard does not allow using @code{cs-pick} and @code{cs-roll} on
   @i{do-sys}. Our system allows it, but it's your job to ensure that for
   every @code{?DO} etc. there is exactly one @code{UNLOOP} on any path
   through the definition (@code{LOOP} etc. compile an @code{UNLOOP} on the
   fall-through path). Also, you have to ensure that all @code{LEAVE}s are
   resolved (by using one of the loop-ending words or @code{DONE}).
   
   Another group of control structure words are
   
   doc-case
   doc-endcase
   doc-of
   doc-endof
   
   @i{case-sys} and @i{of-sys} cannot be processed using @code{cs-pick} and
   @code{cs-roll}.
   
   @subsubsection Programming Style
   
   In order to ensure readability we recommend that you do not create
   arbitrary control structures directly, but define new control structure
   words for the control structure you want and use these words in your
   program.
   
   E.g., instead of writing
   
   @example
   begin
     ...
   if [ 1 cs-roll ]
     ...
   again then
   @end example
   
   we recommend defining control structure words, e.g.,
   
   @example
   : while ( dest -- orig dest )
    POSTPONE if
    1 cs-roll ; immediate
   
   : repeat ( orig dest -- )
    POSTPONE again
    POSTPONE then ; immediate
   @end example
   
   and then using these to create the control structure:
   
   @example
   begin
     ...
   while
     ...
   repeat
   @end example
   
   That's much easier to read, isn't it? Of course, @code{BEGIN} and
   @code{WHILE} are predefined, so in this example it would not be
   necessary to define them.
   
   @node Calls and returns, Exception Handling, Arbitrary control structures, Control Structures
   @subsection Calls and returns
   
   A definition can be called simply be writing the name of the
   definition. When the end of the definition is reached, it returns. An
   earlier return can be forced using
   
   doc-exit
   
   Don't forget to clean up the return stack and @code{UNLOOP} any
   outstanding @code{?DO}...@code{LOOP}s before @code{EXIT}ing. The
   primitive compiled by @code{EXIT} is
   
   doc-;s
   
   @node Exception Handling,  , Calls and returns, Control Structures
   @subsection Exception Handling
   
   doc-catch
   doc-throw
   
   @node Locals, Defining Words, Control Structures, Words
 @section Locals  @section Locals
   
   Local variables can make Forth programming more enjoyable and Forth
   programs easier to read. Unfortunately, the locals of ANS Forth are
   laden with restrictions. Therefore, we provide not only the ANS Forth
   locals wordset, but also our own, more powerful locals wordset (we
   implemented the ANS Forth locals wordset through our locals wordset).
   
   @menu
   * Gforth locals::               
   * ANS Forth locals::            
   @end menu
   
   @node Gforth locals, ANS Forth locals, Locals, Locals
   @subsection Gforth locals
   
   Locals can be defined with
   
   @example
   @{ local1 local2 ... -- comment @}
   @end example
   or
   @example
   @{ local1 local2 ... @}
   @end example
   
   E.g.,
   @example
   : max @{ n1 n2 -- n3 @}
    n1 n2 > if
      n1
    else
      n2
    endif ;
   @end example
   
   The similarity of locals definitions with stack comments is intended. A
   locals definition often replaces the stack comment of a word. The order
   of the locals corresponds to the order in a stack comment and everything
   after the @code{--} is really a comment.
   
   This similarity has one disadvantage: It is too easy to confuse locals
   declarations with stack comments, causing bugs and making them hard to
   find. However, this problem can be avoided by appropriate coding
   conventions: Do not use both notations in the same program. If you do,
   they should be distinguished using additional means, e.g. by position.
   
   The name of the local may be preceded by a type specifier, e.g.,
   @code{F:} for a floating point value:
   
   @example
   : CX* @{ F: Ar F: Ai F: Br F: Bi -- Cr Ci @}
   \ complex multiplication
    Ar Br f* Ai Bi f* f-
    Ar Bi f* Ai Br f* f+ ;
   @end example
   
   Gforth currently supports cells (@code{W:}, @code{W^}), doubles
   (@code{D:}, @code{D^}), floats (@code{F:}, @code{F^}) and characters
   (@code{C:}, @code{C^}) in two flavours: a value-flavoured local (defined
   with @code{W:}, @code{D:} etc.) produces its value and can be changed
   with @code{TO}. A variable-flavoured local (defined with @code{W^} etc.)
   produces its address (which becomes invalid when the variable's scope is
   left). E.g., the standard word @code{emit} can be defined in therms of
   @code{type} like this:
   
   @example
   : emit @{ C^ char* -- @}
       char* 1 type ;
   @end example
   
   A local without type specifier is a @code{W:} local. Both flavours of
   locals are initialized with values from the data or FP stack.
   
   Currently there is no way to define locals with user-defined data
   structures, but we are working on it.
   
   Gforth allows defining locals everywhere in a colon definition. This
   poses the following questions:
   
   @menu
   * Where are locals visible by name?::  
   * How long do locals live?::    
   * Programming Style::           
   * Implementation::              
   @end menu
   
   @node Where are locals visible by name?, How long do locals live?, Gforth locals, Gforth locals
   @subsubsection Where are locals visible by name?
   
   Basically, the answer is that locals are visible where you would expect
   it in block-structured languages, and sometimes a little longer. If you
   want to restrict the scope of a local, enclose its definition in
   @code{SCOPE}...@code{ENDSCOPE}.
   
   doc-scope
   doc-endscope
   
   These words behave like control structure words, so you can use them
   with @code{CS-PICK} and @code{CS-ROLL} to restrict the scope in
   arbitrary ways.
   
   If you want a more exact answer to the visibility question, here's the
   basic principle: A local is visible in all places that can only be
   reached through the definition of the local@footnote{In compiler
   construction terminology, all places dominated by the definition of the
   local.}. In other words, it is not visible in places that can be reached
   without going through the definition of the local. E.g., locals defined
   in @code{IF}...@code{ENDIF} are visible until the @code{ENDIF}, locals
   defined in @code{BEGIN}...@code{UNTIL} are visible after the
   @code{UNTIL} (until, e.g., a subsequent @code{ENDSCOPE}).
   
   The reasoning behind this solution is: We want to have the locals
   visible as long as it is meaningful. The user can always make the
   visibility shorter by using explicit scoping. In a place that can
   only be reached through the definition of a local, the meaning of a
   local name is clear. In other places it is not: How is the local
   initialized at the control flow path that does not contain the
   definition? Which local is meant, if the same name is defined twice in
   two independent control flow paths?
   
   This should be enough detail for nearly all users, so you can skip the
   rest of this section. If you relly must know all the gory details and
   options, read on.
   
   In order to implement this rule, the compiler has to know which places
   are unreachable. It knows this automatically after @code{AHEAD},
   @code{AGAIN}, @code{EXIT} and @code{LEAVE}; in other cases (e.g., after
   most @code{THROW}s), you can use the word @code{UNREACHABLE} to tell the
   compiler that the control flow never reaches that place. If
   @code{UNREACHABLE} is not used where it could, the only consequence is
   that the visibility of some locals is more limited than the rule above
   says. If @code{UNREACHABLE} is used where it should not (i.e., if you
   lie to the compiler), buggy code will be produced.
   
   Another problem with this rule is that at @code{BEGIN}, the compiler
   does not know which locals will be visible on the incoming
   back-edge. All problems discussed in the following are due to this
   ignorance of the compiler (we discuss the problems using @code{BEGIN}
   loops as examples; the discussion also applies to @code{?DO} and other
   loops). Perhaps the most insidious example is:
   @example
   AHEAD
   BEGIN
     x
   [ 1 CS-ROLL ] THEN
     @{ x @}
     ...
   UNTIL
   @end example
   
   This should be legal according to the visibility rule. The use of
   @code{x} can only be reached through the definition; but that appears
   textually below the use.
   
   From this example it is clear that the visibility rules cannot be fully
   implemented without major headaches. Our implementation treats common
   cases as advertised and the exceptions are treated in a safe way: The
   compiler makes a reasonable guess about the locals visible after a
   @code{BEGIN}; if it is too pessimistic, the
   user will get a spurious error about the local not being defined; if the
   compiler is too optimistic, it will notice this later and issue a
   warning. In the case above the compiler would complain about @code{x}
   being undefined at its use. You can see from the obscure examples in
   this section that it takes quite unusual control structures to get the
   compiler into trouble, and even then it will often do fine.
   
   If the @code{BEGIN} is reachable from above, the most optimistic guess
   is that all locals visible before the @code{BEGIN} will also be
   visible after the @code{BEGIN}. This guess is valid for all loops that
   are entered only through the @code{BEGIN}, in particular, for normal
   @code{BEGIN}...@code{WHILE}...@code{REPEAT} and
   @code{BEGIN}...@code{UNTIL} loops and it is implemented in our
   compiler. When the branch to the @code{BEGIN} is finally generated by
   @code{AGAIN} or @code{UNTIL}, the compiler checks the guess and
   warns the user if it was too optimisitic:
   @example
   IF
     @{ x @}
   BEGIN
     \ x ? 
   [ 1 cs-roll ] THEN
     ...
   UNTIL
   @end example
   
   Here, @code{x} lives only until the @code{BEGIN}, but the compiler
   optimistically assumes that it lives until the @code{THEN}. It notices
   this difference when it compiles the @code{UNTIL} and issues a
   warning. The user can avoid the warning, and make sure that @code{x}
   is not used in the wrong area by using explicit scoping:
   @example
   IF
     SCOPE
     @{ x @}
     ENDSCOPE
   BEGIN
   [ 1 cs-roll ] THEN
     ...
   UNTIL
   @end example
   
   Since the guess is optimistic, there will be no spurious error messages
   about undefined locals.
   
   If the @code{BEGIN} is not reachable from above (e.g., after
   @code{AHEAD} or @code{EXIT}), the compiler cannot even make an
   optimistic guess, as the locals visible after the @code{BEGIN} may be
   defined later. Therefore, the compiler assumes that no locals are
   visible after the @code{BEGIN}. However, the user can use
   @code{ASSUME-LIVE} to make the compiler assume that the same locals are
   visible at the BEGIN as at the point where the top control-flow stack
   item was created.
   
   doc-assume-live
   
   E.g.,
   @example
   @{ x @}
   AHEAD
   ASSUME-LIVE
   BEGIN
     x
   [ 1 CS-ROLL ] THEN
     ...
   UNTIL
   @end example
   
   Other cases where the locals are defined before the @code{BEGIN} can be
   handled by inserting an appropriate @code{CS-ROLL} before the
   @code{ASSUME-LIVE} (and changing the control-flow stack manipulation
   behind the @code{ASSUME-LIVE}).
   
   Cases where locals are defined after the @code{BEGIN} (but should be
   visible immediately after the @code{BEGIN}) can only be handled by
   rearranging the loop. E.g., the ``most insidious'' example above can be
   arranged into:
   @example
   BEGIN
     @{ x @}
     ... 0=
   WHILE
     x
   REPEAT
   @end example
   
   @node How long do locals live?, Programming Style, Where are locals visible by name?, Gforth locals
   @subsubsection How long do locals live?
   
   The right answer for the lifetime question would be: A local lives at
   least as long as it can be accessed. For a value-flavoured local this
   means: until the end of its visibility. However, a variable-flavoured
   local could be accessed through its address far beyond its visibility
   scope. Ultimately, this would mean that such locals would have to be
   garbage collected. Since this entails un-Forth-like implementation
   complexities, I adopted the same cowardly solution as some other
   languages (e.g., C): The local lives only as long as it is visible;
   afterwards its address is invalid (and programs that access it
   afterwards are erroneous).
   
   @node Programming Style, Implementation, How long do locals live?, Gforth locals
   @subsubsection Programming Style
   
   The freedom to define locals anywhere has the potential to change
   programming styles dramatically. In particular, the need to use the
   return stack for intermediate storage vanishes. Moreover, all stack
   manipulations (except @code{PICK}s and @code{ROLL}s with run-time
   determined arguments) can be eliminated: If the stack items are in the
   wrong order, just write a locals definition for all of them; then
   write the items in the order you want.
   
   This seems a little far-fetched and eliminating stack manipulations is
   unlikely to become a conscious programming objective. Still, the number
   of stack manipulations will be reduced dramatically if local variables
   are used liberally (e.g., compare @code{max} in @ref{Gforth locals} with
   a traditional implementation of @code{max}).
   
   This shows one potential benefit of locals: making Forth programs more
   readable. Of course, this benefit will only be realized if the
   programmers continue to honour the principle of factoring instead of
   using the added latitude to make the words longer.
   
   Using @code{TO} can and should be avoided.  Without @code{TO},
   every value-flavoured local has only a single assignment and many
   advantages of functional languages apply to Forth. I.e., programs are
   easier to analyse, to optimize and to read: It is clear from the
   definition what the local stands for, it does not turn into something
   different later.
   
   E.g., a definition using @code{TO} might look like this:
   @example
   : strcmp @{ addr1 u1 addr2 u2 -- n @}
    u1 u2 min 0
    ?do
      addr1 c@ addr2 c@ - ?dup
      if
        unloop exit
      then
      addr1 char+ TO addr1
      addr2 char+ TO addr2
    loop
    u1 u2 - ;
   @end example
   Here, @code{TO} is used to update @code{addr1} and @code{addr2} at
   every loop iteration. @code{strcmp} is a typical example of the
   readability problems of using @code{TO}. When you start reading
   @code{strcmp}, you think that @code{addr1} refers to the start of the
   string. Only near the end of the loop you realize that it is something
   else.
   
   This can be avoided by defining two locals at the start of the loop that
   are initialized with the right value for the current iteration.
   @example
   : strcmp @{ addr1 u1 addr2 u2 -- n @}
    addr1 addr2
    u1 u2 min 0 
    ?do @{ s1 s2 @}
      s1 c@ s2 c@ - ?dup 
      if
        unloop exit
      then
      s1 char+ s2 char+
    loop
    2drop
    u1 u2 - ;
   @end example
   Here it is clear from the start that @code{s1} has a different value
   in every loop iteration.
   
   @node Implementation,  , Programming Style, Gforth locals
   @subsubsection Implementation
   
   Gforth uses an extra locals stack. The most compelling reason for
   this is that the return stack is not float-aligned; using an extra stack
   also eliminates the problems and restrictions of using the return stack
   as locals stack. Like the other stacks, the locals stack grows toward
   lower addresses. A few primitives allow an efficient implementation:
   
   doc-@local#
   doc-f@local#
   doc-laddr#
   doc-lp+!#
   doc-lp!
   doc->l
   doc-f>l
   
   In addition to these primitives, some specializations of these
   primitives for commonly occurring inline arguments are provided for
   efficiency reasons, e.g., @code{@@local0} as specialization of
   @code{@@local#} for the inline argument 0. The following compiling words
   compile the right specialized version, or the general version, as
   appropriate:
   
   doc-compile-@local
   doc-compile-f@local
   doc-compile-lp+!
   
   Combinations of conditional branches and @code{lp+!#} like
   @code{?branch-lp+!#} (the locals pointer is only changed if the branch
   is taken) are provided for efficiency and correctness in loops.
   
   A special area in the dictionary space is reserved for keeping the
   local variable names. @code{@{} switches the dictionary pointer to this
   area and @code{@}} switches it back and generates the locals
   initializing code. @code{W:} etc.@ are normal defining words. This
   special area is cleared at the start of every colon definition.
   
   A special feature of Gforth's dictionary is used to implement the
   definition of locals without type specifiers: every wordlist (aka
   vocabulary) has its own methods for searching
   etc. (@pxref{Wordlists}). For the present purpose we defined a wordlist
   with a special search method: When it is searched for a word, it
   actually creates that word using @code{W:}. @code{@{} changes the search
   order to first search the wordlist containing @code{@}}, @code{W:} etc.,
   and then the wordlist for defining locals without type specifiers.
   
   The lifetime rules support a stack discipline within a colon
   definition: The lifetime of a local is either nested with other locals
   lifetimes or it does not overlap them.
   
   At @code{BEGIN}, @code{IF}, and @code{AHEAD} no code for locals stack
   pointer manipulation is generated. Between control structure words
   locals definitions can push locals onto the locals stack. @code{AGAIN}
   is the simplest of the other three control flow words. It has to
   restore the locals stack depth of the corresponding @code{BEGIN}
   before branching. The code looks like this:
   @format
   @code{lp+!#} current-locals-size @minus{} dest-locals-size
   @code{branch} <begin>
   @end format
   
   @code{UNTIL} is a little more complicated: If it branches back, it
   must adjust the stack just like @code{AGAIN}. But if it falls through,
   the locals stack must not be changed. The compiler generates the
   following code:
   @format
   @code{?branch-lp+!#} <begin> current-locals-size @minus{} dest-locals-size
   @end format
   The locals stack pointer is only adjusted if the branch is taken.
   
   @code{THEN} can produce somewhat inefficient code:
   @format
   @code{lp+!#} current-locals-size @minus{} orig-locals-size
   <orig target>:
   @code{lp+!#} orig-locals-size @minus{} new-locals-size
   @end format
   The second @code{lp+!#} adjusts the locals stack pointer from the
   level at the @var{orig} point to the level after the @code{THEN}. The
   first @code{lp+!#} adjusts the locals stack pointer from the current
   level to the level at the orig point, so the complete effect is an
   adjustment from the current level to the right level after the
   @code{THEN}.
   
   In a conventional Forth implementation a dest control-flow stack entry
   is just the target address and an orig entry is just the address to be
   patched. Our locals implementation adds a wordlist to every orig or dest
   item. It is the list of locals visible (or assumed visible) at the point
   described by the entry. Our implementation also adds a tag to identify
   the kind of entry, in particular to differentiate between live and dead
   (reachable and unreachable) orig entries.
   
   A few unusual operations have to be performed on locals wordlists:
   
   doc-common-list
   doc-sub-list?
   doc-list-size
   
   Several features of our locals wordlist implementation make these
   operations easy to implement: The locals wordlists are organised as
   linked lists; the tails of these lists are shared, if the lists
   contain some of the same locals; and the address of a name is greater
   than the address of the names behind it in the list.
   
   Another important implementation detail is the variable
   @code{dead-code}. It is used by @code{BEGIN} and @code{THEN} to
   determine if they can be reached directly or only through the branch
   that they resolve. @code{dead-code} is set by @code{UNREACHABLE},
   @code{AHEAD}, @code{EXIT} etc., and cleared at the start of a colon
   definition, by @code{BEGIN} and usually by @code{THEN}.
   
   Counted loops are similar to other loops in most respects, but
   @code{LEAVE} requires special attention: It performs basically the same
   service as @code{AHEAD}, but it does not create a control-flow stack
   entry. Therefore the information has to be stored elsewhere;
   traditionally, the information was stored in the target fields of the
   branches created by the @code{LEAVE}s, by organizing these fields into a
   linked list. Unfortunately, this clever trick does not provide enough
   space for storing our extended control flow information. Therefore, we
   introduce another stack, the leave stack. It contains the control-flow
   stack entries for all unresolved @code{LEAVE}s.
   
   Local names are kept until the end of the colon definition, even if
   they are no longer visible in any control-flow path. In a few cases
   this may lead to increased space needs for the locals name area, but
   usually less than reclaiming this space would cost in code size.
   
   
   @node ANS Forth locals,  , Gforth locals, Locals
   @subsection ANS Forth locals
   
   The ANS Forth locals wordset does not define a syntax for locals, but
   words that make it possible to define various syntaxes. One of the
   possible syntaxes is a subset of the syntax we used in the Gforth locals
   wordset, i.e.:
   
   @example
   @{ local1 local2 ... -- comment @}
   @end example
   or
   @example
   @{ local1 local2 ... @}
   @end example
   
   The order of the locals corresponds to the order in a stack comment. The
   restrictions are:
   
   @itemize @bullet
   @item
   Locals can only be cell-sized values (no type specifiers are allowed).
   @item
   Locals can be defined only outside control structures.
   @item
   Locals can interfere with explicit usage of the return stack. For the
   exact (and long) rules, see the standard. If you don't use return stack
   accessing words in a definition using locals, you will be all right. The
   purpose of this rule is to make locals implementation on the return
   stack easier.
   @item
   The whole definition must be in one line.
   @end itemize
   
   Locals defined in this way behave like @code{VALUE}s
   (@xref{Values}). I.e., they are initialized from the stack. Using their
   name produces their value. Their value can be changed using @code{TO}.
   
   Since this syntax is supported by Gforth directly, you need not do
   anything to use it. If you want to port a program using this syntax to
   another ANS Forth system, use @file{anslocal.fs} to implement the syntax
   on the other system.
   
   Note that a syntax shown in the standard, section A.13 looks
   similar, but is quite different in having the order of locals
   reversed. Beware!
   
   The ANS Forth locals wordset itself consists of the following word
   
   doc-(local)
   
   The ANS Forth locals extension wordset defines a syntax, but it is so
   awful that we strongly recommend not to use it. We have implemented this
   syntax to make porting to Gforth easy, but do not document it here. The
   problem with this syntax is that the locals are defined in an order
   reversed with respect to the standard stack comment notation, making
   programs harder to read, and easier to misread and miswrite. The only
   merit of this syntax is that it is easy to implement using the ANS Forth
   locals wordset.
   
   @node Defining Words, Wordlists, Locals, Words
   @section Defining Words
   
   @menu
   * Values::                      
   @end menu
   
   @node Values,  , Defining Words, Defining Words
   @subsection Values
   
   @node Wordlists, Files, Defining Words, Words
   @section Wordlists
   
   @node Files, Blocks, Wordlists, Words
   @section Files
   
   @node Blocks, Other I/O, Files, Words
   @section Blocks
   
   @node Other I/O, Programming Tools, Blocks, Words
   @section Other I/O
   
   @node Programming Tools, Assembler and Code words, Other I/O, Words
   @section Programming Tools
   
   @menu
   * Debugging::                   Simple and quick.
   * Assertions::                  Making your programs self-checking.
   @end menu
   
   @node Debugging, Assertions, Programming Tools, Programming Tools
   @subsection Debugging
   
   The simple debugging aids provided in @file{debugging.fs}
   are meant to support a different style of debugging than the
   tracing/stepping debuggers used in languages with long turn-around
   times.
   
   A much better (faster) way in fast-compilig languages is to add
   printing code at well-selected places, let the program run, look at
   the output, see where things went wrong, add more printing code, etc.,
   until the bug is found.
   
   The word @code{~~} is easy to insert. It just prints debugging
   information (by default the source location and the stack contents). It
   is also easy to remove (@kbd{C-x ~} in the Emacs Forth mode to
   query-replace them with nothing). The deferred words
   @code{printdebugdata} and @code{printdebugline} control the output of
   @code{~~}. The default source location output format works well with
   Emacs' compilation mode, so you can step through the program at the
   source level using @kbd{C-x `} (the advantage over a stepping debugger
   is that you can step in any direction and you know where the crash has
   happened or where the strange data has occurred).
   
   Note that the default actions clobber the contents of the pictured
   numeric output string, so you should not use @code{~~}, e.g., between
   @code{<#} and @code{#>}.
   
   doc-~~
   doc-printdebugdata
   doc-printdebugline
   
   @node Assertions,  , Debugging, Programming Tools
   @subsection Assertions
   
   It is a good idea to make your programs self-checking, in particular, if
   you use an assumption (e.g., that a certain field of a data structure is
   never zero) that may become wrong during maintenance. Gforth supports
   assertions for this purpose. They are used like this:
   
   @example
   assert( @var{flag} )
   @end example
   
   The code between @code{assert(} and @code{)} should compute a flag, that
   should be true if everything is alright and false otherwise. It should
   not change anything else on the stack. The overall stack effect of the
   assertion is @code{( -- )}. E.g.
   
   @example
   assert( 1 1 + 2 = ) \ what we learn in school
   assert( dup 0<> ) \ assert that the top of stack is not zero
   assert( false ) \ this code should not be reached
   @end example
   
   The need for assertions is different at different times. During
   debugging, we want more checking, in production we sometimes care more
   for speed. Therefore, assertions can be turned off, i.e., the assertion
   becomes a comment. Depending on the importance of an assertion and the
   time it takes to check it, you may want to turn off some assertions and
   keep others turned on. Gforth provides several levels of assertions for
   this purpose:
   
   doc-assert0(
   doc-assert1(
   doc-assert2(
   doc-assert3(
   doc-assert(
   doc-)
   
   @code{Assert(} is the same as @code{assert1(}. The variable
   @code{assert-level} specifies the highest assertions that are turned
   on. I.e., at the default @code{assert-level} of one, @code{assert0(} and
   @code{assert1(} assertions perform checking, while @code{assert2(} and
   @code{assert3(} assertions are treated as comments.
   
   Note that the @code{assert-level} is evaluated at compile-time, not at
   run-time. I.e., you cannot turn assertions on or off at run-time, you
   have to set the @code{assert-level} appropriately before compiling a
   piece of code. You can compile several pieces of code at several
   @code{assert-level}s (e.g., a trusted library at level 1 and newly
   written code at level 3).
   
   doc-assert-level
   
   If an assertion fails, a message compatible with Emacs' compilation mode
   is produced and the execution is aborted (currently with @code{ABORT"}.
   If there is interest, we will introduce a special throw code. But if you
   intend to @code{catch} a specific condition, using @code{throw} is
   probably more appropriate than an assertion).
   
   @node Assembler and Code words, Threading Words, Programming Tools, Words
   @section Assembler and Code words
   
   Gforth provides some words for defining primitives (words written in
   machine code), and for defining the the machine-code equivalent of
   @code{DOES>}-based defining words. However, the machine-independent
   nature of Gforth poses a few problems: First of all. Gforth runs on
   several architectures, so it can provide no standard assembler. What's
   worse is that the register allocation not only depends on the processor,
   but also on the gcc version and options used.
   
   The words Gforth offers encapsulate some system dependences (e.g., the
   header structure), so a system-independent assembler may be used in
   Gforth. If you do not have an assembler, you can compile machine code
   directly with @code{,} and @code{c,}.
   
   doc-assembler
   doc-code
   doc-end-code
   doc-;code
   doc-flush-icache
   
   If @code{flush-icache} does not work correctly, @code{code} words
   etc. will not work (reliably), either.
   
   These words are rarely used. Therefore they reside in @code{code.fs},
   which is usually not loaded (except @code{flush-icache}, which is always
   present). You can load them with @code{require code.fs}.
   
   Another option for implementing normal and defining words efficiently
   is: adding the wanted functionality to the source of Gforth. For normal
   words you just have to edit @file{primitives}, defining words (for fast
   defined words) probably require changes in @file{engine.c},
   @file{kernal.fs}, @file{prims2x.fs}, and possibly @file{cross.fs}.
   
   
   @node Threading Words,  , Assembler and Code words, Words
   @section Threading Words
   
   These words provide access to code addresses and other threading stuff
   in Gforth (and, possibly, other interpretive Forths). It more or less
   abstracts away the differences between direct and indirect threading
   (and, for direct threading, the machine dependences). However, at
   present this wordset is still inclomplete. It is also pretty low-level;
   some day it will hopefully be made unnecessary by an internals words set
   that abstracts implementation details away completely.
   
   doc->code-address
   doc->does-code
   doc-code-address!
   doc-does-code!
   doc-does-handler!
   doc-/does-handler
   
   The code addresses produced by various defining words are produced by
   the following words:
   
   doc-docol:
   doc-docon:
   doc-dovar:
   doc-douser:
   doc-dodefer:
   doc-dofield:
   
   Currently there is no installation-independent way for recogizing words
   defined by a @code{CREATE}...@code{DOES>} word; however, once you know
   that a word is defined by a @code{CREATE}...@code{DOES>} word, you can
   use @code{>DOES-CODE}.
   
   @node ANS conformance, Model, Words, Top
   @chapter ANS conformance
   
   To the best of our knowledge, Gforth is an
   
   ANS Forth System
   @itemize
   @item providing the Core Extensions word set
   @item providing the Block word set
   @item providing the Block Extensions word set
   @item providing the Double-Number word set
   @item providing the Double-Number Extensions word set
   @item providing the Exception word set
   @item providing the Exception Extensions word set
   @item providing the Facility word set
   @item providing @code{MS} and @code{TIME&DATE} from the Facility Extensions word set
   @item providing the File Access word set
   @item providing the File Access Extensions word set
   @item providing the Floating-Point word set
   @item providing the Floating-Point Extensions word set
   @item providing the Locals word set
   @item providing the Locals Extensions word set
   @item providing the Memory-Allocation word set
   @item providing the Memory-Allocation Extensions word set (that one's easy)
   @item providing the Programming-Tools word set
   @item providing @code{;code}, @code{AHEAD}, @code{ASSEMBLER}, @code{BYE}, @code{CODE}, @code{CS-PICK}, @code{CS-ROLL}, @code{STATE}, @code{[ELSE]}, @code{[IF]}, @code{[THEN]} from the Programming-Tools Extensions word set
   @item providing the Search-Order word set
   @item providing the Search-Order Extensions word set
   @item providing the String word set
   @item providing the String Extensions word set (another easy one)
   @end itemize
   
   In addition, ANS Forth systems are required to document certain
   implementation choices. This chapter tries to meet these
   requirements. In many cases it gives a way to ask the system for the
   information instead of providing the information directly, in
   particular, if the information depends on the processor, the operating
   system or the installation options chosen, or if they are likely to
   change during the maintenance of Gforth.
   
   @comment The framework for the rest has been taken from pfe.
   
   @menu
   * The Core Words::              
   * The optional Block word set::  
   * The optional Double Number word set::  
   * The optional Exception word set::  
   * The optional Facility word set::  
   * The optional File-Access word set::  
   * The optional Floating-Point word set::  
   * The optional Locals word set::  
   * The optional Memory-Allocation word set::  
   * The optional Programming-Tools word set::  
   * The optional Search-Order word set::  
   @end menu
   
   
   @c =====================================================================
   @node The Core Words, The optional Block word set, ANS conformance, ANS conformance
   @comment  node-name,  next,  previous,  up
   @section The Core Words
   @c =====================================================================
   
   @menu
   * core-idef::                   Implementation Defined Options                   
   * core-ambcond::                Ambiguous Conditions                
   * core-other::                  Other System Documentation                  
   @end menu
   
   @c ---------------------------------------------------------------------
   @node core-idef, core-ambcond, The Core Words, The Core Words
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item (Cell) aligned addresses:
   processor-dependent. Gforth's alignment words perform natural alignment
   (e.g., an address aligned for a datum of size 8 is divisible by
   8). Unaligned accesses usually result in a @code{-23 THROW}.
   
   @item @code{EMIT} and non-graphic characters:
   The character is output using the C library function (actually, macro)
   @code{putchar}.
   
   @item character editing of @code{ACCEPT} and @code{EXPECT}:
   This is modeled on the GNU readline library (@pxref{Readline
   Interaction, , Command Line Editing, readline, The GNU Readline
   Library}) with Emacs-like key bindings. @kbd{Tab} deviates a little by
   producing a full word completion every time you type it (instead of
   producing the common prefix of all completions).
   
   @item character set:
   The character set of your computer and display device. Gforth is
   8-bit-clean (but some other component in your system may make trouble).
   
   @item Character-aligned address requirements:
   installation-dependent. Currently a character is represented by a C
   @code{unsigned char}; in the future we might switch to @code{wchar_t}
   (Comments on that requested).
   
   @item character-set extensions and matching of names:
   Any character except the ASCII NUL charcter can be used in a
   name. Matching is case-insensitive. The matching is performed using the
   C function @code{strncasecmp}, whose function is probably influenced by
   the locale. E.g., the @code{C} locale does not know about accents and
   umlauts, so they are matched case-sensitively in that locale. For
   portability reasons it is best to write programs such that they work in
   the @code{C} locale. Then one can use libraries written by a Polish
   programmer (who might use words containing ISO Latin-2 encoded
   characters) and by a French programmer (ISO Latin-1) in the same program
   (of course, @code{WORDS} will produce funny results for some of the
   words (which ones, depends on the font you are using)). Also, the locale
   you prefer may not be available in other operating systems. Hopefully,
   Unicode will solve these problems one day.
   
   @item conditions under which control characters match a space delimiter:
   If @code{WORD} is called with the space character as a delimiter, all
   white-space characters (as identified by the C macro @code{isspace()})
   are delimiters. @code{PARSE}, on the other hand, treats space like other
   delimiters. @code{PARSE-WORD} treats space like @code{WORD}, but behaves
   like @code{PARSE} otherwise. @code{(NAME)}, which is used by the outer
   interpreter (aka text interpreter) by default, treats all white-space
   characters as delimiters.
   
   @item format of the control flow stack:
   The data stack is used as control flow stack. The size of a control flow
   stack item in cells is given by the constant @code{cs-item-size}. At the
   time of this writing, an item consists of a (pointer to a) locals list
   (third), an address in the code (second), and a tag for identifying the
   item (TOS). The following tags are used: @code{defstart},
   @code{live-orig}, @code{dead-orig}, @code{dest}, @code{do-dest},
   @code{scopestart}.
   
   @item conversion of digits > 35
   The characters @code{[\]^_'} are the digits with the decimal value
   36@minus{}41. There is no way to input many of the larger digits.
   
   @item display after input terminates in @code{ACCEPT} and @code{EXPECT}:
   The cursor is moved to the end of the entered string. If the input is
   terminated using the @kbd{Return} key, a space is typed.
   
   @item exception abort sequence of @code{ABORT"}:
   The error string is stored into the variable @code{"error} and a
   @code{-2 throw} is performed.
   
   @item input line terminator:
   For interactive input, @kbd{C-m} and @kbd{C-j} terminate lines. One of
   these characters is typically produced when you type the @kbd{Enter} or
   @kbd{Return} key.
   
   @item maximum size of a counted string:
   @code{s" /counted-string" environment? drop .}. Currently 255 characters
   on all ports, but this may change.
   
   @item maximum size of a parsed string:
   Given by the constant @code{/line}. Currently 255 characters.
   
   @item maximum size of a definition name, in characters:
   31
   
   @item maximum string length for @code{ENVIRONMENT?}, in characters:
   31
   
   @item method of selecting the user input device:
   The user input device is the standard input. There is currently no way to
   change it from within Gforth. However, the input can typically be
   redirected in the command line that starts Gforth.
   
   @item method of selecting the user output device:
   The user output device is the standard output. It cannot be redirected
   from within Gforth, but typically from the command line that starts
   Gforth. Gforth uses buffered output, so output on a terminal does not
   become visible before the next newline or buffer overflow. Output on
   non-terminals is invisible until the buffer overflows.
   
   @item methods of dictionary compilation:
   What are we expected to document here?
   
   @item number of bits in one address unit:
   @code{s" address-units-bits" environment? drop .}. 8 in all current
   ports.
   
   @item number representation and arithmetic:
   Processor-dependent. Binary two's complement on all current ports.
   
   @item ranges for integer types:
   Installation-dependent. Make environmental queries for @code{MAX-N},
   @code{MAX-U}, @code{MAX-D} and @code{MAX-UD}. The lower bounds for
   unsigned (and positive) types is 0. The lower bound for signed types on
   two's complement and one's complement machines machines can be computed
   by adding 1 to the upper bound.
   
   @item read-only data space regions:
   The whole Forth data space is writable.
   
   @item size of buffer at @code{WORD}:
   @code{PAD HERE - .}. 104 characters on 32-bit machines. The buffer is
   shared with the pictured numeric output string. If overwriting
   @code{PAD} is acceptable, it is as large as the remaining dictionary
   space, although only as much can be sensibly used as fits in a counted
   string.
   
   @item size of one cell in address units:
   @code{1 cells .}.
   
   @item size of one character in address units:
   @code{1 chars .}. 1 on all current ports.
   
   @item size of the keyboard terminal buffer:
   Varies. You can determine the size at a specific time using @code{lp@
   tib - .}. It is shared with the locals stack and TIBs of files that
   include the current file. You can change the amount of space for TIBs
   and locals stack at Gforth startup with the command line option
   @code{-l}.
   
   @item size of the pictured numeric output buffer:
   @code{PAD HERE - .}. 104 characters on 32-bit machines. The buffer is
   shared with @code{WORD}.
   
   @item size of the scratch area returned by @code{PAD}:
   The remainder of dictionary space. You can even use the unused part of
   the data stack space. The current size can be computed with @code{sp@
   pad - .}.
   
   @item system case-sensitivity characteristics:
   Dictionary searches are case insensitive. However, as explained above
   under @i{character-set extensions}, the matching for non-ASCII
   characters is determined by the locale you are using. In the default
   @code{C} locale all non-ASCII characters are matched case-sensitively.
   
   @item system prompt:
   @code{ ok} in interpret state, @code{ compiled} in compile state.
   
   @item division rounding:
   installation dependent. @code{s" floored" environment? drop .}. We leave
   the choice to gcc (what to use for @code{/}) and to you (whether to use
   @code{fm/mod}, @code{sm/rem} or simply @code{/}).
   
   @item values of @code{STATE} when true:
   -1.
   
   @item values returned after arithmetic overflow:
   On two's complement machines, arithmetic is performed modulo
   2**bits-per-cell for single arithmetic and 4**bits-per-cell for double
   arithmetic (with appropriate mapping for signed types). Division by zero
   typically results in a @code{-55 throw} (floatingpoint unidentified
   fault), although a @code{-10 throw} (divide by zero) would be more
   appropriate.
   
   @item whether the current definition can be found after @t{DOES>}:
   No.
   
   @end table
   
   @c ---------------------------------------------------------------------
   @node core-ambcond, core-other, core-idef, The Core Words
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item a name is neither a word nor a number:
   @code{-13 throw} (Undefined word)
   
   @item a definition name exceeds the maximum length allowed:
   @code{-19 throw} (Word name too long)
   
   @item addressing a region not inside the various data spaces of the forth system:
   The stacks, code space and name space are accessible. Machine code space is
   typically readable. Accessing other addresses gives results dependent on
   the operating system. On decent systems: @code{-9 throw} (Invalid memory
   address).
   
   @item argument type incompatible with parameter:
   This is usually not caught. Some words perform checks, e.g., the control
   flow words, and issue a @code{ABORT"} or @code{-12 THROW} (Argument type
   mismatch).
   
   @item attempting to obtain the execution token of a word with undefined execution semantics:
   You get an execution token representing the compilation semantics
   instead.
   
   @item dividing by zero:
   typically results in a @code{-55 throw} (floating point unidentified
   fault), although a @code{-10 throw} (divide by zero) would be more
   appropriate.
   
   @item insufficient data stack or return stack space:
   Not checked. This typically results in mysterious illegal memory
   accesses, producing @code{-9 throw} (Invalid memory address) or
   @code{-23 throw} (Address alignment exception).
   
   @item insufficient space for loop control parameters:
   like other return stack overflows.
   
   @item insufficient space in the dictionary:
   Not checked. Similar results as stack overflows. However, typically the
   error appears at a different place when one inserts or removes code.
   
   @item interpreting a word with undefined interpretation semantics:
   For some words, we defined interpretation semantics. For the others:
   @code{-14 throw} (Interpreting a compile-only word). Note that this is
   checked only by the outer (aka text) interpreter; if the word is
   @code{execute}d in some other way, it will typically perform it's
   compilation semantics even in interpret state. (We could change @code{'}
   and relatives not to give the xt of such words, but we think that would
   be too restrictive).
   
   @item modifying the contents of the input buffer or a string literal:
   These are located in writable memory and can be modified.
   
   @item overflow of the pictured numeric output string:
   Not checked.
   
   @item parsed string overflow:
   @code{PARSE} cannot overflow. @code{WORD} does not check for overflow.
   
   @item producing a result out of range:
   On two's complement machines, arithmetic is performed modulo
   2**bits-per-cell for single arithmetic and 4**bits-per-cell for double
   arithmetic (with appropriate mapping for signed types). Division by zero
   typically results in a @code{-55 throw} (floatingpoint unidentified
   fault), although a @code{-10 throw} (divide by zero) would be more
   appropriate. @code{convert} and @code{>number} currently overflow
   silently.
   
   @item reading from an empty data or return stack:
   The data stack is checked by the outer (aka text) interpreter after
   every word executed. If it has underflowed, a @code{-4 throw} (Stack
   underflow) is performed. Apart from that, the stacks are not checked and
   underflows can result in similar behaviour as overflows (of adjacent
   stacks).
   
   @item unexepected end of the input buffer, resulting in an attempt to use a zero-length string as a name:
   @code{Create} and its descendants perform a @code{-16 throw} (Attempt to
   use zero-length string as a name). Words like @code{'} probably will not
   find what they search. Note that it is possible to create zero-length
   names with @code{nextname} (should it not?).
   
   @item @code{>IN} greater than input buffer:
   The next invocation of a parsing word returns a string wih length 0.
   
   @item @code{RECURSE} appears after @code{DOES>}:
   Compiles a recursive call to the defining word not to the defined word.
   
   @item argument input source different than current input source for @code{RESTORE-INPUT}:
   !!???If the argument input source is a valid input source then it gets
   restored. Otherwise causes @code{-12 THROW}, which, unless caught, issues
   the message "argument type mismatch" and aborts.
   
   @item data space containing definitions gets de-allocated:
   Deallocation with @code{allot} is not checked. This typically resuls in
   memory access faults or execution of illegal instructions.
   
   @item data space read/write with incorrect alignment:
   Processor-dependent. Typically results in a @code{-23 throw} (Address
   alignment exception). Under Linux on a 486 or later processor with
   alignment turned on, incorrect alignment results in a @code{-9 throw}
   (Invalid memory address). There are reportedly some processors with
   alignment restrictions that do not report them.
   
   @item data space pointer not properly aligned, @code{,}, @code{C,}:
   Like other alignment errors.
   
   @item less than u+2 stack items (@code{PICK} and @code{ROLL}):
   Not checked. May cause an illegal memory access.
   
   @item loop control parameters not available:
   Not checked. The counted loop words simply assume that the top of return
   stack items are loop control parameters and behave accordingly.
   
   @item most recent definition does not have a name (@code{IMMEDIATE}):
   @code{abort" last word was headerless"}.
   
   @item name not defined by @code{VALUE} used by @code{TO}:
   @code{-32 throw} (Invalid name argument)
   
   @item name not found (@code{'}, @code{POSTPONE}, @code{[']}, @code{[COMPILE]}):
   @code{-13 throw} (Undefined word)
   
   @item parameters are not of the same type (@code{DO}, @code{?DO}, @code{WITHIN}):
   Gforth behaves as if they were of the same type. I.e., you can predict
   the behaviour by interpreting all parameters as, e.g., signed.
   
   @item @code{POSTPONE} or @code{[COMPILE]} applied to @code{TO}:
   Assume @code{: X POSTPONE TO ; IMMEDIATE}. @code{X} is equivalent to
   @code{TO}.
   
   @item String longer than a counted string returned by @code{WORD}:
   Not checked. The string will be ok, but the count will, of course,
   contain only the least significant bits of the length.
   
   @item u greater than or equal to the number of bits in a cell (@code{LSHIFT}, @code{RSHIFT}):
   Processor-dependent. Typical behaviours are returning 0 and using only
   the low bits of the shift count.
   
   @item word not defined via @code{CREATE}:
   @code{>BODY} produces the PFA of the word no matter how it was defined.
   
   @code{DOES>} changes the execution semantics of the last defined word no
   matter how it was defined. E.g., @code{CONSTANT DOES>} is equivalent to
   @code{CREATE , DOES>}.
   
   @item words improperly used outside @code{<#} and @code{#>}:
   Not checked. As usual, you can expect memory faults.
   
   @end table
   
   
   @c ---------------------------------------------------------------------
   @node core-other,  , core-ambcond, The Core Words
   @subsection Other system documentation
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item nonstandard words using @code{PAD}:
   None.
   
   @item operator's terminal facilities available:
   !!??
   
   @item program data space available:
   @code{sp@ here - .} gives the space remaining for dictionary and data
   stack together.
   
   @item return stack space available:
   !!??
   
   @item stack space available:
   @code{sp@ here - .} gives the space remaining for dictionary and data
   stack together.
   
   @item system dictionary space required, in address units:
   Type @code{here forthstart - .} after startup. At the time of this
   writing, this gives 70108 (bytes) on a 32-bit system.
   @end table
   
   
   @c =====================================================================
   @node The optional Block word set, The optional Double Number word set, The Core Words, ANS conformance
   @section The optional Block word set
   @c =====================================================================
   
   @menu
   * block-idef::                  Implementation Defined Options                  
   * block-ambcond::               Ambiguous Conditions               
   * block-other::                 Other System Documentation                 
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node block-idef, block-ambcond, The optional Block word set, The optional Block word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item the format for display by @code{LIST}:
   First the screen number is displayed, then 16 lines of 64 characters,
   each line preceded by the line number.
   
   @item the length of a line affected by @code{\}:
   64 characters.
   @end table
   
   
   @c ---------------------------------------------------------------------
   @node block-ambcond, block-other, block-idef, The optional Block word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item correct block read was not possible:
   Typically results in a @code{throw} of some OS-derived value (between
   -512 and -2048). If the blocks file was just not long enough, blanks are
   supplied for the missing portion.
   
   @item I/O exception in block transfer:
   Typically results in a @code{throw} of some OS-derived value (between
   -512 and -2048).
   
   @item invalid block number:
   @code{-35 throw} (Invalid block number)
   
   @item a program directly alters the contents of @code{BLK}:
   The input stream is switched to that other block, at the same
   position. If the storing to @code{BLK} happens when interpreting
   non-block input, the system will get quite confused when the block ends.
   
   @item no current block buffer for @code{UPDATE}:
   @code{UPDATE} has no effect.
   
   @end table
   
   
   @c ---------------------------------------------------------------------
   @node block-other,  , block-ambcond, The optional Block word set
   @subsection Other system documentation
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item any restrictions a multiprogramming system places on the use of buffer addresses:
   No restrictions (yet).
   
   @item the number of blocks available for source and data:
   depends on your disk space.
   
   @end table
   
   
   @c =====================================================================
   @node The optional Double Number word set, The optional Exception word set, The optional Block word set, ANS conformance
   @section The optional Double Number word set
   @c =====================================================================
   
   @menu
   * double-ambcond::              Ambiguous Conditions              
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node double-ambcond,  , The optional Double Number word set, The optional Double Number word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item @var{d} outside of range of @var{n} in @code{D>S}:
   The least significant cell of @var{d} is produced.
   
   @end table
   
   
   @c =====================================================================
   @node The optional Exception word set, The optional Facility word set, The optional Double Number word set, ANS conformance
   @section The optional Exception word set
   @c =====================================================================
   
   @menu
   * exception-idef::              Implementation Defined Options              
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node exception-idef,  , The optional Exception word set, The optional Exception word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   @item @code{THROW}-codes used in the system:
   The codes -256@minus{}-511 are used for reporting signals (see
   @file{errore.fs}). The codes -512@minus{}-2047 are used for OS errors
   (for file and memory allocation operations). The mapping from OS error
   numbers to throw code is -512@minus{}@var{errno}. One side effect of
   this mapping is that undefined OS errors produce a message with a
   strange number; e.g., @code{-1000 THROW} results in @code{Unknown error
   488} on my system.
   @end table
   
   @c =====================================================================
   @node The optional Facility word set, The optional File-Access word set, The optional Exception word set, ANS conformance
   @section The optional Facility word set
   @c =====================================================================
   
   @menu
   * facility-idef::               Implementation Defined Options               
   * facility-ambcond::            Ambiguous Conditions            
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node facility-idef, facility-ambcond, The optional Facility word set, The optional Facility word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item encoding of keyboard events (@code{EKEY}):
   Not yet implemeted.
   
   @item duration of a system clock tick
   System dependent. With respect to @code{MS}, the time is specified in
   microseconds. How well the OS and the hardware implement this, is
   another question.
   
   @item repeatability to be expected from the execution of @code{MS}:
   System dependent. On Unix, a lot depends on load. If the system is
   lightly loaded, and the delay is short enough that Gforth does not get
   swapped out, the performance should be acceptable. Under MS-DOS and
   other single-tasking systems, it should be good.
   
   @end table
   
   
   @c ---------------------------------------------------------------------
   @node facility-ambcond,  , facility-idef, The optional Facility word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item @code{AT-XY} can't be performed on user output device:
   Largely terminal dependant. No range checks are done on the arguments.
   No errors are reported. You may see some garbage appearing, you may see
   simply nothing happen.
   
   @end table
   
   
   @c =====================================================================
   @node The optional File-Access word set, The optional Floating-Point word set, The optional Facility word set, ANS conformance
   @section The optional File-Access word set
   @c =====================================================================
   
   @menu
   * file-idef::                   Implementation Defined Options                   
   * file-ambcond::                Ambiguous Conditions                
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node file-idef, file-ambcond, The optional File-Access word set, The optional File-Access word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item File access methods used:
   @code{R/O}, @code{R/W} and @code{BIN} work as you would
   expect. @code{W/O} translates into the C file opening mode @code{w} (or
   @code{wb}): The file is cleared, if it exists, and created, if it does
   not (both with @code{open-file} and @code{create-file}).  Under Unix
   @code{create-file} creates a file with 666 permissions modified by your
   umask.
   
   @item file exceptions:
   The file words do not raise exceptions (except, perhaps, memory access
   faults when you pass illegal addresses or file-ids).
   
   @item file line terminator:
   System-dependent. Gforth uses C's newline character as line
   terminator. What the actual character code(s) of this are is
   system-dependent.
   
   @item file name format
   System dependent. Gforth just uses the file name format of your OS.
   
   @item information returned by @code{FILE-STATUS}:
   @code{FILE-STATUS} returns the most powerful file access mode allowed
   for the file: Either @code{R/O}, @code{W/O} or @code{R/W}. If the file
   cannot be accessed, @code{R/O BIN} is returned. @code{BIN} is applicable
   along with the retured mode.
   
   @item input file state after an exception when including source:
   All files that are left via the exception are closed.
   
   @item @var{ior} values and meaning:
   The @var{ior}s returned by the file and memory allocation words are
   intended as throw codes. They typically are in the range
   -512@minus{}-2047 of OS errors.  The mapping from OS error numbers to
   @var{ior}s is -512@minus{}@var{errno}.
   
   @item maximum depth of file input nesting:
   limited by the amount of return stack, locals/TIB stack, and the number
   of open files available. This should not give you troubles.
   
   @item maximum size of input line:
   @code{/line}. Currently 255.
   
   @item methods of mapping block ranges to files:
   Currently, the block words automatically access the file
   @file{blocks.fb} in the currend working directory. More sophisticated
   methods could be implemented if there is demand (and a volunteer).
   
   @item number of string buffers provided by @code{S"}:
   1
   
   @item size of string buffer used by @code{S"}:
   @code{/line}. currently 255.
   
   @end table
   
   @c ---------------------------------------------------------------------
   @node file-ambcond,  , file-idef, The optional File-Access word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item attempting to position a file outside it's boundaries:
   @code{REPOSITION-FILE} is performed as usual: Afterwards,
   @code{FILE-POSITION} returns the value given to @code{REPOSITION-FILE}.
   
   @item attempting to read from file positions not yet written:
   End-of-file, i.e., zero characters are read and no error is reported.
   
   @item @var{file-id} is invalid (@code{INCLUDE-FILE}):
   An appropriate exception may be thrown, but a memory fault or other
   problem is more probable.
   
   @item I/O exception reading or closing @var{file-id} (@code{include-file}, @code{included}):
   The @var{ior} produced by the operation, that discovered the problem, is
   thrown.
   
   @item named file cannot be opened (@code{included}):
   The @var{ior} produced by @code{open-file} is thrown.
   
   @item requesting an unmapped block number:
   There are no unmapped legal block numbers. On some operating systems,
   writing a block with a large number may overflow the file system and
   have an error message as consequence.
   
   @item using @code{source-id} when @code{blk} is non-zero:
   @code{source-id} performs its function. Typically it will give the id of
   the source which loaded the block. (Better ideas?)
   
   @end table
   
   
   @c =====================================================================
   @node  The optional Floating-Point word set, The optional Locals word set, The optional File-Access word set, ANS conformance
   @section The optional Floating-Point word set
   @c =====================================================================
   
   @menu
   * floating-idef::               Implementation Defined Options
   * floating-ambcond::            Ambiguous Conditions            
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node floating-idef, floating-ambcond, The optional Floating-Point word set, The optional Floating-Point word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item format and range of floating point numbers:
   System-dependent; the @code{double} type of C.
   
   @item results of @code{REPRESENT} when @var{float} is out of range:
   System dependent; @code{REPRESENT} is implemented using the C library
   function @code{ecvt()} and inherits its behaviour in this respect.
   
   @item rounding or truncation of floating-point numbers:
   What's the question?!!
   
   @item size of floating-point stack:
   @code{s" FLOATING-STACK" environment? drop .}. Can be changed at startup
   with the command-line option @code{-f}.
   
   @item width of floating-point stack:
   @code{1 floats}.
   
   @end table
   
   
   @c ---------------------------------------------------------------------
   @node floating-ambcond,  , floating-idef, The optional Floating-Point word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item @code{df@@} or @code{df!} used with an address that is not double-float  aligned:
   System-dependent. Typically results in an alignment fault like other
   alignment violations.
   
   @item @code{f@@} or @code{f!} used with an address that is not float  aligned:
   System-dependent. Typically results in an alignment fault like other
   alignment violations.
   
   @item Floating-point result out of range:
   System-dependent. Can result in a @code{-55 THROW} (Floating-point
   unidentified fault), or can produce a special value representing, e.g.,
   Infinity.
   
   @item @code{sf@@} or @code{sf!} used with an address that is not single-float  aligned:
   System-dependent. Typically results in an alignment fault like other
   alignment violations.
   
   @item BASE is not decimal (@code{REPRESENT}, @code{F.}, @code{FE.}, @code{FS.}):
   The floating-point number is converted into decimal nonetheless.
   
   @item Both arguments are equal to zero (@code{FATAN2}):
   System-dependent. @code{FATAN2} is implemented using the C library
   function @code{atan2()}.
   
   @item Using ftan on an argument @var{r1} where cos(@var{r1}) is zero:
   System-dependent. Anyway, typically the cos of @var{r1} will not be zero
   because of small errors and the tan will be a very large (or very small)
   but finite number.
   
   @item @var{d} cannot be presented precisely as a float in @code{D>F}:
   The result is rounded to the nearest float.
   
   @item dividing by zero:
   @code{-55 throw} (Floating-point unidentified fault)
   
   @item exponent too big for conversion (@code{DF!}, @code{DF@@}, @code{SF!}, @code{SF@@}):
   System dependent. On IEEE-FP based systems the number is converted into
   an infinity.
   
   @item @var{float}<1 (@code{facosh}):
   @code{-55 throw} (Floating-point unidentified fault)
   
   @item @var{float}=<-1 (@code{flnp1}):
   @code{-55 throw} (Floating-point unidentified fault). On IEEE-FP systems
   negative infinity is typically produced for @var{float}=-1.
   
   @item @var{float}=<0 (@code{fln}, @code{flog}):
   @code{-55 throw} (Floating-point unidentified fault). On IEEE-FP systems
   negative infinity is typically produced for @var{float}=0.
   
   @item @var{float}<0 (@code{fasinh}, @code{fsqrt}):
   @code{-55 throw} (Floating-point unidentified fault). @code{fasinh}
   produces values for these inputs on my Linux box (Bug in the C library?)
   
   @item |@var{float}|>1 (@code{facos}, @code{fasin}, @code{fatanh}):
   @code{-55 throw} (Floating-point unidentified fault).
   
   @item integer part of float cannot be represented by @var{d} in @code{f>d}:
   @code{-55 throw} (Floating-point unidentified fault).
   
   @item string larger than pictured numeric output area (@code{f.}, @code{fe.}, @code{fs.}):
   This does not happen.
   @end table
   
   
   
   @c =====================================================================
   @node  The optional Locals word set, The optional Memory-Allocation word set, The optional Floating-Point word set, ANS conformance
   @section The optional Locals word set
   @c =====================================================================
   
   @menu
   * locals-idef::                 Implementation Defined Options                 
   * locals-ambcond::              Ambiguous Conditions              
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node locals-idef, locals-ambcond, The optional Locals word set, The optional Locals word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item maximum number of locals in a definition:
   @code{s" #locals" environment? drop .}. Currently 15. This is a lower
   bound, e.g., on a 32-bit machine there can be 41 locals of up to 8
   characters. The number of locals in a definition is bounded by the size
   of locals-buffer, which contains the names of the locals.
   
   @end table
   
   
   @c ---------------------------------------------------------------------
   @node locals-ambcond,  , locals-idef, The optional Locals word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item executing a named local in interpretation state:
   @code{-14 throw} (Interpreting a compile-only word).
   
   @item @var{name} not defined by @code{VALUE} or @code{(LOCAL)} (@code{TO}):
   @code{-32 throw} (Invalid name argument)
   
   @end table
   
   
   @c =====================================================================
   @node  The optional Memory-Allocation word set, The optional Programming-Tools word set, The optional Locals word set, ANS conformance
   @section The optional Memory-Allocation word set
   @c =====================================================================
   
   @menu
   * memory-idef::                 Implementation Defined Options                 
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node memory-idef,  , The optional Memory-Allocation word set, The optional Memory-Allocation word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item values and meaning of @var{ior}:
   The @var{ior}s returned by the file and memory allocation words are
   intended as throw codes. They typically are in the range
   -512@minus{}-2047 of OS errors.  The mapping from OS error numbers to
   @var{ior}s is -512@minus{}@var{errno}.
   
   @end table
   
   @c =====================================================================
   @node  The optional Programming-Tools word set, The optional Search-Order word set, The optional Memory-Allocation word set, ANS conformance
   @section The optional Programming-Tools word set
   @c =====================================================================
   
   @menu
   * programming-idef::            Implementation Defined Options            
   * programming-ambcond::         Ambiguous Conditions         
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node programming-idef, programming-ambcond, The optional Programming-Tools word set, The optional Programming-Tools word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item ending sequence for input following @code{;code} and @code{code}:
   Not implemented (yet).
   
   @item manner of processing input following @code{;code} and @code{code}:
   Not implemented (yet).
   
   @item search order capability for @code{EDITOR} and @code{ASSEMBLER}:
   Not implemented (yet). If they were implemented, they would use the
   search order wordset.
   
   @item source and format of display by @code{SEE}:
   The source for @code{see} is the intermediate code used by the inner
   interpreter.  The current @code{see} tries to output Forth source code
   as well as possible.
   
   @end table
   
   @c ---------------------------------------------------------------------
   @node programming-ambcond,  , programming-idef, The optional Programming-Tools word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item deleting the compilation wordlist (@code{FORGET}):
   Not implemented (yet).
   
   @item fewer than @var{u}+1 items on the control flow stack (@code{CS-PICK}, @code{CS-ROLL}):
   This typically results in an @code{abort"} with a descriptive error
   message (may change into a @code{-22 throw} (Control structure mismatch)
   in the future). You may also get a memory access error. If you are
   unlucky, this ambiguous condition is not caught.
   
   @item @var{name} can't be found (@code{forget}):
   Not implemented (yet).
   
   @item @var{name} not defined via @code{CREATE}:
   @code{;code} is not implemented (yet). If it were, it would behave like
   @code{DOES>} in this respect, i.e., change the execution semantics of
   the last defined word no matter how it was defined.
   
   @item @code{POSTPONE} applied to @code{[IF]}:
   After defining @code{: X POSTPONE [IF] ; IMMEDIATE}. @code{X} is
   equivalent to @code{[IF]}.
   
   @item reaching the end of the input source before matching @code{[ELSE]} or @code{[THEN]}:
   Continue in the same state of conditional compilation in the next outer
   input source. Currently there is no warning to the user about this.
   
   @item removing a needed definition (@code{FORGET}):
   Not implemented (yet).
   
   @end table
   
   
   @c =====================================================================
   @node  The optional Search-Order word set,  , The optional Programming-Tools word set, ANS conformance
   @section The optional Search-Order word set
   @c =====================================================================
   
   @menu
   * search-idef::                 Implementation Defined Options                 
   * search-ambcond::              Ambiguous Conditions              
   @end menu
   
   
   @c ---------------------------------------------------------------------
   @node search-idef, search-ambcond, The optional Search-Order word set, The optional Search-Order word set
   @subsection Implementation Defined Options
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item maximum number of word lists in search order:
   @code{s" wordlists" environment? drop .}. Currently 16.
   
   @item minimum search order:
   @code{root root}.
   
   @end table
   
   @c ---------------------------------------------------------------------
   @node search-ambcond,  , search-idef, The optional Search-Order word set
   @subsection Ambiguous conditions
   @c ---------------------------------------------------------------------
   
   @table @i
   
   @item changing the compilation wordlist (during compilation):
   The definition is put into the wordlist that is the compilation wordlist
   when @code{REVEAL} is executed (by @code{;}, @code{DOES>},
   @code{RECURSIVE}, etc.).
   
   @item search order empty (@code{previous}):
   @code{abort" Vocstack empty"}.
   
   @item too many word lists in search order (@code{also}):
   @code{abort" Vocstack full"}.
   
   @end table
   
   
   @node Model, Emacs and Gforth, ANS conformance, Top
   @chapter Model
   
   @node Emacs and Gforth, Internals, Model, Top
   @chapter Emacs and Gforth
   
   Gforth comes with @file{gforth.el}, an improved version of
   @file{forth.el} by Goran Rydqvist (icluded in the TILE package). The
   improvements are a better (but still not perfect) handling of
   indentation. I have also added comment paragraph filling (@kbd{M-q}),
   commenting (@kbd{C-x \}) and uncommenting (@kbd{C-u C-x \}) regions and
   removing debugging tracers (@kbd{C-x ~}, @pxref{Debugging}). I left the
   stuff I do not use alone, even though some of it only makes sense for
   TILE. To get a description of these features, enter Forth mode and type
   @kbd{C-h m}.
   
   In addition, Gforth supports Emacs quite well: The source code locations
   given in error messages, debugging output (from @code{~~}) and failed
   assertion messages are in the right format for Emacs' compilation mode
   (@pxref{Compilation, , Running Compilations under Emacs, emacs, Emacs
   Manual}) so the source location corresponding to an error or other
   message is only a few keystrokes away (@kbd{C-x `} for the next error,
   @kbd{C-c C-c} for the error under the cursor).
   
   Also, if you @code{include} @file{etags.fs}, a new @file{TAGS} file
   (@pxref{Tags, , Tags Tables, emacs, Emacs Manual}) will be produced that
   contains the definitions of all words defined afterwards. You can then
   find the source for a word using @kbd{M-.}. Note that emacs can use
   several tags files at the same time (e.g., one for the Gforth sources
   and one for your program).
   
   To get all these benefits, add the following lines to your @file{.emacs}
   file:
   
   @example
   (autoload 'forth-mode "gforth.el")
   (setq auto-mode-alist (cons '("\\.fs\\'" . forth-mode) auto-mode-alist))
   @end example
   
   @node Internals, Bugs, Emacs and Gforth, Top
   @chapter Internals
   
   Reading this section is not necessary for programming with Gforth. It
   should be helpful for finding your way in the Gforth sources.
   
   @menu
   * Portability::                 
   * Threading::                   
   * Primitives::                  
   * System Architecture::         
   * Performance::                 
   @end menu
   
   @node Portability, Threading, Internals, Internals
   @section Portability
   
   One of the main goals of the effort is availability across a wide range
   of personal machines. fig-Forth, and, to a lesser extent, F83, achieved
   this goal by manually coding the engine in assembly language for several
   then-popular processors. This approach is very labor-intensive and the
   results are short-lived due to progress in computer architecture.
   
   Others have avoided this problem by coding in C, e.g., Mitch Bradley
   (cforth), Mikael Patel (TILE) and Dirk Zoller (pfe). This approach is
   particularly popular for UNIX-based Forths due to the large variety of
   architectures of UNIX machines. Unfortunately an implementation in C
   does not mix well with the goals of efficiency and with using
   traditional techniques: Indirect or direct threading cannot be expressed
   in C, and switch threading, the fastest technique available in C, is
   significantly slower. Another problem with C is that it's very
   cumbersome to express double integer arithmetic.
   
   Fortunately, there is a portable language that does not have these
   limitations: GNU C, the version of C processed by the GNU C compiler
   (@pxref{C Extensions, , Extensions to the C Language Family, gcc.info,
   GNU C Manual}). Its labels as values feature (@pxref{Labels as Values, ,
   Labels as Values, gcc.info, GNU C Manual}) makes direct and indirect
   threading possible, its @code{long long} type (@pxref{Long Long, ,
   Double-Word Integers, gcc.info, GNU C Manual}) corresponds to Forths
   double numbers. GNU C is available for free on all important (and many
   unimportant) UNIX machines, VMS, 80386s running MS-DOS, the Amiga, and
   the Atari ST, so a Forth written in GNU C can run on all these
   machines.
   
   Writing in a portable language has the reputation of producing code that
   is slower than assembly. For our Forth engine we repeatedly looked at
   the code produced by the compiler and eliminated most compiler-induced
   inefficiencies by appropriate changes in the source-code.
   
   However, register allocation cannot be portably influenced by the
   programmer, leading to some inefficiencies on register-starved
   machines. We use explicit register declarations (@pxref{Explicit Reg
   Vars, , Variables in Specified Registers, gcc.info, GNU C Manual}) to
   improve the speed on some machines. They are turned on by using the
   @code{gcc} switch @code{-DFORCE_REG}. Unfortunately, this feature not
   only depends on the machine, but also on the compiler version: On some
   machines some compiler versions produce incorrect code when certain
   explicit register declarations are used. So by default
   @code{-DFORCE_REG} is not used.
   
   @node Threading, Primitives, Portability, Internals
   @section Threading
   
   GNU C's labels as values extension (available since @code{gcc-2.0},
   @pxref{Labels as Values, , Labels as Values, gcc.info, GNU C Manual})
   makes it possible to take the address of @var{label} by writing
   @code{&&@var{label}}.  This address can then be used in a statement like
   @code{goto *@var{address}}. I.e., @code{goto *&&x} is the same as
   @code{goto x}.
   
   With this feature an indirect threaded NEXT looks like:
   @example
   cfa = *ip++;
   ca = *cfa;
   goto *ca;
   @end example
   For those unfamiliar with the names: @code{ip} is the Forth instruction
   pointer; the @code{cfa} (code-field address) corresponds to ANS Forths
   execution token and points to the code field of the next word to be
   executed; The @code{ca} (code address) fetched from there points to some
   executable code, e.g., a primitive or the colon definition handler
   @code{docol}.
   
   Direct threading is even simpler:
   @example
   ca = *ip++;
   goto *ca;
   @end example
   
   Of course we have packaged the whole thing neatly in macros called
   @code{NEXT} and @code{NEXT1} (the part of NEXT after fetching the cfa).
   
   @menu
   * Scheduling::                  
   * Direct or Indirect Threaded?::  
   * DOES>::                       
   @end menu
   
   @node Scheduling, Direct or Indirect Threaded?, Threading, Threading
   @subsection Scheduling
   
   There is a little complication: Pipelined and superscalar processors,
   i.e., RISC and some modern CISC machines can process independent
   instructions while waiting for the results of an instruction. The
   compiler usually reorders (schedules) the instructions in a way that
   achieves good usage of these delay slots. However, on our first tries
   the compiler did not do well on scheduling primitives. E.g., for
   @code{+} implemented as
   @example
   n=sp[0]+sp[1];
   sp++;
   sp[0]=n;
   NEXT;
   @end example
   the NEXT comes strictly after the other code, i.e., there is nearly no
   scheduling. After a little thought the problem becomes clear: The
   compiler cannot know that sp and ip point to different addresses (and
   the version of @code{gcc} we used would not know it even if it was
   possible), so it could not move the load of the cfa above the store to
   the TOS. Indeed the pointers could be the same, if code on or very near
   the top of stack were executed. In the interest of speed we chose to
   forbid this probably unused ``feature'' and helped the compiler in
   scheduling: NEXT is divided into the loading part (@code{NEXT_P1}) and
   the goto part (@code{NEXT_P2}). @code{+} now looks like:
   @example
   n=sp[0]+sp[1];
   sp++;
   NEXT_P1;
   sp[0]=n;
   NEXT_P2;
   @end example
   This can be scheduled optimally by the compiler.
   
   This division can be turned off with the switch @code{-DCISC_NEXT}. This
   switch is on by default on machines that do not profit from scheduling
   (e.g., the 80386), in order to preserve registers.
   
   @node Direct or Indirect Threaded?, DOES>, Scheduling, Threading
   @subsection Direct or Indirect Threaded?
   
   Both! After packaging the nasty details in macro definitions we
   realized that we could switch between direct and indirect threading by
   simply setting a compilation flag (@code{-DDIRECT_THREADED}) and
   defining a few machine-specific macros for the direct-threading case.
   On the Forth level we also offer access words that hide the
   differences between the threading methods (@pxref{Threading Words}).
   
   Indirect threading is implemented completely
   machine-independently. Direct threading needs routines for creating
   jumps to the executable code (e.g. to docol or dodoes). These routines
   are inherently machine-dependent, but they do not amount to many source
   lines. I.e., even porting direct threading to a new machine is a small
   effort.
   
   @node DOES>,  , Direct or Indirect Threaded?, Threading
   @subsection DOES>
   One of the most complex parts of a Forth engine is @code{dodoes}, i.e.,
   the chunk of code executed by every word defined by a
   @code{CREATE}...@code{DOES>} pair. The main problem here is: How to find
   the Forth code to be executed, i.e. the code after the @code{DOES>} (the
   DOES-code)? There are two solutions:
   
   In fig-Forth the code field points directly to the dodoes and the
   DOES-code address is stored in the cell after the code address
   (i.e. at cfa cell+). It may seem that this solution is illegal in the
   Forth-79 and all later standards, because in fig-Forth this address
   lies in the body (which is illegal in these standards). However, by
   making the code field larger for all words this solution becomes legal
   again. We use this approach for the indirect threaded version. Leaving
   a cell unused in most words is a bit wasteful, but on the machines we
   are targetting this is hardly a problem. The other reason for having a
   code field size of two cells is to avoid having different image files
   for direct and indirect threaded systems (@pxref{System Architecture}).
   
   The other approach is that the code field points or jumps to the cell
   after @code{DOES}. In this variant there is a jump to @code{dodoes} at
   this address. @code{dodoes} can then get the DOES-code address by
   computing the code address, i.e., the address of the jump to dodoes,
   and add the length of that jump field. A variant of this is to have a
   call to @code{dodoes} after the @code{DOES>}; then the return address
   (which can be found in the return register on RISCs) is the DOES-code
   address. Since the two cells available in the code field are usually
   used up by the jump to the code address in direct threading, we use
   this approach for direct threading. We did not want to add another
   cell to the code field.
   
   @node Primitives, System Architecture, Threading, Internals
   @section Primitives
   
   @menu
   * Automatic Generation::        
   * TOS Optimization::            
   * Produced code::               
   @end menu
   
   @node Automatic Generation, TOS Optimization, Primitives, Primitives
   @subsection Automatic Generation
   
   Since the primitives are implemented in a portable language, there is no
   longer any need to minimize the number of primitives. On the contrary,
   having many primitives is an advantage: speed. In order to reduce the
   number of errors in primitives and to make programming them easier, we
   provide a tool, the primitive generator (@file{prims2x.fs}), that
   automatically generates most (and sometimes all) of the C code for a
   primitive from the stack effect notation.  The source for a primitive
   has the following form:
   
   @format
   @var{Forth-name}        @var{stack-effect}      @var{category}  [@var{pronounc.}]
   [@code{""}@var{glossary entry}@code{""}]
   @var{C code}
   [@code{:}
   @var{Forth code}]
   @end format
   
   The items in brackets are optional. The category and glossary fields
   are there for generating the documentation, the Forth code is there
   for manual implementations on machines without GNU C. E.g., the source
   for the primitive @code{+} is:
   @example
   +    n1 n2 -- n    core    plus
   n = n1+n2;
   @end example
   
   This looks like a specification, but in fact @code{n = n1+n2} is C
   code. Our primitive generation tool extracts a lot of information from
   the stack effect notations@footnote{We use a one-stack notation, even
   though we have separate data and floating-point stacks; The separate
   notation can be generated easily from the unified notation.}: The number
   of items popped from and pushed on the stack, their type, and by what
   name they are referred to in the C code. It then generates a C code
   prelude and postlude for each primitive. The final C code for @code{+}
   looks like this:
   
   @example
   I_plus: /* + ( n1 n2 -- n ) */  /* label, stack effect */
   /*  */                          /* documentation */
   @{
   DEF_CA                          /* definition of variable ca (indirect threading) */
   Cell n1;                        /* definitions of variables */
   Cell n2;
   Cell n;
   n1 = (Cell) sp[1];              /* input */
   n2 = (Cell) TOS;
   sp += 1;                        /* stack adjustment */
   NAME("+")                       /* debugging output (with -DDEBUG) */
   @{
   n = n1+n2;                      /* C code taken from the source */
   @}
   NEXT_P1;                        /* NEXT part 1 */
   TOS = (Cell)n;                  /* output */
   NEXT_P2;                        /* NEXT part 2 */
   @}
   @end example
   
   This looks long and inefficient, but the GNU C compiler optimizes quite
   well and produces optimal code for @code{+} on, e.g., the R3000 and the
   HP RISC machines: Defining the @code{n}s does not produce any code, and
   using them as intermediate storage also adds no cost.
   
   There are also other optimizations, that are not illustrated by this
   example: Assignments between simple variables are usually for free (copy
   propagation). If one of the stack items is not used by the primitive
   (e.g.  in @code{drop}), the compiler eliminates the load from the stack
   (dead code elimination). On the other hand, there are some things that
   the compiler does not do, therefore they are performed by
   @file{prims2x.fs}: The compiler does not optimize code away that stores
   a stack item to the place where it just came from (e.g., @code{over}).
   
   While programming a primitive is usually easy, there are a few cases
   where the programmer has to take the actions of the generator into
   account, most notably @code{?dup}, but also words that do not (always)
   fall through to NEXT.
   
   @node TOS Optimization, Produced code, Automatic Generation, Primitives
   @subsection TOS Optimization
   
   An important optimization for stack machine emulators, e.g., Forth
   engines, is keeping  one or more of the top stack items in
   registers.  If a word has the stack effect @var{in1}...@var{inx} @code{--}
   @var{out1}...@var{outy}, keeping the top @var{n} items in registers
   @itemize
   @item
   is better than keeping @var{n-1} items, if @var{x>=n} and @var{y>=n},
   due to fewer loads from and stores to the stack.
   @item is slower than keeping @var{n-1} items, if @var{x<>y} and @var{x<n} and
   @var{y<n}, due to additional moves between registers.
   @end itemize
   
   In particular, keeping one item in a register is never a disadvantage,
   if there are enough registers. Keeping two items in registers is a
   disadvantage for frequent words like @code{?branch}, constants,
   variables, literals and @code{i}. Therefore our generator only produces
   code that keeps zero or one items in registers. The generated C code
   covers both cases; the selection between these alternatives is made at
   C-compile time using the switch @code{-DUSE_TOS}. @code{TOS} in the C
   code for @code{+} is just a simple variable name in the one-item case,
   otherwise it is a macro that expands into @code{sp[0]}. Note that the
   GNU C compiler tries to keep simple variables like @code{TOS} in
   registers, and it usually succeeds, if there are enough registers.
   
   The primitive generator performs the TOS optimization for the
   floating-point stack, too (@code{-DUSE_FTOS}). For floating-point
   operations the benefit of this optimization is even larger:
   floating-point operations take quite long on most processors, but can be
   performed in parallel with other operations as long as their results are
   not used. If the FP-TOS is kept in a register, this works. If
   it is kept on the stack, i.e., in memory, the store into memory has to
   wait for the result of the floating-point operation, lengthening the
   execution time of the primitive considerably.
   
   The TOS optimization makes the automatic generation of primitives a
   bit more complicated. Just replacing all occurrences of @code{sp[0]} by
   @code{TOS} is not sufficient. There are some special cases to
   consider:
   @itemize
   @item In the case of @code{dup ( w -- w w )} the generator must not
   eliminate the store to the original location of the item on the stack,
   if the TOS optimization is turned on.
   @item Primitives with stack effects of the form @code{--}
   @var{out1}...@var{outy} must store the TOS to the stack at the start.
   Likewise, primitives with the stack effect @var{in1}...@var{inx} @code{--}
   must load the TOS from the stack at the end. But for the null stack
   effect @code{--} no stores or loads should be generated.
   @end itemize
   
   @node Produced code,  , TOS Optimization, Primitives
   @subsection Produced code
   
   To see what assembly code is produced for the primitives on your machine
   with your compiler and your flag settings, type @code{make engine.s} and
   look at the resulting file @file{engine.s}.
   
   @node System Architecture, Performance, Primitives, Internals
   @section System Architecture
   
   Our Forth system consists not only of primitives, but also of
   definitions written in Forth. Since the Forth compiler itself belongs
   to those definitions, it is not possible to start the system with the
   primitives and the Forth source alone. Therefore we provide the Forth
   code as an image file in nearly executable form. At the start of the
   system a C routine loads the image file into memory, sets up the
   memory (stacks etc.) according to information in the image file, and
   starts executing Forth code.
   
   The image file format is a compromise between the goals of making it
   easy to generate image files and making them portable. The easiest way
   to generate an image file is to just generate a memory dump. However,
   this kind of image file cannot be used on a different machine, or on
   the next version of the engine on the same machine, it even might not
   work with the same engine compiled by a different version of the C
   compiler. We would like to have as few versions of the image file as
   possible, because we do not want to distribute many versions of the
   same image file, and to make it easy for the users to use their image
   files on many machines. We currently need to create a different image
   file for machines with different cell sizes and different byte order
   (little- or big-endian)@footnote{We are considering adding information to the
   image file that enables the loader to change the byte order.}.
   
   Forth code that is going to end up in a portable image file has to
   comply to some restrictions: addresses have to be stored in memory with
   special words (@code{A!}, @code{A,}, etc.) in order to make the code
   relocatable. Cells, floats, etc., have to be stored at the natural
   alignment boundaries@footnote{E.g., store floats (8 bytes) at an address
   dividable by~8. This happens automatically in our system when you use
   the ANS Forth alignment words.}, in order to avoid alignment faults on
   machines with stricter alignment. The image file is produced by a
   metacompiler (@file{cross.fs}).
   
   So, unlike the image file of Mitch Bradleys @code{cforth}, our image
   file is not directly executable, but has to undergo some manipulations
   during loading. Address relocation is performed at image load-time, not
   at run-time. The loader also has to replace tokens standing for
   primitive calls with the appropriate code-field addresses (or code
   addresses in the case of direct threading).
   
   @node  Performance,  , System Architecture, Internals
   @section Performance
   
   On RISCs the Gforth engine is very close to optimal; i.e., it is usually
   impossible to write a significantly faster engine.
   
   On register-starved machines like the 386 architecture processors
   improvements are possible, because @code{gcc} does not utilize the
   registers as well as a human, even with explicit register declarations;
   e.g., Bernd Beuster wrote a Forth system fragment in assembly language
   and hand-tuned it for the 486; this system is 1.19 times faster on the
   Sieve benchmark on a 486DX2/66 than Gforth compiled with
   @code{gcc-2.6.3} with @code{-DFORCE_REG}.
   
   However, this potential advantage of assembly language implementations
   is not necessarily realized in complete Forth systems: We compared
   Gforth (compiled with @code{gcc-2.6.3} and @code{-DFORCE_REG}) with
   Win32Forth 1.2093 and LMI's NT Forth (Beta, May 1994), two systems
   written in assembly, and with two systems written in C: PFE-0.9.11
   (compiled with @code{gcc-2.6.3} with the default configuration for
   Linux: @code{-O2 -fomit-frame-pointer -DUSE_REGS}) and ThisForth Beta
   (compiled with gcc-2.6.3 -O3 -fomit-frame-pointer). We benchmarked
   Gforth, PFE and ThisForth on a 486DX2/66 under Linux. Kenneth O'Heskin
   kindly provided the results for Win32Forth and NT Forth on a 486DX2/66
   with similar memory performance under Windows NT.
    
   We used four small benchmarks: the ubiquitous Sieve; bubble-sorting and
   matrix multiplication come from the Stanford integer benchmarks and have
   been translated into Forth by Martin Fraeman; we used the versions
   included in the TILE Forth package; and a recursive Fibonacci number
   computation for benchmark calling performance. The following table shows
   the time taken for the benchmarks scaled by the time taken by Gforth (in
   other words, it shows the speedup factor that Gforth achieved over the
   other systems).
   
   @example
   relative             Win32-        NT               This-
     time     Gforth     Forth     Forth       PFE     Forth
   sieve        1.00      1.30      1.07      1.67      2.98
   bubble       1.00      1.30      1.40      1.66
   matmul       1.00      1.40      1.29      2.24
   fib          1.00      1.44      1.26      1.82      2.82
   @end example
   
   You may find the good performance of Gforth compared with the systems
   written in assembly language quite surprising. One important reason for
   the disappointing performance of these systems is probably that they are
   not written optimally for the 486 (e.g., they use the @code{lods}
   instruction). In addition, Win32Forth uses a comfortable, but costly
   method for relocating the Forth image: like @code{cforth}, it computes
   the actual addresses at run time, resulting in two address computations
   per NEXT (@pxref{System Architecture}).
   
   The speedup of Gforth over PFE and ThisForth can be easily explained
   with the self-imposed restriction to standard C (although the measured
   implementation of PFE uses a GNU C extension: global register
   variables), which makes efficient threading impossible.  Moreover,
   current C compilers have a hard time optimizing other aspects of the
   ThisForth source.
   
   Note that the performance of Gforth on 386 architecture processors
   varies widely with the version of @code{gcc} used. E.g., @code{gcc-2.5.8}
   failed to allocate any of the virtual machine registers into real
   machine registers by itself and would not work correctly with explicit
   register declarations, giving a 1.3 times slower engine (on a 486DX2/66
   running the Sieve) than the one measured above.
   
   @node Bugs, Pedigree, Internals, Top
   @chapter Bugs
   
   Known bugs are described in the file BUGS in the Gforth distribution.
   
   If you find a bug, please send a bug report to !!. A bug report should
   describe the Gforth version used (it is announced at the start of an
   interactive Gforth session), the machine and operating system (on Unix
   systems you can use @code{uname -a} to produce this information), the
   installation options (!! a way to find them out), and a complete list of
   changes you (or your installer) have made to the Gforth sources (if
   any); it should contain a program (or a sequence of keyboard commands)
   that reproduces the bug and a description of what you think constitutes
   the buggy behaviour.
   
   For a thorough guide on reporting bugs read @ref{Bug Reporting, , How
   to Report Bugs, gcc.info, GNU C Manual}.
   
   
   @node Pedigree, Word Index, Bugs, Top
   @chapter Pedigree
   
   Gforth descends from BigForth (1993) and fig-Forth. Gforth and PFE (by
   Dirk Zoller) will cross-fertilize each other. Of course, a significant part of the design of Gforth was prescribed by ANS Forth.
   
   Bernd Paysan wrote BigForth, a child of VolksForth.
   
   VolksForth descends from F83. !! Authors? When?
   
   Laxen and Perry wrote F83 as a model implementation of the
   Forth-83 standard. !! Pedigree? When?
   
   A team led by Bill Ragsdale implemented fig-Forth on many processors in
   1979. Dean Sanderson and Bill Ragsdale developed the original
   implementation of fig-Forth based on microForth.
   
   !! microForth pedigree
   
   A part of the information in this section comes from @cite{The Evolution
   of Forth} by Elizabeth D. Rather, Donald R. Colburn and Charles
   H. Moore, presented at the HOPL-II conference and preprinted in SIGPLAN
   Notices 28(3), 1993.  You can find more historical and genealogical
   information about Forth there.
   
   @node Word Index, Node Index, Pedigree, Top
   @chapter Word Index
   
   This index is as incomplete as the manual. Each word is listed with
   stack effect and wordset.
   
   @printindex fn
   
   @node Node Index,  , Word Index, Top
   @chapter Node Index
   
   This index is even less complete than the manual.
   
 @contents  @contents
 @bye  @bye

Removed from v.1.1  
changed lines
  Added in v.1.20


FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>