Annotation of gforth/gforth.ds, revision 1.15

1.1       anton       1: \input texinfo   @c -*-texinfo-*-
                      2: @comment The source is gforth.ds, from which gforth.texi is generated
                      3: @comment %**start of header (This is for running Texinfo on a region.)
1.4       anton       4: @setfilename gforth.info
1.1       anton       5: @settitle GNU Forth Manual
1.4       anton       6: @comment @setchapternewpage odd
1.1       anton       7: @comment %**end of header (This is for running Texinfo on a region.)
                      8: 
                      9: @ifinfo
                     10: This file documents GNU Forth 0.0
                     11: 
                     12: Copyright @copyright{} 1994 GNU Forth Development Group
                     13: 
                     14:      Permission is granted to make and distribute verbatim copies of
                     15:      this manual provided the copyright notice and this permission notice
                     16:      are preserved on all copies.
                     17:      
1.4       anton      18: @ignore
1.1       anton      19:      Permission is granted to process this file through TeX and print the
                     20:      results, provided the printed document carries a copying permission
                     21:      notice identical to this one except for the removal of this paragraph
                     22:      (this paragraph not being relevant to the printed manual).
                     23:      
1.4       anton      24: @end ignore
1.1       anton      25:      Permission is granted to copy and distribute modified versions of this
                     26:      manual under the conditions for verbatim copying, provided also that the
                     27:      sections entitled "Distribution" and "General Public License" are
                     28:      included exactly as in the original, and provided that the entire
                     29:      resulting derived work is distributed under the terms of a permission
                     30:      notice identical to this one.
                     31:      
                     32:      Permission is granted to copy and distribute translations of this manual
                     33:      into another language, under the above conditions for modified versions,
                     34:      except that the sections entitled "Distribution" and "General Public
                     35:      License" may be included in a translation approved by the author instead
                     36:      of in the original English.
                     37: @end ifinfo
                     38: 
                     39: @titlepage
                     40: @sp 10
                     41: @center @titlefont{GNU Forth Manual}
                     42: @sp 2
                     43: @center for version 0.0
                     44: @sp 2
                     45: @center Anton Ertl
                     46: 
                     47: @comment  The following two commands start the copyright page.
                     48: @page
                     49: @vskip 0pt plus 1filll
                     50: Copyright @copyright{} 1994 GNU Forth Development Group
                     51: 
                     52: @comment !! Published by ... or You can get a copy of this manual ...
                     53: 
                     54:      Permission is granted to make and distribute verbatim copies of
                     55:      this manual provided the copyright notice and this permission notice
                     56:      are preserved on all copies.
                     57:      
                     58:      Permission is granted to copy and distribute modified versions of this
                     59:      manual under the conditions for verbatim copying, provided also that the
                     60:      sections entitled "Distribution" and "General Public License" are
                     61:      included exactly as in the original, and provided that the entire
                     62:      resulting derived work is distributed under the terms of a permission
                     63:      notice identical to this one.
                     64:      
                     65:      Permission is granted to copy and distribute translations of this manual
                     66:      into another language, under the above conditions for modified versions,
                     67:      except that the sections entitled "Distribution" and "General Public
                     68:      License" may be included in a translation approved by the author instead
                     69:      of in the original English.
                     70: @end titlepage
                     71: 
                     72: 
                     73: @node Top, License, (dir), (dir)
                     74: @ifinfo
                     75: GNU Forth is a free implementation of ANS Forth available on many
                     76: personal machines. This manual corresponds to version 0.0.
                     77: @end ifinfo
                     78: 
                     79: @menu
1.4       anton      80: * License::                     
                     81: * Goals::                       About the GNU Forth Project
                     82: * Other Books::                 Things you might want to read
                     83: * Invocation::                  Starting GNU Forth
                     84: * Words::                       Forth words available in GNU Forth
                     85: * ANS conformance::             Implementation-defined options etc.
                     86: * Model::                       The abstract machine of GNU Forth
                     87: * Emacs and GForth::            The GForth Mode
                     88: * Internals::                   Implementation details
                     89: * Bugs::                        How to report them
                     90: * Pedigree::                    Ancestors of GNU Forth
                     91: * Word Index::                  An item for each Forth word
                     92: * Node Index::                  An item for each node
1.1       anton      93: @end menu
                     94: 
                     95: @node License, Goals, Top, Top
                     96: @unnumbered License
                     97: !! Insert GPL here
                     98: 
                     99: @iftex
                    100: @unnumbered Preface
                    101: This manual documents GNU Forth. The reader is expected to know
                    102: Forth. This manual is primarily a reference manual. @xref{Other Books}
                    103: for introductory material.
                    104: @end iftex
                    105: 
                    106: @node    Goals, Other Books, License, Top
                    107: @comment node-name,     next,           previous, up
                    108: @chapter Goals of GNU Forth
                    109: @cindex Goals
                    110: The goal of the GNU Forth Project is to develop a standard model for
                    111: ANSI Forth. This can be split into several subgoals:
                    112: 
                    113: @itemize @bullet
                    114: @item
                    115: GNU Forth should conform to the ANSI Forth standard.
                    116: @item
                    117: It should be a model, i.e. it should define all the
                    118: implementation-dependent things.
                    119: @item
                    120: It should become standard, i.e. widely accepted and used. This goal
                    121: is the most difficult one.
                    122: @end itemize
                    123: 
                    124: To achieve these goals GNU Forth should be
                    125: @itemize @bullet
                    126: @item
                    127: Similar to previous models (fig-Forth, F83)
                    128: @item
                    129: Powerful. It should provide for all the things that are considered
                    130: necessary today and even some that are not yet considered necessary.
                    131: @item
                    132: Efficient. It should not get the reputation of being exceptionally
                    133: slow.
                    134: @item
                    135: Free.
                    136: @item
                    137: Available on many machines/easy to port.
                    138: @end itemize
                    139: 
                    140: Have we achieved these goals? GNU Forth conforms to the ANS Forth
                    141: standard; it may be considered a model, but we have not yet documented
                    142: which parts of the model are stable and which parts we are likely to
                    143: change; it certainly has not yet become a de facto standard. It has some
                    144: similarities and some differences to previous models; It has some
                    145: powerful features, but not yet everything that we envisioned; on RISCs
                    146: it is as fast as interpreters programmed in assembly, on
                    147: register-starved machines it is not so fast, but still faster than any
                    148: other C-based interpretive implementation; it is free and available on
                    149: many machines.
                    150: 
                    151: @node Other Books, Invocation, Goals, Top
                    152: @chapter Other books on ANS Forth
                    153: 
                    154: As the standard is relatively new, there are not many books out yet. It
                    155: is not recommended to learn Forth by using GNU Forth and a book that is
                    156: not written for ANS Forth, as you will not know your mistakes from the
                    157: deviations of the book.
                    158: 
                    159: There is, of course, the standard, the definite reference if you want to
                    160: write ANS Forth programs. It will be available in printed form from
                    161: Global Engineering Documents !! somtime in spring or summer 1994. If you
                    162: are lucky, you can still get dpANS6 (the draft that was approved as
                    163: standard) by aftp from ftp.uu.net:/vendor/minerva/x3j14.
                    164: 
                    165: @cite{Forth: The new model} by Jack Woehr (!! Publisher) is an
                    166: introductory book based on a draft version of the standard. It does not
                    167: cover the whole standard. It also contains interesting background
                    168: information (Jack Woehr was in the ANS Forth Technical Committe). It is
                    169: not appropriate for complete newbies, but programmers experienced in
                    170: other languages should find it ok.
                    171: 
                    172: @node Invocation, Words, Other Books, Top
                    173: @chapter Invocation
                    174: 
                    175: You will usually just say @code{gforth}. In many other cases the default
                    176: GNU Forth image will be invoked like this:
                    177: 
                    178: @example
                    179: gforth [files] [-e forth-code]
                    180: @end example
                    181: 
                    182: executing the contents of the files and the Forth code in the order they
                    183: are given.
                    184: 
                    185: In general, the command line looks like this:
                    186: 
                    187: @example
                    188: gforth [initialization options] [image-specific options]
                    189: @end example
                    190: 
                    191: The initialization options must come before the rest of the command
                    192: line. They are:
                    193: 
                    194: @table @code
                    195: @item --image-file @var{file}
                    196: Loads the Forth image @var{file} instead of the default
                    197: @file{gforth.fi}.
                    198: 
                    199: @item --path @var{path}
                    200: Uses @var{path} for searching the image file and Forth source code
                    201: files instead of the default in the environment variable
                    202: @code{GFORTHPATH} or the path specified at installation time (typically
                    203: @file{/usr/local/lib/gforth:.}). A path is given as a @code{:}-separated
                    204: list.
                    205: 
                    206: @item --dictionary-size @var{size}
                    207: @item -m @var{size}
                    208: Allocate @var{size} space for the Forth dictionary space instead of
                    209: using the default specified in the image (typically 256K). The
                    210: @var{size} specification consists of an integer and a unit (e.g.,
                    211: @code{4M}). The unit can be one of @code{b} (bytes), @code{e} (element
                    212: size, in this case Cells), @code{k} (kilobytes), and @code{M}
                    213: (Megabytes). If no unit is specified, @code{e} is used.
                    214: 
                    215: @item --data-stack-size @var{size}
                    216: @item -d @var{size}
                    217: Allocate @var{size} space for the data stack instead of using the
                    218: default specified in the image (typically 16K).
                    219: 
                    220: @item --return-stack-size @var{size}
                    221: @item -r @var{size}
                    222: Allocate @var{size} space for the return stack instead of using the
                    223: default specified in the image (typically 16K).
                    224: 
                    225: @item --fp-stack-size @var{size}
                    226: @item -f @var{size}
                    227: Allocate @var{size} space for the floating point stack instead of
                    228: using the default specified in the image (typically 16K). In this case
                    229: the unit specifier @code{e} refers to floating point numbers.
                    230: 
                    231: @item --locals-stack-size @var{size}
                    232: @item -l @var{size}
                    233: Allocate @var{size} space for the locals stack instead of using the
                    234: default specified in the image (typically 16K).
                    235: 
                    236: @end table
                    237: 
                    238: As explained above, the image-specific command-line arguments for the
                    239: default image @file{gforth.fi} consist of a sequence of filenames and
                    240: @code{-e @var{forth-code}} options that are interpreted in the seqence
                    241: in which they are given. The @code{-e @var{forth-code}} or
                    242: @code{--evaluate @var{forth-code}} option evaluates the forth
                    243: code. This option takes only one argument; if you want to evaluate more
                    244: Forth words, you have to quote them or use several @code{-e}s. To exit
                    245: after processing the command line (instead of entering interactive mode)
                    246: append @code{-e bye} to the command line.
                    247: 
                    248: Not yet implemented:
                    249: On startup the system first executes the system initialization file
                    250: (unless the option @code{--no-init-file} is given; note that the system
                    251: resulting from using this option may not be ANS Forth conformant). Then
                    252: the user initialization file @file{.gforth.fs} is executed, unless the
                    253: option @code{--no-rc} is given; this file is first searched in @file{.},
                    254: then in @file{~}, then in the normal path (see above).
                    255: 
1.4       anton     256: @node Words, ANS conformance, Invocation, Top
1.1       anton     257: @chapter Forth Words
                    258: 
                    259: @menu
1.4       anton     260: * Notation::                    
                    261: * Arithmetic::                  
                    262: * Stack Manipulation::          
                    263: * Memory access::               
                    264: * Control Structures::          
                    265: * Locals::                      
                    266: * Defining Words::              
                    267: * Wordlists::                   
                    268: * Files::                       
                    269: * Blocks::                      
                    270: * Other I/O::                   
                    271: * Programming Tools::           
                    272: * Threading Words::             
1.1       anton     273: @end menu
                    274: 
                    275: @node Notation, Arithmetic, Words, Words
                    276: @section Notation
                    277: 
                    278: The Forth words are described in this section in the glossary notation
                    279: that has become a de-facto standard for Forth texts, i.e.
                    280: 
1.4       anton     281: @format
1.1       anton     282: @var{word}     @var{Stack effect}   @var{wordset}   @var{pronunciation}
1.4       anton     283: @end format
1.1       anton     284: @var{Description}
                    285: 
                    286: @table @var
                    287: @item word
                    288: The name of the word. BTW, GNU Forth is case insensitive, so you can
1.14      anton     289: type the words in in lower case (However, @pxref{core-idef}).
1.1       anton     290: 
                    291: @item Stack effect
                    292: The stack effect is written in the notation @code{@var{before} --
                    293: @var{after}}, where @var{before} and @var{after} describe the top of
                    294: stack entries before and after the execution of the word. The rest of
                    295: the stack is not touched by the word. The top of stack is rightmost,
                    296: i.e., a stack sequence is written as it is typed in. Note that GNU Forth
                    297: uses a separate floating point stack, but a unified stack
                    298: notation. Also, return stack effects are not shown in @var{stack
                    299: effect}, but in @var{Description}. The name of a stack item describes
                    300: the type and/or the function of the item. See below for a discussion of
                    301: the types.
                    302: 
                    303: @item pronunciation
                    304: How the word is pronounced
                    305: 
                    306: @item wordset
                    307: The ANS Forth standard is divided into several wordsets. A standard
                    308: system need not support all of them. So, the fewer wordsets your program
                    309: uses the more portable it will be in theory. However, we suspect that
                    310: most ANS Forth systems on personal machines will feature all
                    311: wordsets. Words that are not defined in the ANS standard have
                    312: @code{gforth} as wordset.
                    313: 
                    314: @item Description
                    315: A description of the behaviour of the word.
                    316: @end table
                    317: 
1.4       anton     318: The type of a stack item is specified by the character(s) the name
                    319: starts with:
1.1       anton     320: 
                    321: @table @code
                    322: @item f
                    323: Bool, i.e. @code{false} or @code{true}.
                    324: @item c
                    325: Char
                    326: @item w
                    327: Cell, can contain an integer or an address
                    328: @item n
                    329: signed integer
                    330: @item u
                    331: unsigned integer
                    332: @item d
                    333: double sized signed integer
                    334: @item ud
                    335: double sized unsigned integer
                    336: @item r
                    337: Float
                    338: @item a_
                    339: Cell-aligned address
                    340: @item c_
                    341: Char-aligned address (note that a Char is two bytes in Windows NT)
                    342: @item f_
                    343: Float-aligned address
                    344: @item df_
                    345: Address aligned for IEEE double precision float
                    346: @item sf_
                    347: Address aligned for IEEE single precision float
                    348: @item xt
                    349: Execution token, same size as Cell
                    350: @item wid
                    351: Wordlist ID, same size as Cell
                    352: @item f83name
                    353: Pointer to a name structure
                    354: @end table
                    355: 
1.4       anton     356: @node Arithmetic, Stack Manipulation, Notation, Words
1.1       anton     357: @section Arithmetic
                    358: Forth arithmetic is not checked, i.e., you will not hear about integer
                    359: overflow on addition or multiplication, you may hear about division by
                    360: zero if you are lucky. The operator is written after the operands, but
                    361: the operands are still in the original order. I.e., the infix @code{2-1}
                    362: corresponds to @code{2 1 -}. Forth offers a variety of division
                    363: operators. If you perform division with potentially negative operands,
                    364: you do not want to use @code{/} or @code{/mod} with its undefined
                    365: behaviour, but rather @code{fm/mod} or @code{sm/mod} (probably the
1.4       anton     366: former, @pxref{Mixed precision}).
                    367: 
                    368: @menu
                    369: * Single precision::            
                    370: * Bitwise operations::          
                    371: * Mixed precision::             operations with single and double-cell integers
                    372: * Double precision::            Double-cell integer arithmetic
                    373: * Floating Point::              
                    374: @end menu
1.1       anton     375: 
1.4       anton     376: @node Single precision, Bitwise operations, Arithmetic, Arithmetic
1.1       anton     377: @subsection Single precision
                    378: doc-+
                    379: doc--
                    380: doc-*
                    381: doc-/
                    382: doc-mod
                    383: doc-/mod
                    384: doc-negate
                    385: doc-abs
                    386: doc-min
                    387: doc-max
                    388: 
1.4       anton     389: @node Bitwise operations, Mixed precision, Single precision, Arithmetic
1.1       anton     390: @subsection Bitwise operations
                    391: doc-and
                    392: doc-or
                    393: doc-xor
                    394: doc-invert
                    395: doc-2*
                    396: doc-2/
                    397: 
1.4       anton     398: @node Mixed precision, Double precision, Bitwise operations, Arithmetic
1.1       anton     399: @subsection Mixed precision
                    400: doc-m+
                    401: doc-*/
                    402: doc-*/mod
                    403: doc-m*
                    404: doc-um*
                    405: doc-m*/
                    406: doc-um/mod
                    407: doc-fm/mod
                    408: doc-sm/rem
                    409: 
1.4       anton     410: @node Double precision, Floating Point, Mixed precision, Arithmetic
1.1       anton     411: @subsection Double precision
                    412: doc-d+
                    413: doc-d-
                    414: doc-dnegate
                    415: doc-dabs
                    416: doc-dmin
                    417: doc-dmax
                    418: 
1.4       anton     419: @node Floating Point,  , Double precision, Arithmetic
                    420: @subsection Floating Point
                    421: 
                    422: Angles in floating point operations are given in radians (a full circle
                    423: has 2 pi radians). Note, that gforth has a separate floating point
                    424: stack, but we use the unified notation.
                    425: 
                    426: Floating point numbers have a number of unpleasant surprises for the
                    427: unwary (e.g., floating point addition is not associative) and even a few
                    428: for the wary. You should not use them unless you know what you are doing
                    429: or you don't care that the results you get are totally bogus. If you
                    430: want to learn about the problems of floating point numbers (and how to
1.11      anton     431: avoid them), you might start with @cite{David Goldberg, What Every
1.6       anton     432: Computer Scientist Should Know About Floating-Point Arithmetic, ACM
                    433: Computing Surveys 23(1):5@minus{}48, March 1991}.
1.4       anton     434: 
                    435: doc-f+
                    436: doc-f-
                    437: doc-f*
                    438: doc-f/
                    439: doc-fnegate
                    440: doc-fabs
                    441: doc-fmax
                    442: doc-fmin
                    443: doc-floor
                    444: doc-fround
                    445: doc-f**
                    446: doc-fsqrt
                    447: doc-fexp
                    448: doc-fexpm1
                    449: doc-fln
                    450: doc-flnp1
                    451: doc-flog
1.6       anton     452: doc-falog
1.4       anton     453: doc-fsin
                    454: doc-fcos
                    455: doc-fsincos
                    456: doc-ftan
                    457: doc-fasin
                    458: doc-facos
                    459: doc-fatan
                    460: doc-fatan2
                    461: doc-fsinh
                    462: doc-fcosh
                    463: doc-ftanh
                    464: doc-fasinh
                    465: doc-facosh
                    466: doc-fatanh
                    467: 
                    468: @node Stack Manipulation, Memory access, Arithmetic, Words
1.1       anton     469: @section Stack Manipulation
                    470: 
                    471: gforth has a data stack (aka parameter stack) for characters, cells,
                    472: addresses, and double cells, a floating point stack for floating point
                    473: numbers, a return stack for storing the return addresses of colon
                    474: definitions and other data, and a locals stack for storing local
                    475: variables. Note that while every sane Forth has a separate floating
                    476: point stack, this is not strictly required; an ANS Forth system could
                    477: theoretically keep floating point numbers on the data stack. As an
                    478: additional difficulty, you don't know how many cells a floating point
                    479: number takes. It is reportedly possible to write words in a way that
                    480: they work also for a unified stack model, but we do not recommend trying
1.4       anton     481: it. Instead, just say that your program has an environmental dependency
                    482: on a separate FP stack.
                    483: 
                    484: Also, a Forth system is allowed to keep the local variables on the
1.1       anton     485: return stack. This is reasonable, as local variables usually eliminate
                    486: the need to use the return stack explicitly. So, if you want to produce
                    487: a standard complying program and if you are using local variables in a
                    488: word, forget about return stack manipulations in that word (see the
                    489: standard document for the exact rules).
                    490: 
1.4       anton     491: @menu
                    492: * Data stack::                  
                    493: * Floating point stack::        
                    494: * Return stack::                
                    495: * Locals stack::                
                    496: * Stack pointer manipulation::  
                    497: @end menu
                    498: 
                    499: @node Data stack, Floating point stack, Stack Manipulation, Stack Manipulation
1.1       anton     500: @subsection Data stack
                    501: doc-drop
                    502: doc-nip
                    503: doc-dup
                    504: doc-over
                    505: doc-tuck
                    506: doc-swap
                    507: doc-rot
                    508: doc--rot
                    509: doc-?dup
                    510: doc-pick
                    511: doc-roll
                    512: doc-2drop
                    513: doc-2nip
                    514: doc-2dup
                    515: doc-2over
                    516: doc-2tuck
                    517: doc-2swap
                    518: doc-2rot
                    519: 
1.4       anton     520: @node Floating point stack, Return stack, Data stack, Stack Manipulation
1.1       anton     521: @subsection Floating point stack
                    522: doc-fdrop
                    523: doc-fnip
                    524: doc-fdup
                    525: doc-fover
                    526: doc-ftuck
                    527: doc-fswap
                    528: doc-frot
                    529: 
1.4       anton     530: @node Return stack, Locals stack, Floating point stack, Stack Manipulation
1.1       anton     531: @subsection Return stack
                    532: doc->r
                    533: doc-r>
                    534: doc-r@
                    535: doc-rdrop
                    536: doc-2>r
                    537: doc-2r>
                    538: doc-2r@
                    539: doc-2rdrop
                    540: 
1.4       anton     541: @node Locals stack, Stack pointer manipulation, Return stack, Stack Manipulation
1.1       anton     542: @subsection Locals stack
                    543: 
1.4       anton     544: @node Stack pointer manipulation,  , Locals stack, Stack Manipulation
1.1       anton     545: @subsection Stack pointer manipulation
                    546: doc-sp@
                    547: doc-sp!
                    548: doc-fp@
                    549: doc-fp!
                    550: doc-rp@
                    551: doc-rp!
                    552: doc-lp@
                    553: doc-lp!
                    554: 
1.4       anton     555: @node Memory access, Control Structures, Stack Manipulation, Words
1.1       anton     556: @section Memory access
                    557: 
1.4       anton     558: @menu
                    559: * Stack-Memory transfers::      
                    560: * Address arithmetic::          
                    561: * Memory block access::         
                    562: @end menu
                    563: 
                    564: @node Stack-Memory transfers, Address arithmetic, Memory access, Memory access
1.1       anton     565: @subsection Stack-Memory transfers
                    566: 
                    567: doc-@
                    568: doc-!
                    569: doc-+!
                    570: doc-c@
                    571: doc-c!
                    572: doc-2@
                    573: doc-2!
                    574: doc-f@
                    575: doc-f!
                    576: doc-sf@
                    577: doc-sf!
                    578: doc-df@
                    579: doc-df!
                    580: 
1.4       anton     581: @node Address arithmetic, Memory block access, Stack-Memory transfers, Memory access
1.1       anton     582: @subsection Address arithmetic
                    583: 
                    584: ANS Forth does not specify the sizes of the data types. Instead, it
                    585: offers a number of words for computing sizes and doing address
                    586: arithmetic. Basically, address arithmetic is performed in terms of
                    587: address units (aus); on most systems the address unit is one byte. Note
                    588: that a character may have more than one au, so @code{chars} is no noop
                    589: (on systems where it is a noop, it compiles to nothing).
                    590: 
                    591: ANS Forth also defines words for aligning addresses for specific
                    592: addresses. Many computers require that accesses to specific data types
                    593: must only occur at specific addresses; e.g., that cells may only be
                    594: accessed at addresses divisible by 4. Even if a machine allows unaligned
                    595: accesses, it can usually perform aligned accesses faster. 
                    596: 
                    597: For the performance-concious: alignment operations are usually only
                    598: necessary during the definition of a data structure, not during the
                    599: (more frequent) accesses to it.
                    600: 
                    601: ANS Forth defines no words for character-aligning addresses. This is not
                    602: an oversight, but reflects the fact that addresses that are not
                    603: char-aligned have no use in the standard and therefore will not be
                    604: created.
                    605: 
                    606: The standard guarantees that addresses returned by @code{CREATE}d words
                    607: are cell-aligned; in addition, gforth guarantees that these addresses
                    608: are aligned for all purposes.
                    609: 
1.9       anton     610: Note that the standard defines a word @code{char}, which has nothing to
                    611: do with address arithmetic.
                    612: 
1.1       anton     613: doc-chars
                    614: doc-char+
                    615: doc-cells
                    616: doc-cell+
                    617: doc-align
                    618: doc-aligned
                    619: doc-floats
                    620: doc-float+
                    621: doc-falign
                    622: doc-faligned
                    623: doc-sfloats
                    624: doc-sfloat+
                    625: doc-sfalign
                    626: doc-sfaligned
                    627: doc-dfloats
                    628: doc-dfloat+
                    629: doc-dfalign
                    630: doc-dfaligned
1.10      anton     631: doc-maxalign
                    632: doc-maxaligned
                    633: doc-cfalign
                    634: doc-cfaligned
1.1       anton     635: doc-address-unit-bits
                    636: 
1.4       anton     637: @node Memory block access,  , Address arithmetic, Memory access
1.1       anton     638: @subsection Memory block access
                    639: 
                    640: doc-move
                    641: doc-erase
                    642: 
                    643: While the previous words work on address units, the rest works on
                    644: characters.
                    645: 
                    646: doc-cmove
                    647: doc-cmove>
                    648: doc-fill
                    649: doc-blank
                    650: 
1.4       anton     651: @node Control Structures, Locals, Memory access, Words
1.1       anton     652: @section Control Structures
                    653: 
                    654: Control structures in Forth cannot be used in interpret state, only in
                    655: compile state, i.e., in a colon definition. We do not like this
                    656: limitation, but have not seen a satisfying way around it yet, although
                    657: many schemes have been proposed.
                    658: 
1.4       anton     659: @menu
                    660: * Selection::                   
                    661: * Simple Loops::                
                    662: * Counted Loops::               
                    663: * Arbitrary control structures::  
                    664: * Calls and returns::           
                    665: * Exception Handling::          
                    666: @end menu
                    667: 
                    668: @node Selection, Simple Loops, Control Structures, Control Structures
1.1       anton     669: @subsection Selection
                    670: 
                    671: @example
                    672: @var{flag}
                    673: IF
                    674:   @var{code}
                    675: ENDIF
                    676: @end example
                    677: or
                    678: @example
                    679: @var{flag}
                    680: IF
                    681:   @var{code1}
                    682: ELSE
                    683:   @var{code2}
                    684: ENDIF
                    685: @end example
                    686: 
1.4       anton     687: You can use @code{THEN} instead of @code{ENDIF}. Indeed, @code{THEN} is
1.1       anton     688: standard, and @code{ENDIF} is not, although it is quite popular. We
                    689: recommend using @code{ENDIF}, because it is less confusing for people
                    690: who also know other languages (and is not prone to reinforcing negative
                    691: prejudices against Forth in these people). Adding @code{ENDIF} to a
                    692: system that only supplies @code{THEN} is simple:
                    693: @example
                    694: : endif   POSTPONE then ; immediate
                    695: @end example
                    696: 
                    697: [According to @cite{Webster's New Encyclopedic Dictionary}, @dfn{then
                    698: (adv.)}  has the following meanings:
                    699: @quotation
                    700: ... 2b: following next after in order ... 3d: as a necessary consequence
                    701: (if you were there, then you saw them).
                    702: @end quotation
                    703: Forth's @code{THEN} has the meaning 2b, whereas @code{THEN} in Pascal
                    704: and many other programming languages has the meaning 3d.]
                    705: 
                    706: We also provide the words @code{?dup-if} and @code{?dup-0=-if}, so you
                    707: can avoid using @code{?dup}.
                    708: 
                    709: @example
                    710: @var{n}
                    711: CASE
                    712:   @var{n1} OF @var{code1} ENDOF
                    713:   @var{n2} OF @var{code2} ENDOF
1.4       anton     714:   @dots{}
1.1       anton     715: ENDCASE
                    716: @end example
                    717: 
                    718: Executes the first @var{codei}, where the @var{ni} is equal to
                    719: @var{n}. A default case can be added by simply writing the code after
                    720: the last @code{ENDOF}. It may use @var{n}, which is on top of the stack,
                    721: but must not consume it.
                    722: 
1.4       anton     723: @node Simple Loops, Counted Loops, Selection, Control Structures
1.1       anton     724: @subsection Simple Loops
                    725: 
                    726: @example
                    727: BEGIN
                    728:   @var{code1}
                    729:   @var{flag}
                    730: WHILE
                    731:   @var{code2}
                    732: REPEAT
                    733: @end example
                    734: 
                    735: @var{code1} is executed and @var{flag} is computed. If it is true,
                    736: @var{code2} is executed and the loop is restarted; If @var{flag} is false, execution continues after the @code{REPEAT}.
                    737: 
                    738: @example
                    739: BEGIN
                    740:   @var{code}
                    741:   @var{flag}
                    742: UNTIL
                    743: @end example
                    744: 
                    745: @var{code} is executed. The loop is restarted if @code{flag} is false.
                    746: 
                    747: @example
                    748: BEGIN
                    749:   @var{code}
                    750: AGAIN
                    751: @end example
                    752: 
                    753: This is an endless loop.
                    754: 
1.4       anton     755: @node Counted Loops, Arbitrary control structures, Simple Loops, Control Structures
1.1       anton     756: @subsection Counted Loops
                    757: 
                    758: The basic counted loop is:
                    759: @example
                    760: @var{limit} @var{start}
                    761: ?DO
                    762:   @var{body}
                    763: LOOP
                    764: @end example
                    765: 
                    766: This performs one iteration for every integer, starting from @var{start}
                    767: and up to, but excluding @var{limit}. The counter, aka index, can be
                    768: accessed with @code{i}. E.g., the loop
                    769: @example
                    770: 10 0 ?DO
                    771:   i .
                    772: LOOP
                    773: @end example
                    774: prints
                    775: @example
                    776: 0 1 2 3 4 5 6 7 8 9
                    777: @end example
                    778: The index of the innermost loop can be accessed with @code{i}, the index
                    779: of the next loop with @code{j}, and the index of the third loop with
                    780: @code{k}.
                    781: 
                    782: The loop control data are kept on the return stack, so there are some
                    783: restrictions on mixing return stack accesses and counted loop
                    784: words. E.g., if you put values on the return stack outside the loop, you
                    785: cannot read them inside the loop. If you put values on the return stack
                    786: within a loop, you have to remove them before the end of the loop and
                    787: before accessing the index of the loop.
                    788: 
                    789: There are several variations on the counted loop:
                    790: 
                    791: @code{LEAVE} leaves the innermost counted loop immediately.
                    792: 
                    793: @code{LOOP} can be replaced with @code{@var{n} +LOOP}; this updates the
                    794: index by @var{n} instead of by 1. The loop is terminated when the border
                    795: between @var{limit-1} and @var{limit} is crossed. E.g.:
                    796: 
1.2       anton     797: @code{4 0 ?DO  i .  2 +LOOP}   prints @code{0 2}
1.1       anton     798: 
1.2       anton     799: @code{4 1 ?DO  i .  2 +LOOP}   prints @code{1 3}
1.1       anton     800: 
                    801: The behaviour of @code{@var{n} +LOOP} is peculiar when @var{n} is negative:
                    802: 
1.2       anton     803: @code{-1 0 ?DO  i .  -1 +LOOP}  prints @code{0 -1}
1.1       anton     804: 
1.2       anton     805: @code{ 0 0 ?DO  i .  -1 +LOOP}  prints nothing
1.1       anton     806: 
                    807: Therefore we recommend avoiding using @code{@var{n} +LOOP} with negative
                    808: @var{n}. One alternative is @code{@var{n} S+LOOP}, where the negative
                    809: case behaves symmetrical to the positive case:
                    810: 
1.7       pazsan    811: @code{-2 0 ?DO  i .  -1 S+LOOP}  prints @code{0 -1}
1.1       anton     812: 
1.7       pazsan    813: @code{-1 0 ?DO  i .  -1 S+LOOP}  prints @code{0}
1.1       anton     814: 
1.7       pazsan    815: @code{ 0 0 ?DO  i .  -1 S+LOOP}  prints nothing
1.1       anton     816: 
1.2       anton     817: The loop is terminated when the border between @var{limit@minus{}sgn(n)} and
1.1       anton     818: @var{limit} is crossed. However, @code{S+LOOP} is not part of the ANS
                    819: Forth standard.
                    820: 
                    821: @code{?DO} can be replaced by @code{DO}. @code{DO} enters the loop even
                    822: when the start and the limit value are equal. We do not recommend using
                    823: @code{DO}. It will just give you maintenance troubles.
                    824: 
                    825: @code{UNLOOP} is used to prepare for an abnormal loop exit, e.g., via
                    826: @code{EXIT}. @code{UNLOOP} removes the loop control parameters from the
                    827: return stack so @code{EXIT} can get to its return address.
                    828: 
                    829: Another counted loop is
                    830: @example
                    831: @var{n}
                    832: FOR
                    833:   @var{body}
                    834: NEXT
                    835: @end example
                    836: This is the preferred loop of native code compiler writers who are too
                    837: lazy to optimize @code{?DO} loops properly. In GNU Forth, this loop
                    838: iterates @var{n+1} times; @code{i} produces values starting with @var{n}
                    839: and ending with 0. Other Forth systems may behave differently, even if
                    840: they support @code{FOR} loops.
                    841: 
1.4       anton     842: @node Arbitrary control structures, Calls and returns, Counted Loops, Control Structures
1.2       anton     843: @subsection Arbitrary control structures
                    844: 
                    845: ANS Forth permits and supports using control structures in a non-nested
                    846: way. Information about incomplete control structures is stored on the
                    847: control-flow stack. This stack may be implemented on the Forth data
                    848: stack, and this is what we have done in gforth.
                    849: 
                    850: An @i{orig} entry represents an unresolved forward branch, a @i{dest}
                    851: entry represents a backward branch target. A few words are the basis for
                    852: building any control structure possible (except control structures that
                    853: need storage, like calls, coroutines, and backtracking).
                    854: 
1.3       anton     855: doc-if
                    856: doc-ahead
                    857: doc-then
                    858: doc-begin
                    859: doc-until
                    860: doc-again
                    861: doc-cs-pick
                    862: doc-cs-roll
1.2       anton     863: 
                    864: On many systems control-flow stack items take one word, in gforth they
                    865: currently take three (this may change in the future). Therefore it is a
                    866: really good idea to manipulate the control flow stack with
                    867: @code{cs-pick} and @code{cs-roll}, not with data stack manipulation
                    868: words.
                    869: 
                    870: Some standard control structure words are built from these words:
                    871: 
1.3       anton     872: doc-else
                    873: doc-while
                    874: doc-repeat
1.2       anton     875: 
                    876: Counted loop words constitute a separate group of words:
                    877: 
1.3       anton     878: doc-?do
                    879: doc-do
                    880: doc-for
                    881: doc-loop
                    882: doc-s+loop
                    883: doc-+loop
                    884: doc-next
                    885: doc-leave
                    886: doc-?leave
                    887: doc-unloop
1.10      anton     888: doc-done
1.2       anton     889: 
                    890: The standard does not allow using @code{cs-pick} and @code{cs-roll} on
                    891: @i{do-sys}. Our system allows it, but it's your job to ensure that for
                    892: every @code{?DO} etc. there is exactly one @code{UNLOOP} on any path
1.3       anton     893: through the definition (@code{LOOP} etc. compile an @code{UNLOOP} on the
                    894: fall-through path). Also, you have to ensure that all @code{LEAVE}s are
1.7       pazsan    895: resolved (by using one of the loop-ending words or @code{DONE}).
1.2       anton     896: 
                    897: Another group of control structure words are
                    898: 
1.3       anton     899: doc-case
                    900: doc-endcase
                    901: doc-of
                    902: doc-endof
1.2       anton     903: 
                    904: @i{case-sys} and @i{of-sys} cannot be processed using @code{cs-pick} and
                    905: @code{cs-roll}.
                    906: 
1.3       anton     907: @subsubsection Programming Style
                    908: 
                    909: In order to ensure readability we recommend that you do not create
                    910: arbitrary control structures directly, but define new control structure
                    911: words for the control structure you want and use these words in your
                    912: program.
                    913: 
                    914: E.g., instead of writing
                    915: 
                    916: @example
                    917: begin
                    918:   ...
                    919: if [ 1 cs-roll ]
                    920:   ...
                    921: again then
                    922: @end example
                    923: 
                    924: we recommend defining control structure words, e.g.,
                    925: 
                    926: @example
                    927: : while ( dest -- orig dest )
                    928:  POSTPONE if
                    929:  1 cs-roll ; immediate
                    930: 
                    931: : repeat ( orig dest -- )
                    932:  POSTPONE again
                    933:  POSTPONE then ; immediate
                    934: @end example
                    935: 
                    936: and then using these to create the control structure:
                    937: 
                    938: @example
                    939: begin
                    940:   ...
                    941: while
                    942:   ...
                    943: repeat
                    944: @end example
                    945: 
                    946: That's much easier to read, isn't it? Of course, @code{BEGIN} and
                    947: @code{WHILE} are predefined, so in this example it would not be
                    948: necessary to define them.
                    949: 
1.4       anton     950: @node Calls and returns, Exception Handling, Arbitrary control structures, Control Structures
1.3       anton     951: @subsection Calls and returns
                    952: 
                    953: A definition can be called simply be writing the name of the
                    954: definition. When the end of the definition is reached, it returns. An earlier return can be forced using
                    955: 
                    956: doc-exit
                    957: 
                    958: Don't forget to clean up the return stack and @code{UNLOOP} any
                    959: outstanding @code{?DO}...@code{LOOP}s before @code{EXIT}ing. The
                    960: primitive compiled by @code{EXIT} is
                    961: 
                    962: doc-;s
                    963: 
1.4       anton     964: @node Exception Handling,  , Calls and returns, Control Structures
1.3       anton     965: @subsection Exception Handling
                    966: 
                    967: doc-catch
                    968: doc-throw
                    969: 
1.4       anton     970: @node Locals, Defining Words, Control Structures, Words
1.1       anton     971: @section Locals
                    972: 
1.2       anton     973: Local variables can make Forth programming more enjoyable and Forth
                    974: programs easier to read. Unfortunately, the locals of ANS Forth are
                    975: laden with restrictions. Therefore, we provide not only the ANS Forth
                    976: locals wordset, but also our own, more powerful locals wordset (we
                    977: implemented the ANS Forth locals wordset through our locals wordset).
                    978: 
                    979: @menu
1.4       anton     980: * gforth locals::               
                    981: * ANS Forth locals::            
1.2       anton     982: @end menu
                    983: 
1.4       anton     984: @node gforth locals, ANS Forth locals, Locals, Locals
1.2       anton     985: @subsection gforth locals
                    986: 
                    987: Locals can be defined with
                    988: 
                    989: @example
                    990: @{ local1 local2 ... -- comment @}
                    991: @end example
                    992: or
                    993: @example
                    994: @{ local1 local2 ... @}
                    995: @end example
                    996: 
                    997: E.g.,
                    998: @example
                    999: : max @{ n1 n2 -- n3 @}
                   1000:  n1 n2 > if
                   1001:    n1
                   1002:  else
                   1003:    n2
                   1004:  endif ;
                   1005: @end example
                   1006: 
                   1007: The similarity of locals definitions with stack comments is intended. A
                   1008: locals definition often replaces the stack comment of a word. The order
                   1009: of the locals corresponds to the order in a stack comment and everything
                   1010: after the @code{--} is really a comment.
                   1011: 
                   1012: This similarity has one disadvantage: It is too easy to confuse locals
                   1013: declarations with stack comments, causing bugs and making them hard to
                   1014: find. However, this problem can be avoided by appropriate coding
                   1015: conventions: Do not use both notations in the same program. If you do,
                   1016: they should be distinguished using additional means, e.g. by position.
                   1017: 
                   1018: The name of the local may be preceded by a type specifier, e.g.,
                   1019: @code{F:} for a floating point value:
                   1020: 
                   1021: @example
                   1022: : CX* @{ F: Ar F: Ai F: Br F: Bi -- Cr Ci @}
                   1023: \ complex multiplication
                   1024:  Ar Br f* Ai Bi f* f-
                   1025:  Ar Bi f* Ai Br f* f+ ;
                   1026: @end example
                   1027: 
                   1028: GNU Forth currently supports cells (@code{W:}, @code{W^}), doubles
                   1029: (@code{D:}, @code{D^}), floats (@code{F:}, @code{F^}) and characters
                   1030: (@code{C:}, @code{C^}) in two flavours: a value-flavoured local (defined
                   1031: with @code{W:}, @code{D:} etc.) produces its value and can be changed
                   1032: with @code{TO}. A variable-flavoured local (defined with @code{W^} etc.)
                   1033: produces its address (which becomes invalid when the variable's scope is
                   1034: left). E.g., the standard word @code{emit} can be defined in therms of
                   1035: @code{type} like this:
                   1036: 
                   1037: @example
                   1038: : emit @{ C^ char* -- @}
                   1039:     char* 1 type ;
                   1040: @end example
                   1041: 
                   1042: A local without type specifier is a @code{W:} local. Both flavours of
                   1043: locals are initialized with values from the data or FP stack.
                   1044: 
                   1045: Currently there is no way to define locals with user-defined data
                   1046: structures, but we are working on it.
                   1047: 
1.7       pazsan   1048: GNU Forth allows defining locals everywhere in a colon definition. This
                   1049: poses the following questions:
1.2       anton    1050: 
1.4       anton    1051: @menu
                   1052: * Where are locals visible by name?::  
1.14      anton    1053: * How long do locals live?::    
1.4       anton    1054: * Programming Style::           
                   1055: * Implementation::              
                   1056: @end menu
                   1057: 
                   1058: @node Where are locals visible by name?, How long do locals live?, gforth locals, gforth locals
1.2       anton    1059: @subsubsection Where are locals visible by name?
                   1060: 
                   1061: Basically, the answer is that locals are visible where you would expect
                   1062: it in block-structured languages, and sometimes a little longer. If you
                   1063: want to restrict the scope of a local, enclose its definition in
                   1064: @code{SCOPE}...@code{ENDSCOPE}.
                   1065: 
                   1066: doc-scope
                   1067: doc-endscope
                   1068: 
                   1069: These words behave like control structure words, so you can use them
                   1070: with @code{CS-PICK} and @code{CS-ROLL} to restrict the scope in
                   1071: arbitrary ways.
                   1072: 
                   1073: If you want a more exact answer to the visibility question, here's the
                   1074: basic principle: A local is visible in all places that can only be
                   1075: reached through the definition of the local@footnote{In compiler
                   1076: construction terminology, all places dominated by the definition of the
                   1077: local.}. In other words, it is not visible in places that can be reached
                   1078: without going through the definition of the local. E.g., locals defined
                   1079: in @code{IF}...@code{ENDIF} are visible until the @code{ENDIF}, locals
                   1080: defined in @code{BEGIN}...@code{UNTIL} are visible after the
                   1081: @code{UNTIL} (until, e.g., a subsequent @code{ENDSCOPE}).
                   1082: 
                   1083: The reasoning behind this solution is: We want to have the locals
                   1084: visible as long as it is meaningful. The user can always make the
                   1085: visibility shorter by using explicit scoping. In a place that can
                   1086: only be reached through the definition of a local, the meaning of a
                   1087: local name is clear. In other places it is not: How is the local
                   1088: initialized at the control flow path that does not contain the
                   1089: definition? Which local is meant, if the same name is defined twice in
                   1090: two independent control flow paths?
                   1091: 
                   1092: This should be enough detail for nearly all users, so you can skip the
                   1093: rest of this section. If you relly must know all the gory details and
                   1094: options, read on.
                   1095: 
                   1096: In order to implement this rule, the compiler has to know which places
                   1097: are unreachable. It knows this automatically after @code{AHEAD},
                   1098: @code{AGAIN}, @code{EXIT} and @code{LEAVE}; in other cases (e.g., after
                   1099: most @code{THROW}s), you can use the word @code{UNREACHABLE} to tell the
                   1100: compiler that the control flow never reaches that place. If
                   1101: @code{UNREACHABLE} is not used where it could, the only consequence is
                   1102: that the visibility of some locals is more limited than the rule above
                   1103: says. If @code{UNREACHABLE} is used where it should not (i.e., if you
                   1104: lie to the compiler), buggy code will be produced.
                   1105: 
                   1106: Another problem with this rule is that at @code{BEGIN}, the compiler
1.3       anton    1107: does not know which locals will be visible on the incoming
                   1108: back-edge. All problems discussed in the following are due to this
                   1109: ignorance of the compiler (we discuss the problems using @code{BEGIN}
                   1110: loops as examples; the discussion also applies to @code{?DO} and other
1.2       anton    1111: loops). Perhaps the most insidious example is:
                   1112: @example
                   1113: AHEAD
                   1114: BEGIN
                   1115:   x
                   1116: [ 1 CS-ROLL ] THEN
1.4       anton    1117:   @{ x @}
1.2       anton    1118:   ...
                   1119: UNTIL
                   1120: @end example
                   1121: 
                   1122: This should be legal according to the visibility rule. The use of
                   1123: @code{x} can only be reached through the definition; but that appears
                   1124: textually below the use.
                   1125: 
                   1126: From this example it is clear that the visibility rules cannot be fully
                   1127: implemented without major headaches. Our implementation treats common
                   1128: cases as advertised and the exceptions are treated in a safe way: The
                   1129: compiler makes a reasonable guess about the locals visible after a
                   1130: @code{BEGIN}; if it is too pessimistic, the
                   1131: user will get a spurious error about the local not being defined; if the
                   1132: compiler is too optimistic, it will notice this later and issue a
                   1133: warning. In the case above the compiler would complain about @code{x}
                   1134: being undefined at its use. You can see from the obscure examples in
                   1135: this section that it takes quite unusual control structures to get the
                   1136: compiler into trouble, and even then it will often do fine.
                   1137: 
                   1138: If the @code{BEGIN} is reachable from above, the most optimistic guess
                   1139: is that all locals visible before the @code{BEGIN} will also be
                   1140: visible after the @code{BEGIN}. This guess is valid for all loops that
                   1141: are entered only through the @code{BEGIN}, in particular, for normal
                   1142: @code{BEGIN}...@code{WHILE}...@code{REPEAT} and
                   1143: @code{BEGIN}...@code{UNTIL} loops and it is implemented in our
                   1144: compiler. When the branch to the @code{BEGIN} is finally generated by
                   1145: @code{AGAIN} or @code{UNTIL}, the compiler checks the guess and
                   1146: warns the user if it was too optimisitic:
                   1147: @example
                   1148: IF
1.4       anton    1149:   @{ x @}
1.2       anton    1150: BEGIN
                   1151:   \ x ? 
                   1152: [ 1 cs-roll ] THEN
                   1153:   ...
                   1154: UNTIL
                   1155: @end example
                   1156: 
                   1157: Here, @code{x} lives only until the @code{BEGIN}, but the compiler
                   1158: optimistically assumes that it lives until the @code{THEN}. It notices
                   1159: this difference when it compiles the @code{UNTIL} and issues a
                   1160: warning. The user can avoid the warning, and make sure that @code{x}
                   1161: is not used in the wrong area by using explicit scoping:
                   1162: @example
                   1163: IF
                   1164:   SCOPE
1.4       anton    1165:   @{ x @}
1.2       anton    1166:   ENDSCOPE
                   1167: BEGIN
                   1168: [ 1 cs-roll ] THEN
                   1169:   ...
                   1170: UNTIL
                   1171: @end example
                   1172: 
                   1173: Since the guess is optimistic, there will be no spurious error messages
                   1174: about undefined locals.
                   1175: 
                   1176: If the @code{BEGIN} is not reachable from above (e.g., after
                   1177: @code{AHEAD} or @code{EXIT}), the compiler cannot even make an
                   1178: optimistic guess, as the locals visible after the @code{BEGIN} may be
                   1179: defined later. Therefore, the compiler assumes that no locals are
                   1180: visible after the @code{BEGIN}. However, the useer can use
                   1181: @code{ASSUME-LIVE} to make the compiler assume that the same locals are
                   1182: visible at the BEGIN as at the point where the item was created.
                   1183: 
                   1184: doc-assume-live
                   1185: 
                   1186: E.g.,
                   1187: @example
1.4       anton    1188: @{ x @}
1.2       anton    1189: AHEAD
                   1190: ASSUME-LIVE
                   1191: BEGIN
                   1192:   x
                   1193: [ 1 CS-ROLL ] THEN
                   1194:   ...
                   1195: UNTIL
                   1196: @end example
                   1197: 
                   1198: Other cases where the locals are defined before the @code{BEGIN} can be
                   1199: handled by inserting an appropriate @code{CS-ROLL} before the
                   1200: @code{ASSUME-LIVE} (and changing the control-flow stack manipulation
                   1201: behind the @code{ASSUME-LIVE}).
                   1202: 
                   1203: Cases where locals are defined after the @code{BEGIN} (but should be
                   1204: visible immediately after the @code{BEGIN}) can only be handled by
                   1205: rearranging the loop. E.g., the ``most insidious'' example above can be
                   1206: arranged into:
                   1207: @example
                   1208: BEGIN
1.4       anton    1209:   @{ x @}
1.2       anton    1210:   ... 0=
                   1211: WHILE
                   1212:   x
                   1213: REPEAT
                   1214: @end example
                   1215: 
1.4       anton    1216: @node How long do locals live?, Programming Style, Where are locals visible by name?, gforth locals
1.2       anton    1217: @subsubsection How long do locals live?
                   1218: 
                   1219: The right answer for the lifetime question would be: A local lives at
                   1220: least as long as it can be accessed. For a value-flavoured local this
                   1221: means: until the end of its visibility. However, a variable-flavoured
                   1222: local could be accessed through its address far beyond its visibility
                   1223: scope. Ultimately, this would mean that such locals would have to be
                   1224: garbage collected. Since this entails un-Forth-like implementation
                   1225: complexities, I adopted the same cowardly solution as some other
                   1226: languages (e.g., C): The local lives only as long as it is visible;
                   1227: afterwards its address is invalid (and programs that access it
                   1228: afterwards are erroneous).
                   1229: 
1.4       anton    1230: @node Programming Style, Implementation, How long do locals live?, gforth locals
1.2       anton    1231: @subsubsection Programming Style
                   1232: 
                   1233: The freedom to define locals anywhere has the potential to change
                   1234: programming styles dramatically. In particular, the need to use the
                   1235: return stack for intermediate storage vanishes. Moreover, all stack
                   1236: manipulations (except @code{PICK}s and @code{ROLL}s with run-time
                   1237: determined arguments) can be eliminated: If the stack items are in the
                   1238: wrong order, just write a locals definition for all of them; then
                   1239: write the items in the order you want.
                   1240: 
                   1241: This seems a little far-fetched and eliminating stack manipulations is
1.4       anton    1242: unlikely to become a conscious programming objective. Still, the number
                   1243: of stack manipulations will be reduced dramatically if local variables
                   1244: are used liberally (e.g., compare @code{max} in @ref{gforth locals} with
                   1245: a traditional implementation of @code{max}).
1.2       anton    1246: 
                   1247: This shows one potential benefit of locals: making Forth programs more
                   1248: readable. Of course, this benefit will only be realized if the
                   1249: programmers continue to honour the principle of factoring instead of
                   1250: using the added latitude to make the words longer.
                   1251: 
                   1252: Using @code{TO} can and should be avoided.  Without @code{TO},
                   1253: every value-flavoured local has only a single assignment and many
                   1254: advantages of functional languages apply to Forth. I.e., programs are
                   1255: easier to analyse, to optimize and to read: It is clear from the
                   1256: definition what the local stands for, it does not turn into something
                   1257: different later.
                   1258: 
                   1259: E.g., a definition using @code{TO} might look like this:
                   1260: @example
                   1261: : strcmp @{ addr1 u1 addr2 u2 -- n @}
                   1262:  u1 u2 min 0
                   1263:  ?do
                   1264:    addr1 c@ addr2 c@ - ?dup
                   1265:    if
                   1266:      unloop exit
                   1267:    then
                   1268:    addr1 char+ TO addr1
                   1269:    addr2 char+ TO addr2
                   1270:  loop
                   1271:  u1 u2 - ;
                   1272: @end example
                   1273: Here, @code{TO} is used to update @code{addr1} and @code{addr2} at
                   1274: every loop iteration. @code{strcmp} is a typical example of the
                   1275: readability problems of using @code{TO}. When you start reading
                   1276: @code{strcmp}, you think that @code{addr1} refers to the start of the
                   1277: string. Only near the end of the loop you realize that it is something
                   1278: else.
                   1279: 
                   1280: This can be avoided by defining two locals at the start of the loop that
                   1281: are initialized with the right value for the current iteration.
                   1282: @example
                   1283: : strcmp @{ addr1 u1 addr2 u2 -- n @}
                   1284:  addr1 addr2
                   1285:  u1 u2 min 0 
                   1286:  ?do @{ s1 s2 @}
                   1287:    s1 c@ s2 c@ - ?dup 
                   1288:    if
                   1289:      unloop exit
                   1290:    then
                   1291:    s1 char+ s2 char+
                   1292:  loop
                   1293:  2drop
                   1294:  u1 u2 - ;
                   1295: @end example
                   1296: Here it is clear from the start that @code{s1} has a different value
                   1297: in every loop iteration.
                   1298: 
1.4       anton    1299: @node Implementation,  , Programming Style, gforth locals
1.2       anton    1300: @subsubsection Implementation
                   1301: 
                   1302: GNU Forth uses an extra locals stack. The most compelling reason for
                   1303: this is that the return stack is not float-aligned; using an extra stack
                   1304: also eliminates the problems and restrictions of using the return stack
                   1305: as locals stack. Like the other stacks, the locals stack grows toward
                   1306: lower addresses. A few primitives allow an efficient implementation:
                   1307: 
                   1308: doc-@local#
                   1309: doc-f@local#
                   1310: doc-laddr#
                   1311: doc-lp+!#
                   1312: doc-lp!
                   1313: doc->l
                   1314: doc-f>l
                   1315: 
                   1316: In addition to these primitives, some specializations of these
                   1317: primitives for commonly occurring inline arguments are provided for
                   1318: efficiency reasons, e.g., @code{@@local0} as specialization of
                   1319: @code{@@local#} for the inline argument 0. The following compiling words
                   1320: compile the right specialized version, or the general version, as
                   1321: appropriate:
                   1322: 
1.12      anton    1323: doc-compile-@local
                   1324: doc-compile-f@local
1.2       anton    1325: doc-compile-lp+!
                   1326: 
                   1327: Combinations of conditional branches and @code{lp+!#} like
                   1328: @code{?branch-lp+!#} (the locals pointer is only changed if the branch
                   1329: is taken) are provided for efficiency and correctness in loops.
                   1330: 
                   1331: A special area in the dictionary space is reserved for keeping the
                   1332: local variable names. @code{@{} switches the dictionary pointer to this
                   1333: area and @code{@}} switches it back and generates the locals
                   1334: initializing code. @code{W:} etc.@ are normal defining words. This
                   1335: special area is cleared at the start of every colon definition.
                   1336: 
                   1337: A special feature of GNU Forths dictionary is used to implement the
                   1338: definition of locals without type specifiers: every wordlist (aka
                   1339: vocabulary) has its own methods for searching
1.4       anton    1340: etc. (@pxref{Wordlists}). For the present purpose we defined a wordlist
1.2       anton    1341: with a special search method: When it is searched for a word, it
                   1342: actually creates that word using @code{W:}. @code{@{} changes the search
                   1343: order to first search the wordlist containing @code{@}}, @code{W:} etc.,
                   1344: and then the wordlist for defining locals without type specifiers.
                   1345: 
                   1346: The lifetime rules support a stack discipline within a colon
                   1347: definition: The lifetime of a local is either nested with other locals
                   1348: lifetimes or it does not overlap them.
                   1349: 
                   1350: At @code{BEGIN}, @code{IF}, and @code{AHEAD} no code for locals stack
                   1351: pointer manipulation is generated. Between control structure words
                   1352: locals definitions can push locals onto the locals stack. @code{AGAIN}
                   1353: is the simplest of the other three control flow words. It has to
                   1354: restore the locals stack depth of the corresponding @code{BEGIN}
                   1355: before branching. The code looks like this:
                   1356: @format
                   1357: @code{lp+!#} current-locals-size @minus{} dest-locals-size
                   1358: @code{branch} <begin>
                   1359: @end format
                   1360: 
                   1361: @code{UNTIL} is a little more complicated: If it branches back, it
                   1362: must adjust the stack just like @code{AGAIN}. But if it falls through,
                   1363: the locals stack must not be changed. The compiler generates the
                   1364: following code:
                   1365: @format
                   1366: @code{?branch-lp+!#} <begin> current-locals-size @minus{} dest-locals-size
                   1367: @end format
                   1368: The locals stack pointer is only adjusted if the branch is taken.
                   1369: 
                   1370: @code{THEN} can produce somewhat inefficient code:
                   1371: @format
                   1372: @code{lp+!#} current-locals-size @minus{} orig-locals-size
                   1373: <orig target>:
                   1374: @code{lp+!#} orig-locals-size @minus{} new-locals-size
                   1375: @end format
                   1376: The second @code{lp+!#} adjusts the locals stack pointer from the
1.4       anton    1377: level at the @var{orig} point to the level after the @code{THEN}. The
1.2       anton    1378: first @code{lp+!#} adjusts the locals stack pointer from the current
                   1379: level to the level at the orig point, so the complete effect is an
                   1380: adjustment from the current level to the right level after the
                   1381: @code{THEN}.
                   1382: 
                   1383: In a conventional Forth implementation a dest control-flow stack entry
                   1384: is just the target address and an orig entry is just the address to be
                   1385: patched. Our locals implementation adds a wordlist to every orig or dest
                   1386: item. It is the list of locals visible (or assumed visible) at the point
                   1387: described by the entry. Our implementation also adds a tag to identify
                   1388: the kind of entry, in particular to differentiate between live and dead
                   1389: (reachable and unreachable) orig entries.
                   1390: 
                   1391: A few unusual operations have to be performed on locals wordlists:
                   1392: 
                   1393: doc-common-list
                   1394: doc-sub-list?
                   1395: doc-list-size
                   1396: 
                   1397: Several features of our locals wordlist implementation make these
                   1398: operations easy to implement: The locals wordlists are organised as
                   1399: linked lists; the tails of these lists are shared, if the lists
                   1400: contain some of the same locals; and the address of a name is greater
                   1401: than the address of the names behind it in the list.
                   1402: 
                   1403: Another important implementation detail is the variable
                   1404: @code{dead-code}. It is used by @code{BEGIN} and @code{THEN} to
                   1405: determine if they can be reached directly or only through the branch
                   1406: that they resolve. @code{dead-code} is set by @code{UNREACHABLE},
                   1407: @code{AHEAD}, @code{EXIT} etc., and cleared at the start of a colon
                   1408: definition, by @code{BEGIN} and usually by @code{THEN}.
                   1409: 
                   1410: Counted loops are similar to other loops in most respects, but
                   1411: @code{LEAVE} requires special attention: It performs basically the same
                   1412: service as @code{AHEAD}, but it does not create a control-flow stack
                   1413: entry. Therefore the information has to be stored elsewhere;
                   1414: traditionally, the information was stored in the target fields of the
                   1415: branches created by the @code{LEAVE}s, by organizing these fields into a
                   1416: linked list. Unfortunately, this clever trick does not provide enough
                   1417: space for storing our extended control flow information. Therefore, we
                   1418: introduce another stack, the leave stack. It contains the control-flow
                   1419: stack entries for all unresolved @code{LEAVE}s.
                   1420: 
                   1421: Local names are kept until the end of the colon definition, even if
                   1422: they are no longer visible in any control-flow path. In a few cases
                   1423: this may lead to increased space needs for the locals name area, but
                   1424: usually less than reclaiming this space would cost in code size.
                   1425: 
                   1426: 
1.4       anton    1427: @node ANS Forth locals,  , gforth locals, Locals
1.2       anton    1428: @subsection ANS Forth locals
                   1429: 
                   1430: The ANS Forth locals wordset does not define a syntax for locals, but
                   1431: words that make it possible to define various syntaxes. One of the
                   1432: possible syntaxes is a subset of the syntax we used in the gforth locals
                   1433: wordset, i.e.:
                   1434: 
                   1435: @example
                   1436: @{ local1 local2 ... -- comment @}
                   1437: @end example
                   1438: or
                   1439: @example
                   1440: @{ local1 local2 ... @}
                   1441: @end example
                   1442: 
                   1443: The order of the locals corresponds to the order in a stack comment. The
                   1444: restrictions are:
1.1       anton    1445: 
1.2       anton    1446: @itemize @bullet
                   1447: @item
                   1448: Locals can only be cell-sized values (no type specifers are allowed).
                   1449: @item
                   1450: Locals can be defined only outside control structures.
                   1451: @item
                   1452: Locals can interfere with explicit usage of the return stack. For the
                   1453: exact (and long) rules, see the standard. If you don't use return stack
                   1454: accessing words in a definition using locals, you will we all right. The
                   1455: purpose of this rule is to make locals implementation on the return
                   1456: stack easier.
                   1457: @item
                   1458: The whole definition must be in one line.
                   1459: @end itemize
                   1460: 
                   1461: Locals defined in this way behave like @code{VALUE}s
1.4       anton    1462: (@xref{Values}). I.e., they are initialized from the stack. Using their
1.2       anton    1463: name produces their value. Their value can be changed using @code{TO}.
                   1464: 
                   1465: Since this syntax is supported by gforth directly, you need not do
                   1466: anything to use it. If you want to port a program using this syntax to
                   1467: another ANS Forth system, use @file{anslocal.fs} to implement the syntax
                   1468: on the other system.
                   1469: 
                   1470: Note that a syntax shown in the standard, section A.13 looks
                   1471: similar, but is quite different in having the order of locals
                   1472: reversed. Beware!
                   1473: 
                   1474: The ANS Forth locals wordset itself consists of the following word
                   1475: 
                   1476: doc-(local)
                   1477: 
                   1478: The ANS Forth locals extension wordset defines a syntax, but it is so
                   1479: awful that we strongly recommend not to use it. We have implemented this
                   1480: syntax to make porting to gforth easy, but do not document it here. The
                   1481: problem with this syntax is that the locals are defined in an order
                   1482: reversed with respect to the standard stack comment notation, making
                   1483: programs harder to read, and easier to misread and miswrite. The only
                   1484: merit of this syntax is that it is easy to implement using the ANS Forth
                   1485: locals wordset.
1.3       anton    1486: 
1.4       anton    1487: @node Defining Words, Wordlists, Locals, Words
                   1488: @section Defining Words
                   1489: 
1.14      anton    1490: @menu
                   1491: * Values::                      
                   1492: @end menu
                   1493: 
1.4       anton    1494: @node Values,  , Defining Words, Defining Words
                   1495: @subsection Values
                   1496: 
                   1497: @node Wordlists, Files, Defining Words, Words
                   1498: @section Wordlists
                   1499: 
                   1500: @node Files, Blocks, Wordlists, Words
                   1501: @section Files
                   1502: 
                   1503: @node Blocks, Other I/O, Files, Words
                   1504: @section Blocks
                   1505: 
                   1506: @node Other I/O, Programming Tools, Blocks, Words
                   1507: @section Other I/O
                   1508: 
                   1509: @node Programming Tools, Threading Words, Other I/O, Words
                   1510: @section Programming Tools
                   1511: 
1.5       anton    1512: @menu
                   1513: * Debugging::                   Simple and quick.
                   1514: * Assertions::                  Making your programs self-checking.
                   1515: @end menu
                   1516: 
                   1517: @node Debugging, Assertions, Programming Tools, Programming Tools
1.4       anton    1518: @subsection Debugging
                   1519: 
                   1520: The simple debugging aids provided in @file{debugging.fs}
                   1521: are meant to support a different style of debugging than the
                   1522: tracing/stepping debuggers used in languages with long turn-around
                   1523: times.
                   1524: 
                   1525: A much better (faster) way in fast-compilig languages is to add
                   1526: printing code at well-selected places, let the program run, look at
                   1527: the output, see where things went wrong, add more printing code, etc.,
                   1528: until the bug is found.
                   1529: 
                   1530: The word @code{~~} is easy to insert. It just prints debugging
                   1531: information (by default the source location and the stack contents). It
                   1532: is also easy to remove (@kbd{C-x ~} in the Emacs Forth mode to
                   1533: query-replace them with nothing). The deferred words
                   1534: @code{printdebugdata} and @code{printdebugline} control the output of
                   1535: @code{~~}. The default source location output format works well with
                   1536: Emacs' compilation mode, so you can step through the program at the
1.5       anton    1537: source level using @kbd{C-x `} (the advantage over a stepping debugger
                   1538: is that you can step in any direction and you know where the crash has
                   1539: happened or where the strange data has occurred).
1.4       anton    1540: 
                   1541: Note that the default actions clobber the contents of the pictured
                   1542: numeric output string, so you should not use @code{~~}, e.g., between
                   1543: @code{<#} and @code{#>}.
                   1544: 
                   1545: doc-~~
                   1546: doc-printdebugdata
                   1547: doc-printdebugline
                   1548: 
1.5       anton    1549: @node Assertions,  , Debugging, Programming Tools
1.4       anton    1550: @subsection Assertions
                   1551: 
1.5       anton    1552: It is a good idea to make your programs self-checking, in particular, if
                   1553: you use an assumption (e.g., that a certain field of a data structure is
                   1554: never zero) that may become wrong during maintenance. GForth supports
                   1555: assertions for this purpose. They are used like this:
                   1556: 
                   1557: @example
                   1558: assert( @var{flag} )
                   1559: @end example
                   1560: 
                   1561: The code between @code{assert(} and @code{)} should compute a flag, that
                   1562: should be true if everything is alright and false otherwise. It should
                   1563: not change anything else on the stack. The overall stack effect of the
                   1564: assertion is @code{( -- )}. E.g.
                   1565: 
                   1566: @example
                   1567: assert( 1 1 + 2 = ) \ what we learn in school
                   1568: assert( dup 0<> ) \ assert that the top of stack is not zero
                   1569: assert( false ) \ this code should not be reached
                   1570: @end example
                   1571: 
                   1572: The need for assertions is different at different times. During
                   1573: debugging, we want more checking, in production we sometimes care more
                   1574: for speed. Therefore, assertions can be turned off, i.e., the assertion
                   1575: becomes a comment. Depending on the importance of an assertion and the
                   1576: time it takes to check it, you may want to turn off some assertions and
                   1577: keep others turned on. GForth provides several levels of assertions for
                   1578: this purpose:
                   1579: 
                   1580: doc-assert0(
                   1581: doc-assert1(
                   1582: doc-assert2(
                   1583: doc-assert3(
                   1584: doc-assert(
                   1585: doc-)
                   1586: 
                   1587: @code{Assert(} is the same as @code{assert1(}. The variable
                   1588: @code{assert-level} specifies the highest assertions that are turned
                   1589: on. I.e., at the default @code{assert-level} of one, @code{assert0(} and
                   1590: @code{assert1(} assertions perform checking, while @code{assert2(} and
                   1591: @code{assert3(} assertions are treated as comments.
                   1592: 
                   1593: Note that the @code{assert-level} is evaluated at compile-time, not at
                   1594: run-time. I.e., you cannot turn assertions on or off at run-time, you
                   1595: have to set the @code{assert-level} appropriately before compiling a
                   1596: piece of code. You can compile several pieces of code at several
                   1597: @code{assert-level}s (e.g., a trusted library at level 1 and newly
                   1598: written code at level 3).
                   1599: 
                   1600: doc-assert-level
                   1601: 
                   1602: If an assertion fails, a message compatible with Emacs' compilation mode
                   1603: is produced and the execution is aborted (currently with @code{ABORT"}.
                   1604: If there is interest, we will introduce a special throw code. But if you
                   1605: intend to @code{catch} a specific condition, using @code{throw} is
                   1606: probably more appropriate than an assertion).
                   1607: 
1.4       anton    1608: @node Threading Words,  , Programming Tools, Words
                   1609: @section Threading Words
                   1610: 
                   1611: These words provide access to code addresses and other threading stuff
                   1612: in gforth (and, possibly, other interpretive Forths). It more or less
                   1613: abstracts away the differences between direct and indirect threading
                   1614: (and, for direct threading, the machine dependences). However, at
                   1615: present this wordset is still inclomplete. It is also pretty low-level;
                   1616: some day it will hopefully be made unnecessary by an internals words set
                   1617: that abstracts implementation details away completely.
                   1618: 
                   1619: doc->code-address
                   1620: doc->does-code
                   1621: doc-code-address!
                   1622: doc-does-code!
                   1623: doc-does-handler!
                   1624: doc-/does-handler
                   1625: 
1.14      anton    1626: 
                   1627: 
1.4       anton    1628: @node ANS conformance, Model, Words, Top
                   1629: @chapter ANS conformance
                   1630: 
1.15    ! anton    1631: To the best of our knowledge, gforth is an
1.14      anton    1632: 
1.15    ! anton    1633: ANS Forth System
        !          1634: @itemize
        !          1635: @item providing the Core Extensions word set
        !          1636: @item providing the Block word set
        !          1637: @item providing the Block Extensions word set
        !          1638: @item providing the Double-Number word set
        !          1639: @item providing the Double-Number Extensions word set
        !          1640: @item providing the Exception word set
        !          1641: @item providing the Exception Extensions word set
        !          1642: @item providing the Facility word set
        !          1643: @item providing @code{MS} and @code{TIME&DATE} from the Facility Extensions word set
        !          1644: @item providing the File Access word set
        !          1645: @item providing the File Access Extensions word set
        !          1646: @item providing the Floating-Point word set
        !          1647: @item providing the Floating-Point Extensions word set
        !          1648: @item providing the Locals word set
        !          1649: @item providing the Locals Extensions word set
        !          1650: @item providing the Memory-Allocation word set
        !          1651: @item providing the Memory-Allocation Extensions word set (that one's easy)
        !          1652: @item providing the Programming-Tools word set
        !          1653: @item providing @code{AHEAD}, @code{BYE}, @code{CS-PICK}, @code{CS-ROLL}, @code{STATE}, @code{[ELSE]}, @code{[IF]}, @code{[THEN]} from the Programming-Tools Extensions word set
        !          1654: @item providing the Search-Order word set
        !          1655: @item providing the Search-Order Extensions word set
        !          1656: @item providing the String word set
        !          1657: @item providing the String Extensions word set (another easy one)
        !          1658: @end itemize
        !          1659: 
        !          1660: In addition, ANS Forth systems are required to document certain
        !          1661: implementation choices. This chapter tries to meet these
        !          1662: requirements. In many cases it gives a way to ask the system for the
        !          1663: information instead of providing the information directly, in
        !          1664: particular, if the information depends on the processor, the operating
        !          1665: system or the installation options chosen, or if they are likely to
        !          1666: change during the maintenance of gforth.
        !          1667: 
1.14      anton    1668: @comment The framework for the rest has been taken from pfe.
                   1669: 
                   1670: @menu
                   1671: * The Core Words::              
                   1672: * The optional Block word set::  
                   1673: * The optional Double Number word set::  
                   1674: * The optional Exception word set::  
                   1675: * The optional Facility word set::  
                   1676: * The optional File-Access word set::  
                   1677: * The optional Floating-Point word set::  
                   1678: * The optional Locals word set::  
                   1679: * The optional Memory-Allocation word set::  
                   1680: * The optional Programming-Tools word set::  
                   1681: * The optional Search-Order word set::  
                   1682: @end menu
                   1683: 
                   1684: 
                   1685: @c =====================================================================
                   1686: @node The Core Words, The optional Block word set, ANS conformance, ANS conformance
                   1687: @comment  node-name,  next,  previous,  up
                   1688: @section The Core Words
                   1689: @c =====================================================================
                   1690: 
                   1691: @menu
1.15    ! anton    1692: * core-idef::                   Implementation Defined Options                   
        !          1693: * core-ambcond::                Ambiguous Conditions                
        !          1694: * core-other::                  Other System Documentation                  
1.14      anton    1695: @end menu
                   1696: 
                   1697: @c ---------------------------------------------------------------------
                   1698: @node core-idef, core-ambcond, The Core Words, The Core Words
                   1699: @subsection Implementation Defined Options
                   1700: @c ---------------------------------------------------------------------
                   1701: 
                   1702: @table @i
                   1703: 
                   1704: @item (Cell) aligned addresses:
                   1705: processor-dependent. Gforths alignment words perform natural alignment
                   1706: (e.g., an address aligned for a datum of size 8 is divisible by
                   1707: 8). Unaligned accesses usually result in a @code{-23 THROW}.
                   1708: 
                   1709: @item @code{EMIT} and non-graphic characters:
                   1710: The character is output using the C library function (actually, macro)
                   1711: @code{putchar}.
                   1712: 
                   1713: @item character editing of @code{ACCEPT} and @code{EXPECT}:
                   1714: This is modeled on the GNU readline library (@pxref{Readline
                   1715: Interaction, , Command Line Editing, readline, The GNU Readline
                   1716: Library}) with Emacs-like key bindings. @kbd{Tab} deviates a little by
                   1717: producing a full word completion every time you type it (instead of
                   1718: producing the common prefix of all completions).
                   1719: 
                   1720: @item character set:
                   1721: The character set of your computer and display device. Gforth is
                   1722: 8-bit-clean (but some other component in your system may make trouble).
                   1723: 
                   1724: @item Character-aligned address requirements:
                   1725: installation-dependent. Currently a character is represented by a C
                   1726: @code{unsigned char}; in the future we might switch to @code{wchar_t}
                   1727: (Comments on that requested).
                   1728: 
                   1729: @item character-set extensions and matching of names:
                   1730: Any character except 0 can be used in a name. Matching is
                   1731: case-insensitive. The matching is performed using the C function
                   1732: @code{strncasecmp}, whose function is probably influenced by the
                   1733: locale. E.g., the @code{C} locale does not know about accents and
                   1734: umlauts, so they are matched case-sensitively in that locale. For
                   1735: portability reasons it is best to write programs such that they work in
                   1736: the @code{C} locale. Then one can use libraries written by a Polish
                   1737: programmer (who might use words containing ISO Latin-2 encoded
                   1738: characters) and by a French programmer (ISO Latin-1) in the same program
                   1739: (of course, @code{WORDS} will produce funny results for some of the
                   1740: words (which ones, depends on the font you are using)). Also, the locale
                   1741: you prefer may not be available in other operating systems. Hopefully,
                   1742: Unicode will solve these problems one day.
                   1743: 
                   1744: @item conditions under which control characters match a space delimiter:
                   1745: If @code{WORD} is called with the space character as a delimiter, all
                   1746: white-space characters (as identified by the C macro @code{isspace()})
                   1747: are delimiters. @code{PARSE}, on the other hand, treats space like other
                   1748: delimiters. @code{PARSE-WORD} treats space like @code{WORD}, but behaves
                   1749: like @code{PARSE} otherwise. @code{(NAME)}, which is used by the outer
                   1750: interpreter (aka text interpreter) by default, treats all white-space
                   1751: characters as delimiters.
                   1752: 
                   1753: @item format of the control flow stack:
                   1754: The data stack is used as control flow stack. The size of a control flow
                   1755: stack item in cells is given by the constant @code{cs-item-size}. At the
                   1756: time of this writing, an item consists of a (pointer to a) locals list
                   1757: (third), an address in the code (second), and a tag for identifying the
                   1758: item (TOS). The following tags are used: @code{defstart},
                   1759: @code{live-orig}, @code{dead-orig}, @code{dest}, @code{do-dest},
                   1760: @code{scopestart}.
                   1761: 
                   1762: @item conversion of digits > 35
                   1763: The characters @code{[\]^_'} are the digits with the decimal value
                   1764: 36@minus{}41. There is no way to input many of the larger digits.
                   1765: 
                   1766: @item display after input terminates in @code{ACCEPT} and @code{EXPECT}:
                   1767: The cursor is moved to the end of the entered string. If the input is
                   1768: terminated using the @kbd{Return} key, a space is typed.
                   1769: 
                   1770: @item exception abort sequence of @code{ABORT"}:
                   1771: The error string is stored into the variable @code{"error} and a
                   1772: @code{-2 throw} is performed.
                   1773: 
                   1774: @item input line terminator:
                   1775: For interactive input, @kbd{C-m} and @kbd{C-j} terminate lines. One of
                   1776: these characters is typically produced when you type the @kbd{Enter} or
                   1777: @kbd{Return} key.
                   1778: 
                   1779: @item maximum size of a counted string:
                   1780: @code{s" /counted-string" environment? drop .}. Currently 255 characters
                   1781: on all ports, but this may change.
                   1782: 
                   1783: @item maximum size of a parsed string:
                   1784: Given by the constant @code{/line}. Currently 255 characters.
                   1785: 
                   1786: @item maximum size of a definition name, in characters:
                   1787: 31
                   1788: 
                   1789: @item maximum string length for @code{ENVIRONMENT?}, in characters:
                   1790: 31
                   1791: 
                   1792: @item method of selecting the user input device:
                   1793: The user input device is the standard input. There is current no way to
                   1794: change it from within gforth. However, the input can typically be
                   1795: redirected in the command line that starts gforth.
                   1796: 
                   1797: @item method of selecting the user output device:
                   1798: The user output device is the standard output. It cannot be redirected
                   1799: from within gforth, but typically from the command line that starts
                   1800: gforth. Gforth uses buffered output, so output on a terminal does not
                   1801: become visible before the next newline or buffer overflow. Output on
                   1802: non-terminals is invisible until the buffer overflows.
                   1803: 
                   1804: @item methods of dictionary compilation:
                   1805: Waht are we expected to document here?
                   1806: 
                   1807: @item number of bits in one address unit:
                   1808: @code{s" address-units-bits" environment? drop .}. 8 in all current
                   1809: ports.
                   1810: 
                   1811: @item number representation and arithmetic:
                   1812: Processor-dependent. Binary two's complement on all current ports.
                   1813: 
                   1814: @item ranges for integer types:
                   1815: Installation-dependent. Make environmental queries for @code{MAX-N},
                   1816: @code{MAX-U}, @code{MAX-D} and @code{MAX-UD}. The lower bounds for
                   1817: unsigned (and positive) types is 0. The lower bound for signed types on
                   1818: two's complement and one's complement machines machines can be computed
                   1819: by adding 1 to the upper bound.
                   1820: 
                   1821: @item read-only data space regions:
                   1822: The whole Forth data space is writable.
                   1823: 
                   1824: @item size of buffer at @code{WORD}:
                   1825: @code{PAD HERE - .}. 104 characters on 32-bit machines. The buffer is
                   1826: shared with the pictured numeric output string. If overwriting
                   1827: @code{PAD} is acceptable, it is as large as the remaining dictionary
                   1828: space, although only as much can be sensibly used as fits in a counted
                   1829: string.
                   1830: 
                   1831: @item size of one cell in address units:
                   1832: @code{1 cells .}.
                   1833: 
                   1834: @item size of one character in address units:
                   1835: @code{1 chars .}. 1 on all current ports.
                   1836: 
                   1837: @item size of the keyboard terminal buffer:
                   1838: Varies. You can determine the size at a specific time using @code{lp@
                   1839: tib - .}. It is shared with the locals stack and TIBs of files that
                   1840: include the current file. You can change the amount of space for TIBs
                   1841: and locals stack at gforth startup with the command line option
                   1842: @code{-l}.
                   1843: 
                   1844: @item size of the pictured numeric output buffer:
                   1845: @code{PAD HERE - .}. 104 characters on 32-bit machines. The buffer is
                   1846: shared with @code{WORD}.
                   1847: 
                   1848: @item size of the scratch area returned by @code{PAD}:
                   1849: The remainder of dictionary space. You can even use the unused part of
                   1850: the data stack space. The current size can be computed with @code{sp@
                   1851: pad - .}.
                   1852: 
                   1853: @item system case-sensitivity characteristics:
                   1854: Dictionary searches are case insensitive. However, as explained above
                   1855: under @i{character-set extensions}, the matching for non-ASCII
                   1856: characters is determined by the locale you are using. In the default
                   1857: @code{C} locale all non-ASCII characters are matched case-sensitively.
                   1858: 
                   1859: @item system prompt:
                   1860: @code{ ok} in interpret state, @code{ compiled} in compile state.
                   1861: 
                   1862: @item division rounding:
                   1863: installation dependent. @code{s" floored" environment? drop .}. We leave
                   1864: the choice to gcc (what to use for @code{/}) and to you (whether to use
                   1865: @code{fm/mod}, @code{sm/rem} or simply @code{/}).
                   1866: 
                   1867: @item values of @code{STATE} when true:
                   1868: -1.
                   1869: 
                   1870: @item values returned after arithmetic overflow:
                   1871: On two's complement machines, arithmetic is performed modulo
                   1872: 2**bits-per-cell for single arithmetic and 4**bits-per-cell for double
                   1873: arithmetic (with appropriate mapping for signed types). Division by zero
                   1874: typically results in a @code{-55 throw} (floatingpoint unidentified
                   1875: fault), although a @code{-10 throw} (divide by zero) would be more
                   1876: appropriate.
                   1877: 
                   1878: @item whether the current definition can be found after @t{DOES>}:
                   1879: No.
                   1880: 
                   1881: @end table
                   1882: 
                   1883: @c ---------------------------------------------------------------------
                   1884: @node core-ambcond, core-other, core-idef, The Core Words
                   1885: @subsection Ambiguous conditions
                   1886: @c ---------------------------------------------------------------------
                   1887: 
                   1888: @table @i
                   1889: 
                   1890: @item a name is neither a word nor a number:
                   1891: @code{-13 throw} (Undefined word)
                   1892: 
                   1893: @item a definition name exceeds the maximum length allowed:
                   1894: @code{-19 throw} (Word name too long)
                   1895: 
                   1896: @item addressing a region not inside the various data spaces of the forth system:
                   1897: The stacks, code space and name space are accessible. Machine code space is
                   1898: typically readable. Accessing other addresses gives results dependent on
                   1899: the operating system. On decent systems: @code{-9 throw} (Invalid memory
                   1900: address).
                   1901: 
                   1902: @item argument type incompatible with parameter:
                   1903: This is usually not caught. Some words perform checks, e.g., the control
                   1904: flow words, and issue a @code{ABORT"} or @code{-12 THROW} (Argument type
                   1905: mismatch).
                   1906: 
                   1907: @item attempting to obtain the execution token of a word with undefined execution semantics:
                   1908: You get an execution token representing the compilation semantics
                   1909: instead.
                   1910: 
                   1911: @item dividing by zero:
                   1912: typically results in a @code{-55 throw} (floating point unidentified
                   1913: fault), although a @code{-10 throw} (divide by zero) would be more
                   1914: appropriate.
                   1915: 
                   1916: @item insufficient data stack or return stack space:
                   1917: Not checked. This typically results in mysterious illegal memory
                   1918: accesses, producing @code{-9 throw} (Invalid memory address) or
                   1919: @code{-23 throw} (Address alignment exception).
                   1920: 
                   1921: @item insufficient space for loop control parameters:
                   1922: like other return stack overflows.
                   1923: 
                   1924: @item insufficient space in the dictionary:
                   1925: Not checked. Similar results as stack overflows. However, typically the
                   1926: error appears at a different place when one inserts or removes code.
                   1927: 
                   1928: @item interpreting a word with undefined interpretation semantics:
                   1929: For some words, we defined interpretation semantics. For the others:
                   1930: @code{-14 throw} (Interpreting a compile-only word). Note that this is
                   1931: checked only by the outer (aka text) interpreter; if the word is
                   1932: @code{execute}d in some other way, it will typically perform it's
                   1933: compilation semantics even in interpret state. (We could change @code{'}
                   1934: and relatives not to give the xt of such words, but we think that would
                   1935: be too restrictive).
                   1936: 
                   1937: @item modifying the contents of the input buffer or a string literal:
                   1938: These are located in writable memory and can be modified.
                   1939: 
                   1940: @item overflow of the pictured numeric output string:
                   1941: Not checked.
                   1942: 
                   1943: @item parsed string overflow:
                   1944: @code{PARSE} cannot overflow. @code{WORD} does not check for overflow.
                   1945: 
                   1946: @item producing a result out of range:
                   1947: On two's complement machines, arithmetic is performed modulo
                   1948: 2**bits-per-cell for single arithmetic and 4**bits-per-cell for double
                   1949: arithmetic (with appropriate mapping for signed types). Division by zero
                   1950: typically results in a @code{-55 throw} (floatingpoint unidentified
                   1951: fault), although a @code{-10 throw} (divide by zero) would be more
                   1952: appropriate. @code{convert} and @code{>number} currently overflow
                   1953: silently.
                   1954: 
                   1955: @item reading from an empty data or return stack:
                   1956: The data stack is checked by the outer (aka text) interpreter after
                   1957: every word executed. If it has underflowed, a @code{-4 throw} (Stack
                   1958: underflow) is performed. Apart from that, the stacks are not checked and
                   1959: underflows can result in similar behaviour as overflows (of adjacent
                   1960: stacks).
                   1961: 
                   1962: @item unexepected end of the input buffer, resulting in an attempt to use a zero-length string as a name:
                   1963: @code{Create} and its descendants perform a @code{-16 throw} (Attempt to
                   1964: use zero-length string as a name). Words like @code{'} probably will not
                   1965: find what they search. Note that it is possible to create zero-length
                   1966: names with @code{nextname} (should it not?).
                   1967: 
                   1968: @item @code{>IN} greater than input buffer:
                   1969: The next invocation of a parsing word returns a string wih length 0.
                   1970: 
                   1971: @item @code{RECURSE} appears after @code{DOES>}:
                   1972: Compiles a recursive call to the defining word not to the defined word.
                   1973: 
                   1974: @item argument input source different than current input source for @code{RESTORE-INPUT}:
                   1975: !!???If the argument input source is a valid input source then it gets
                   1976: restored. Otherwise causes @code{-12 THROW} which unless caught issues
                   1977: the message "argument type mismatch" and aborts.
                   1978: 
                   1979: @item data space containing definitions gets de-allocated:
                   1980: Deallocation with @code{allot} is not checked. This typically resuls in
                   1981: memory access faults or execution of illegal instructions.
                   1982: 
                   1983: @item data space read/write with incorrect alignment:
                   1984: Processor-dependent. Typically results in a @code{-23 throw} (Address
                   1985: alignment exception). Under Linux on a 486 or later processor with
                   1986: alignment turned on, incorrect alignment results in a @code{-9 throw}
                   1987: (Invalid memory address). There are reportedly some processors with
                   1988: alignment restrictions that do not report them.
                   1989: 
                   1990: @item data space pointer not properly aligned, @code{,}, @code{C,}:
                   1991: Like other alignment errors.
                   1992: 
                   1993: @item less than u+2 stack items (@code{PICK} and @code{ROLL}):
                   1994: Not checked. May cause an illegal memory access.
                   1995: 
                   1996: @item loop control parameters not available:
                   1997: Not checked. The counted loop words simply assume that the top of return
                   1998: stack items are loop control parameters and behave accordingly.
                   1999: 
                   2000: @item most recent definition does not have a name (@code{IMMEDIATE}):
                   2001: @code{abort" last word was headerless"}.
                   2002: 
                   2003: @item name not defined by @code{VALUE} used by @code{TO}:
                   2004: @code{-32 throw} (Invalid name argument)
                   2005: 
1.15    ! anton    2006: @item name not found (@code{'}, @code{POSTPONE}, @code{[']}, @code{[COMPILE]}):
1.14      anton    2007: @code{-13 throw} (Undefined word)
                   2008: 
                   2009: @item parameters are not of the same type (@code{DO}, @code{?DO}, @code{WITHIN}):
                   2010: Gforth behaves as if they were of the same type. I.e., you can predict
                   2011: the behaviour by interpreting all parameters as, e.g., signed.
                   2012: 
                   2013: @item @code{POSTPONE} or @code{[COMPILE]} applied to @code{TO}:
                   2014: Assume @code{: X POSTPONE TO ; IMMEDIATE}. @code{X} is equivalent to
                   2015: @code{TO}.
                   2016: 
                   2017: @item String longer than a counted string returned by @code{WORD}:
                   2018: Not checked. The string will be ok, but the count will, of course,
                   2019: contain only the least significant bits of the length.
                   2020: 
1.15    ! anton    2021: @item u greater than or equal to the number of bits in a cell (@code{LSHIFT}, @code{RSHIFT}):
1.14      anton    2022: Processor-dependent. Typical behaviours are returning 0 and using only
                   2023: the low bits of the shift count.
                   2024: 
                   2025: @item word not defined via @code{CREATE}:
                   2026: @code{>BODY} produces the PFA of the word no matter how it was defined.
                   2027: 
                   2028: @code{DOES>} changes the execution semantics of the last defined word no
                   2029: matter how it was defined. E.g., @code{CONSTANT DOES>} is equivalent to
                   2030: @code{CREATE , DOES>}.
                   2031: 
                   2032: @item words improperly used outside @code{<#} and @code{#>}:
                   2033: Not checked. As usual, you can expect memory faults.
                   2034: 
                   2035: @end table
                   2036: 
                   2037: 
                   2038: @c ---------------------------------------------------------------------
                   2039: @node core-other,  , core-ambcond, The Core Words
                   2040: @subsection Other system documentation
                   2041: @c ---------------------------------------------------------------------
                   2042: 
                   2043: @table @i
                   2044: 
                   2045: @item nonstandard words using @code{PAD}:
                   2046: None.
                   2047: 
                   2048: @item operator's terminal facilities available:
                   2049: !!??
                   2050: 
                   2051: @item program data space available:
                   2052: @code{sp@ here - .} gives the space remaining for dictionary and data
                   2053: stack together.
                   2054: 
                   2055: @item return stack space available:
                   2056: !!??
                   2057: 
                   2058: @item stack space available:
                   2059: @code{sp@ here - .} gives the space remaining for dictionary and data
                   2060: stack together.
                   2061: 
                   2062: @item system dictionary space required, in address units:
                   2063: Type @code{here forthstart - .} after startup. At the time of this
                   2064: writing, this gives 70108 (bytes) on a 32-bit system.
                   2065: @end table
                   2066: 
                   2067: 
                   2068: @c =====================================================================
                   2069: @node The optional Block word set, The optional Double Number word set, The Core Words, ANS conformance
                   2070: @section The optional Block word set
                   2071: @c =====================================================================
                   2072: 
                   2073: @menu
1.15    ! anton    2074: * block-idef::                  Implementation Defined Options                  
        !          2075: * block-ambcond::               Ambiguous Conditions               
        !          2076: * block-other::                 Other System Documentation                 
1.14      anton    2077: @end menu
                   2078: 
                   2079: 
                   2080: @c ---------------------------------------------------------------------
                   2081: @node block-idef, block-ambcond, The optional Block word set, The optional Block word set
                   2082: @subsection Implementation Defined Options
                   2083: @c ---------------------------------------------------------------------
                   2084: 
                   2085: @table @i
                   2086: 
                   2087: @item the format for display by @code{LIST}:
                   2088: First the screen number is displayed, then 16 lines of 64 characters,
                   2089: each line preceded by the line number.
                   2090: 
                   2091: @item the length of a line affected by @code{\}:
                   2092: 64 characters.
                   2093: @end table
                   2094: 
                   2095: 
                   2096: @c ---------------------------------------------------------------------
                   2097: @node block-ambcond, block-other, block-idef, The optional Block word set
                   2098: @subsection Ambiguous conditions
                   2099: @c ---------------------------------------------------------------------
                   2100: 
                   2101: @table @i
                   2102: 
                   2103: @item correct block read was not possible:
                   2104: Typically results in a @code{throw} of some OS-derived value (between
                   2105: -512 and -2048). If the blocks file was just not long enough, blanks are
                   2106: supplied for the missing portion.
                   2107: 
                   2108: @item I/O exception in block transfer:
                   2109: Typically results in a @code{throw} of some OS-derived value (between
                   2110: -512 and -2048).
                   2111: 
                   2112: @item invalid block number:
                   2113: @code{-35 throw} (Invalid block number)
                   2114: 
                   2115: @item a program directly alters the contents of @code{BLK}:
                   2116: The input stream is switched to that other block, at the same
                   2117: position. If the storing to @code{BLK} happens when interpreting
                   2118: non-block input, the system will get quite confused when the block ends.
                   2119: 
                   2120: @item no current block buffer for @code{UPDATE}:
                   2121: @code{UPDATE} has no effect.
                   2122: 
                   2123: @end table
                   2124: 
                   2125: 
                   2126: @c ---------------------------------------------------------------------
                   2127: @node block-other,  , block-ambcond, The optional Block word set
                   2128: @subsection Other system documentation
                   2129: @c ---------------------------------------------------------------------
                   2130: 
                   2131: @table @i
                   2132: 
                   2133: @item any restrictions a multiprogramming system places on the use of buffer addresses:
                   2134: No restrictions (yet).
                   2135: 
                   2136: @item the number of blocks available for source and data:
                   2137: depends on your disk space.
                   2138: 
                   2139: @end table
                   2140: 
                   2141: 
                   2142: @c =====================================================================
                   2143: @node The optional Double Number word set, The optional Exception word set, The optional Block word set, ANS conformance
                   2144: @section The optional Double Number word set
                   2145: @c =====================================================================
                   2146: 
                   2147: @menu
1.15    ! anton    2148: * double-ambcond::              Ambiguous Conditions              
1.14      anton    2149: @end menu
                   2150: 
                   2151: 
                   2152: @c ---------------------------------------------------------------------
1.15    ! anton    2153: @node double-ambcond,  , The optional Double Number word set, The optional Double Number word set
1.14      anton    2154: @subsection Ambiguous conditions
                   2155: @c ---------------------------------------------------------------------
                   2156: 
                   2157: @table @i
                   2158: 
1.15    ! anton    2159: @item @var{d} outside of range of @var{n} in @code{D>S}:
1.14      anton    2160: The least significant cell of @var{d} is produced.
                   2161: 
                   2162: @end table
                   2163: 
                   2164: 
                   2165: @c =====================================================================
                   2166: @node The optional Exception word set, The optional Facility word set, The optional Double Number word set, ANS conformance
                   2167: @section The optional Exception word set
                   2168: @c =====================================================================
                   2169: 
                   2170: @menu
1.15    ! anton    2171: * exception-idef::              Implementation Defined Options              
1.14      anton    2172: @end menu
                   2173: 
                   2174: 
                   2175: @c ---------------------------------------------------------------------
1.15    ! anton    2176: @node exception-idef,  , The optional Exception word set, The optional Exception word set
1.14      anton    2177: @subsection Implementation Defined Options
                   2178: @c ---------------------------------------------------------------------
                   2179: 
                   2180: @table @i
                   2181: @item @code{THROW}-codes used in the system:
                   2182: The codes -256@minus{}-511 are used for reporting signals (see
                   2183: @file{errore.fs}). The codes -512@minus{}-2047 are used for OS errors
                   2184: (for file and memory allocation operations). The mapping from OS error
                   2185: numbers to throw code is -512@minus{}@var{errno}. One side effect of
                   2186: this mapping is that undefined OS errors produce a message with a
                   2187: strange number; e.g., @code{-1000 THROW} results in @code{Unknown error
                   2188: 488} on my system.
                   2189: @end table
                   2190: 
                   2191: @c =====================================================================
                   2192: @node The optional Facility word set, The optional File-Access word set, The optional Exception word set, ANS conformance
                   2193: @section The optional Facility word set
                   2194: @c =====================================================================
                   2195: 
                   2196: @menu
1.15    ! anton    2197: * facility-idef::               Implementation Defined Options               
        !          2198: * facility-ambcond::            Ambiguous Conditions            
1.14      anton    2199: @end menu
                   2200: 
                   2201: 
                   2202: @c ---------------------------------------------------------------------
                   2203: @node facility-idef, facility-ambcond, The optional Facility word set, The optional Facility word set
                   2204: @subsection Implementation Defined Options
                   2205: @c ---------------------------------------------------------------------
                   2206: 
                   2207: @table @i
                   2208: 
                   2209: @item encoding of keyboard events (@code{EKEY}):
                   2210: Not yet implemeted.
                   2211: 
                   2212: @item duration of a system clock tick
                   2213: System dependent. With respect to @code{MS}, the time is specified in
                   2214: microseconds. How well the OS and the hardware implement this, is
                   2215: another question.
                   2216: 
                   2217: @item repeatability to be expected from the execution of @code{MS}:
                   2218: System dependent. On Unix, a lot depends on load. If the system is
                   2219: lightly loaded, and the delay is short enough that gforth does not get
                   2220: swapped out, the performance should be acceptable. Under MS-DOS and
                   2221: other single-tasking systems, it should be good.
                   2222: 
                   2223: @end table
                   2224: 
                   2225: 
                   2226: @c ---------------------------------------------------------------------
1.15    ! anton    2227: @node facility-ambcond,  , facility-idef, The optional Facility word set
1.14      anton    2228: @subsection Ambiguous conditions
                   2229: @c ---------------------------------------------------------------------
                   2230: 
                   2231: @table @i
                   2232: 
                   2233: @item @code{AT-XY} can't be performed on user output device:
                   2234: Largely terminal dependant. No range checks are done on the arguments.
                   2235: No errors are reported. You may see some garbage appearing, you may see
                   2236: simply nothing happen.
                   2237: 
                   2238: @end table
                   2239: 
                   2240: 
                   2241: @c =====================================================================
                   2242: @node The optional File-Access word set, The optional Floating-Point word set, The optional Facility word set, ANS conformance
                   2243: @section The optional File-Access word set
                   2244: @c =====================================================================
                   2245: 
                   2246: @menu
1.15    ! anton    2247: * file-idef::                   Implementation Defined Options                   
        !          2248: * file-ambcond::                Ambiguous Conditions                
1.14      anton    2249: @end menu
                   2250: 
                   2251: 
                   2252: @c ---------------------------------------------------------------------
                   2253: @node file-idef, file-ambcond, The optional File-Access word set, The optional File-Access word set
                   2254: @subsection Implementation Defined Options
                   2255: @c ---------------------------------------------------------------------
                   2256: 
                   2257: @table @i
                   2258: 
                   2259: @item File access methods used:
                   2260: @code{R/O}, @code{R/W} and @code{BIN} work as you would
                   2261: expect. @code{W/O} translates into the C file opening mode @code{w} (or
                   2262: @code{wb}): The file is cleared, if it exists, and created, if it does
1.15    ! anton    2263: not (both with @code{open-file} and @code{create-file}).  Under Unix
1.14      anton    2264: @code{create-file} creates a file with 666 permissions modified by your
                   2265: umask.
                   2266: 
                   2267: @item file exceptions:
                   2268: The file words do not raise exceptions (except, perhaps, memory access
                   2269: faults when you pass illegal addresses or file-ids).
                   2270: 
                   2271: @item file line terminator:
                   2272: System-dependent. Gforth uses C's newline character as line
                   2273: terminator. What the actual character code(s) of this are is
                   2274: system-dependent.
                   2275: 
                   2276: @item file name format
                   2277: System dependent. Gforth just uses the file name format of your OS.
                   2278: 
                   2279: @item information returned by @code{FILE-STATUS}:
                   2280: @code{FILE-STATUS} returns the most powerful file access mode allowed
                   2281: for the file: Either @code{R/O}, @code{W/O} or @code{R/W}. If the file
                   2282: cannot be accessed, @code{R/O BIN} is returned. @code{BIN} is applicable
                   2283: along with the retured mode.
                   2284: 
                   2285: @item input file state after an exception when including source:
                   2286: All files that are left via the exception are closed.
                   2287: 
                   2288: @item @var{ior} values and meaning:
1.15    ! anton    2289: The @var{ior}s returned by the file and memory allocation words are
        !          2290: intended as throw codes. They typically are in the range
        !          2291: -512@minus{}-2047 of OS errors.  The mapping from OS error numbers to
        !          2292: @var{ior}s is -512@minus{}@var{errno}.
1.14      anton    2293: 
                   2294: @item maximum depth of file input nesting:
                   2295: limited by the amount of return stack, locals/TIB stack, and the number
                   2296: of open files available. This should not give you troubles.
                   2297: 
                   2298: @item maximum size of input line:
                   2299: @code{/line}. Currently 255.
                   2300: 
                   2301: @item methods of mapping block ranges to files:
                   2302: Currently, the block words automatically access the file
                   2303: @file{blocks.fb} in the currend working directory. More sophisticated
                   2304: methods could be implemented if there is demand (and a volunteer).
                   2305: 
                   2306: @item number of string buffers provided by @code{S"}:
                   2307: 1
                   2308: 
                   2309: @item size of string buffer used by @code{S"}:
                   2310: @code{/line}. currently 255.
                   2311: 
                   2312: @end table
                   2313: 
                   2314: @c ---------------------------------------------------------------------
1.15    ! anton    2315: @node file-ambcond,  , file-idef, The optional File-Access word set
1.14      anton    2316: @subsection Ambiguous conditions
                   2317: @c ---------------------------------------------------------------------
                   2318: 
                   2319: @table @i
                   2320: 
                   2321: @item attempting to position a file outside it's boundaries:
                   2322: @code{REPOSITION-FILE} is performed as usual: Afterwards,
                   2323: @code{FILE-POSITION} returns the value given to @code{REPOSITION-FILE}.
                   2324: 
                   2325: @item attempting to read from file positions not yet written:
                   2326: End-of-file, i.e., zero characters are read and no error is reported.
                   2327: 
                   2328: @item @var{file-id} is invalid (@code{INCLUDE-FILE}):
                   2329: An appropriate exception may be thrown, but a memory fault or other
                   2330: problem is more probable.
                   2331: 
                   2332: @item I/O exception reading or closing @var{file-id} (@code{include-file}, @code{included}):
                   2333: The @var{ior} produced by the operation, that discovered the problem, is
                   2334: thrown.
                   2335: 
                   2336: @item named file cannot be opened (@code{included}):
                   2337: The @var{ior} produced by @code{open-file} is thrown.
                   2338: 
                   2339: @item requesting an unmapped block number:
                   2340: There are no unmapped legal block numbers. On some operating systems,
                   2341: writing a block with a large number may overflow the file system and
                   2342: have an error message as consequence.
                   2343: 
                   2344: @item using @code{source-id} when @code{blk} is non-zero:
                   2345: @code{source-id} performs its function. Typically it will give the id of
                   2346: the source which loaded the block. (Better ideas?)
                   2347: 
                   2348: @end table
                   2349: 
                   2350: 
                   2351: @c =====================================================================
                   2352: @node  The optional Floating-Point word set, The optional Locals word set, The optional File-Access word set, ANS conformance
1.15    ! anton    2353: @section The optional Floating-Point word set
1.14      anton    2354: @c =====================================================================
                   2355: 
                   2356: @menu
1.15    ! anton    2357: * floating-idef::               Implementation Defined Options
        !          2358: * floating-ambcond::            Ambiguous Conditions            
1.14      anton    2359: @end menu
                   2360: 
                   2361: 
                   2362: @c ---------------------------------------------------------------------
                   2363: @node floating-idef, floating-ambcond, The optional Floating-Point word set, The optional Floating-Point word set
                   2364: @subsection Implementation Defined Options
                   2365: @c ---------------------------------------------------------------------
                   2366: 
                   2367: @table @i
                   2368: 
1.15    ! anton    2369: @item format and range of floating point numbers:
        !          2370: System-dependent; the @code{double} type of C.
1.14      anton    2371: 
1.15    ! anton    2372: @item results of @code{REPRESENT} when @var{float} is out of range:
        !          2373: System dependent; @code{REPRESENT} is implemented using the C library
        !          2374: function @code{ecvt()} and inherits its behaviour in this respect.
1.14      anton    2375: 
1.15    ! anton    2376: @item rounding or truncation of floating-point numbers:
        !          2377: What's the question?!!
1.14      anton    2378: 
1.15    ! anton    2379: @item size of floating-point stack:
        !          2380: @code{s" FLOATING-STACK" environment? drop .}. Can be changed at startup
        !          2381: with the command-line option @code{-f}.
1.14      anton    2382: 
1.15    ! anton    2383: @item width of floating-point stack:
        !          2384: @code{1 floats}.
1.14      anton    2385: 
                   2386: @end table
                   2387: 
                   2388: 
                   2389: @c ---------------------------------------------------------------------
1.15    ! anton    2390: @node floating-ambcond,  , floating-idef, The optional Floating-Point word set
        !          2391: @subsection Ambiguous conditions
1.14      anton    2392: @c ---------------------------------------------------------------------
                   2393: 
                   2394: @table @i
                   2395: 
1.15    ! anton    2396: @item @code{df@@} or @code{df!} used with an address that is not double-float  aligned:
        !          2397: System-dependent. Typically results in an alignment fault like other
        !          2398: alignment violations.
1.14      anton    2399: 
1.15    ! anton    2400: @item @code{f@@} or @code{f!} used with an address that is not float  aligned:
        !          2401: System-dependent. Typically results in an alignment fault like other
        !          2402: alignment violations.
1.14      anton    2403: 
1.15    ! anton    2404: @item Floating-point result out of range:
        !          2405: System-dependent. Can result in a @code{-55 THROW} (Floating-point
        !          2406: unidentified fault), or can produce a special value representing, e.g.,
        !          2407: Infinity.
1.14      anton    2408: 
1.15    ! anton    2409: @item @code{sf@@} or @code{sf!} used with an address that is not single-float  aligned:
        !          2410: System-dependent. Typically results in an alignment fault like other
        !          2411: alignment violations.
1.14      anton    2412: 
1.15    ! anton    2413: @item BASE is not decimal (@code{REPRESENT}, @code{F.}, @code{FE.}, @code{FS.}):
        !          2414: The floating-point number is converted into decimal nonetheless.
1.14      anton    2415: 
1.15    ! anton    2416: @item Both arguments are equal to zero (@code{FATAN2}):
        !          2417: System-dependent. @code{FATAN2} is implemented using the C library
        !          2418: function @code{atan2()}.
1.14      anton    2419: 
1.15    ! anton    2420: @item Using ftan on an argument @var{r1} where cos(@var{r1}) is zero:
        !          2421: System-dependent. Anyway, typically the cos of @var{r1} will not be zero
        !          2422: because of small errors and the tan will be a very large (or very small)
        !          2423: but finite number.
1.14      anton    2424: 
1.15    ! anton    2425: @item @var{d} cannot be presented precisely as a float in @code{D>F}:
        !          2426: The result is rounded to the nearest float.
1.14      anton    2427: 
1.15    ! anton    2428: @item dividing by zero:
        !          2429: @code{-55 throw} (Floating-point unidentified fault)
1.14      anton    2430: 
1.15    ! anton    2431: @item exponent too big for conversion (@code{DF!}, @code{DF@@}, @code{SF!}, @code{SF@@}):
        !          2432: System dependent. On IEEE-FP based systems the number is converted into
        !          2433: an infinity.
1.14      anton    2434: 
1.15    ! anton    2435: @item @var{float}<1 (@code{facosh}):
        !          2436: @code{-55 throw} (Floating-point unidentified fault)
1.14      anton    2437: 
1.15    ! anton    2438: @item @var{float}=<-1 (@code{flnp1}):
        !          2439: @code{-55 throw} (Floating-point unidentified fault). On IEEE-FP systems
        !          2440: negative infinity is typically produced for @var{float}=-1.
1.14      anton    2441: 
1.15    ! anton    2442: @item @var{float}=<0 (@code{fln}, @code{flog}):
        !          2443: @code{-55 throw} (Floating-point unidentified fault). On IEEE-FP systems
        !          2444: negative infinity is typically produced for @var{float}=0.
1.14      anton    2445: 
1.15    ! anton    2446: @item @var{float}<0 (@code{fasinh}, @code{fsqrt}):
        !          2447: @code{-55 throw} (Floating-point unidentified fault). @code{fasinh}
        !          2448: produces values for these inputs on my Linux box (Bug in the C library?)
1.14      anton    2449: 
1.15    ! anton    2450: @item |@var{float}|>1 (@code{facos}, @code{fasin}, @code{fatanh}):
        !          2451: @code{-55 throw} (Floating-point unidentified fault).
1.14      anton    2452: 
1.15    ! anton    2453: @item integer part of float cannot be represented by @var{d} in @code{f>d}:
        !          2454: @code{-55 throw} (Floating-point unidentified fault).
1.14      anton    2455: 
1.15    ! anton    2456: @item string larger than pictured numeric output area (@code{f.}, @code{fe.}, @code{fs.}):
        !          2457: This does not happen.
        !          2458: @end table
1.14      anton    2459: 
                   2460: 
                   2461: 
                   2462: @c =====================================================================
1.15    ! anton    2463: @node  The optional Locals word set, The optional Memory-Allocation word set, The optional Floating-Point word set, ANS conformance
        !          2464: @section The optional Locals word set
1.14      anton    2465: @c =====================================================================
                   2466: 
                   2467: @menu
1.15    ! anton    2468: * locals-idef::                 Implementation Defined Options                 
        !          2469: * locals-ambcond::              Ambiguous Conditions              
1.14      anton    2470: @end menu
                   2471: 
                   2472: 
                   2473: @c ---------------------------------------------------------------------
1.15    ! anton    2474: @node locals-idef, locals-ambcond, The optional Locals word set, The optional Locals word set
1.14      anton    2475: @subsection Implementation Defined Options
                   2476: @c ---------------------------------------------------------------------
                   2477: 
                   2478: @table @i
                   2479: 
1.15    ! anton    2480: @item maximum number of locals in a definition:
        !          2481: @code{s" #locals" environment? drop .}. Currently 15. This is a lower
        !          2482: bound, e.g., on a 32-bit machine there can be 41 locals of up to 8
        !          2483: characters. The number of locals in a definition is bounded by the size
        !          2484: of locals-buffer, which contains the names of the locals.
1.14      anton    2485: 
                   2486: @end table
                   2487: 
                   2488: 
                   2489: @c ---------------------------------------------------------------------
1.15    ! anton    2490: @node locals-ambcond,  , locals-idef, The optional Locals word set
1.14      anton    2491: @subsection Ambiguous conditions
                   2492: @c ---------------------------------------------------------------------
                   2493: 
                   2494: @table @i
                   2495: 
1.15    ! anton    2496: @item executing a named local in interpretation state:
        !          2497: @code{-14 throw} (Interpreting a compile-only word).
1.14      anton    2498: 
1.15    ! anton    2499: @item @var{name} not defined by @code{VALUE} or @code{(LOCAL)} (@code{TO}):
        !          2500: @code{-32 throw} (Invalid name argument)
1.14      anton    2501: 
                   2502: @end table
                   2503: 
                   2504: 
                   2505: @c =====================================================================
1.15    ! anton    2506: @node  The optional Memory-Allocation word set, The optional Programming-Tools word set, The optional Locals word set, ANS conformance
        !          2507: @section The optional Memory-Allocation word set
1.14      anton    2508: @c =====================================================================
                   2509: 
                   2510: @menu
1.15    ! anton    2511: * memory-idef::                 Implementation Defined Options                 
1.14      anton    2512: @end menu
                   2513: 
                   2514: 
                   2515: @c ---------------------------------------------------------------------
1.15    ! anton    2516: @node memory-idef,  , The optional Memory-Allocation word set, The optional Memory-Allocation word set
1.14      anton    2517: @subsection Implementation Defined Options
                   2518: @c ---------------------------------------------------------------------
                   2519: 
                   2520: @table @i
                   2521: 
1.15    ! anton    2522: @item values and meaning of @var{ior}:
        !          2523: The @var{ior}s returned by the file and memory allocation words are
        !          2524: intended as throw codes. They typically are in the range
        !          2525: -512@minus{}-2047 of OS errors.  The mapping from OS error numbers to
        !          2526: @var{ior}s is -512@minus{}@var{errno}.
1.14      anton    2527: 
                   2528: @end table
                   2529: 
                   2530: @c =====================================================================
1.15    ! anton    2531: @node  The optional Programming-Tools word set, The optional Search-Order word set, The optional Memory-Allocation word set, ANS conformance
        !          2532: @section The optional Programming-Tools word set
1.14      anton    2533: @c =====================================================================
                   2534: 
                   2535: @menu
1.15    ! anton    2536: * programming-idef::            Implementation Defined Options            
        !          2537: * programming-ambcond::         Ambiguous Conditions         
1.14      anton    2538: @end menu
                   2539: 
                   2540: 
                   2541: @c ---------------------------------------------------------------------
1.15    ! anton    2542: @node programming-idef, programming-ambcond, The optional Programming-Tools word set, The optional Programming-Tools word set
1.14      anton    2543: @subsection Implementation Defined Options
                   2544: @c ---------------------------------------------------------------------
                   2545: 
                   2546: @table @i
                   2547: 
1.15    ! anton    2548: @item ending sequence for input following @code{;code} and @code{code}:
        !          2549: Not implemented (yet).
1.14      anton    2550: 
1.15    ! anton    2551: @item manner of processing input following @code{;code} and @code{code}:
        !          2552: Not implemented (yet).
        !          2553: 
        !          2554: @item search order capability for @code{EDITOR} and @code{ASSEMBLER}:
        !          2555: Not implemented (yet). If they were implemented, they would use the
        !          2556: search order wordset.
        !          2557: 
        !          2558: @item source and format of display by @code{SEE}:
        !          2559: The source for @code{see} is the intermediate code used by the inner
        !          2560: interpreter.  The current @code{see} tries to output Forth source code
        !          2561: as well as possible.
        !          2562: 
1.14      anton    2563: @end table
                   2564: 
                   2565: @c ---------------------------------------------------------------------
1.15    ! anton    2566: @node programming-ambcond,  , programming-idef, The optional Programming-Tools word set
1.14      anton    2567: @subsection Ambiguous conditions
                   2568: @c ---------------------------------------------------------------------
                   2569: 
                   2570: @table @i
                   2571: 
1.15    ! anton    2572: @item deleting the compilation wordlist (@code{FORGET}):
        !          2573: Not implemented (yet).
1.14      anton    2574: 
1.15    ! anton    2575: @item fewer than @var{u}+1 items on the control flow stack (@code{CS-PICK}, @code{CS-ROLL}):
        !          2576: This typically results in an @code{abort"} with a descriptive error
        !          2577: message (may change into a @code{-22 throw} (Control structure mismatch)
        !          2578: in the future). You may also get a memory access error. If you are
        !          2579: unlucky, this ambiguous condition is not caught.
        !          2580: 
        !          2581: @item @var{name} can't be found (@code{forget}):
        !          2582: Not implemented (yet).
1.14      anton    2583: 
1.15    ! anton    2584: @item @var{name} not defined via @code{CREATE}:
        !          2585: @code{;code} is not implemented (yet). If it were, it would behave like
        !          2586: @code{DOES>} in this respect, i.e., change the execution semantics of
        !          2587: the last defined word no matter how it was defined.
1.14      anton    2588: 
1.15    ! anton    2589: @item @code{POSTPONE} applied to @code{[IF]}:
        !          2590: After defining @code{: X POSTPONE [IF] ; IMMEDIATE}. @code{X} is
        !          2591: equivalent to @code{[IF]}.
1.14      anton    2592: 
1.15    ! anton    2593: @item reaching the end of the input source before matching @code{[ELSE]} or @code{[THEN]}:
        !          2594: Continue in the same state of conditional compilation in the next outer
        !          2595: input source. Currently there is no warning to the user about this.
1.14      anton    2596: 
1.15    ! anton    2597: @item removing a needed definition (@code{FORGET}):
        !          2598: Not implemented (yet).
1.14      anton    2599: 
                   2600: @end table
                   2601: 
                   2602: 
                   2603: @c =====================================================================
1.15    ! anton    2604: @node  The optional Search-Order word set,  , The optional Programming-Tools word set, ANS conformance
        !          2605: @section The optional Search-Order word set
1.14      anton    2606: @c =====================================================================
                   2607: 
                   2608: @menu
1.15    ! anton    2609: * search-idef::                 Implementation Defined Options                 
        !          2610: * search-ambcond::              Ambiguous Conditions              
1.14      anton    2611: @end menu
                   2612: 
                   2613: 
                   2614: @c ---------------------------------------------------------------------
1.15    ! anton    2615: @node search-idef, search-ambcond, The optional Search-Order word set, The optional Search-Order word set
1.14      anton    2616: @subsection Implementation Defined Options
                   2617: @c ---------------------------------------------------------------------
                   2618: 
                   2619: @table @i
                   2620: 
1.15    ! anton    2621: @item maximum number of word lists in search order:
        !          2622: @code{s" wordlists" environment? drop .}. Currently 16.
        !          2623: 
        !          2624: @item minimum search order:
        !          2625: @code{root root}.
1.14      anton    2626: 
                   2627: @end table
                   2628: 
                   2629: @c ---------------------------------------------------------------------
1.15    ! anton    2630: @node search-ambcond,  , search-idef, The optional Search-Order word set
1.14      anton    2631: @subsection Ambiguous conditions
                   2632: @c ---------------------------------------------------------------------
                   2633: 
                   2634: @table @i
                   2635: 
1.15    ! anton    2636: @item changing the compilation wordlist (during compilation):
        !          2637: The definition is put into the wordlist that is the compilation wordlist
        !          2638: when @code{REVEAL} is executed (by @code{;}, @code{DOES>},
        !          2639: @code{RECURSIVE}, etc.).
1.14      anton    2640: 
1.15    ! anton    2641: @item search order empty (@code{previous}):
        !          2642: @code{abort" Vocstack empty"}.
1.14      anton    2643: 
1.15    ! anton    2644: @item too many word lists in search order (@code{also}):
        !          2645: @code{abort" Vocstack full"}.
1.14      anton    2646: 
                   2647: @end table
1.13      anton    2648: 
                   2649: 
1.4       anton    2650: @node Model, Emacs and GForth, ANS conformance, Top
                   2651: @chapter Model
                   2652: 
                   2653: @node Emacs and GForth, Internals, Model, Top
                   2654: @chapter Emacs and GForth
                   2655: 
                   2656: GForth comes with @file{gforth.el}, an improved version of
                   2657: @file{forth.el} by Goran Rydqvist (icluded in the TILE package). The
                   2658: improvements are a better (but still not perfect) handling of
                   2659: indentation. I have also added comment paragraph filling (@kbd{M-q}),
1.8       anton    2660: commenting (@kbd{C-x \}) and uncommenting (@kbd{C-u C-x \}) regions and
                   2661: removing debugging tracers (@kbd{C-x ~}, @pxref{Debugging}). I left the
                   2662: stuff I do not use alone, even though some of it only makes sense for
                   2663: TILE. To get a description of these features, enter Forth mode and type
                   2664: @kbd{C-h m}.
1.4       anton    2665: 
                   2666: In addition, GForth supports Emacs quite well: The source code locations
                   2667: given in error messages, debugging output (from @code{~~}) and failed
                   2668: assertion messages are in the right format for Emacs' compilation mode
                   2669: (@pxref{Compilation, , Running Compilations under Emacs, emacs, Emacs
                   2670: Manual}) so the source location corresponding to an error or other
                   2671: message is only a few keystrokes away (@kbd{C-x `} for the next error,
                   2672: @kbd{C-c C-c} for the error under the cursor).
                   2673: 
                   2674: Also, if you @code{include} @file{etags.fs}, a new @file{TAGS} file
                   2675: (@pxref{Tags, , Tags Tables, emacs, Emacs Manual}) will be produced that
                   2676: contains the definitions of all words defined afterwards. You can then
                   2677: find the source for a word using @kbd{M-.}. Note that emacs can use
                   2678: several tags files at the same time (e.g., one for the gforth sources
                   2679: and one for your program).
                   2680: 
                   2681: To get all these benefits, add the following lines to your @file{.emacs}
                   2682: file:
                   2683: 
                   2684: @example
                   2685: (autoload 'forth-mode "gforth.el")
                   2686: (setq auto-mode-alist (cons '("\\.fs\\'" . forth-mode) auto-mode-alist))
                   2687: @end example
                   2688: 
                   2689: @node Internals, Bugs, Emacs and GForth, Top
1.3       anton    2690: @chapter Internals
                   2691: 
                   2692: Reading this section is not necessary for programming with gforth. It
                   2693: should be helpful for finding your way in the gforth sources.
                   2694: 
1.4       anton    2695: @menu
                   2696: * Portability::                 
                   2697: * Threading::                   
                   2698: * Primitives::                  
                   2699: * System Architecture::         
                   2700: @end menu
                   2701: 
                   2702: @node Portability, Threading, Internals, Internals
1.3       anton    2703: @section Portability
                   2704: 
                   2705: One of the main goals of the effort is availability across a wide range
                   2706: of personal machines. fig-Forth, and, to a lesser extent, F83, achieved
                   2707: this goal by manually coding the engine in assembly language for several
                   2708: then-popular processors. This approach is very labor-intensive and the
                   2709: results are short-lived due to progress in computer architecture.
                   2710: 
                   2711: Others have avoided this problem by coding in C, e.g., Mitch Bradley
                   2712: (cforth), Mikael Patel (TILE) and Dirk Zoller (pfe). This approach is
                   2713: particularly popular for UNIX-based Forths due to the large variety of
                   2714: architectures of UNIX machines. Unfortunately an implementation in C
                   2715: does not mix well with the goals of efficiency and with using
                   2716: traditional techniques: Indirect or direct threading cannot be expressed
                   2717: in C, and switch threading, the fastest technique available in C, is
                   2718: significantly slower. Another problem with C is that it's very
                   2719: cumbersome to express double integer arithmetic.
                   2720: 
                   2721: Fortunately, there is a portable language that does not have these
                   2722: limitations: GNU C, the version of C processed by the GNU C compiler
                   2723: (@pxref{C Extensions, , Extensions to the C Language Family, gcc.info,
                   2724: GNU C Manual}). Its labels as values feature (@pxref{Labels as Values, ,
                   2725: Labels as Values, gcc.info, GNU C Manual}) makes direct and indirect
                   2726: threading possible, its @code{long long} type (@pxref{Long Long, ,
                   2727: Double-Word Integers, gcc.info, GNU C Manual}) corresponds to Forths
                   2728: double numbers. GNU C is available for free on all important (and many
                   2729: unimportant) UNIX machines, VMS, 80386s running MS-DOS, the Amiga, and
                   2730: the Atari ST, so a Forth written in GNU C can run on all these
                   2731: machines@footnote{Due to Apple's look-and-feel lawsuit it is not
1.5       anton    2732: available on the Mac (@pxref{Boycott, , Protect Your Freedom---Fight
1.3       anton    2733: ``Look And Feel'', gcc.info, GNU C Manual}).}.
                   2734: 
                   2735: Writing in a portable language has the reputation of producing code that
                   2736: is slower than assembly. For our Forth engine we repeatedly looked at
                   2737: the code produced by the compiler and eliminated most compiler-induced
                   2738: inefficiencies by appropriate changes in the source-code.
                   2739: 
                   2740: However, register allocation cannot be portably influenced by the
                   2741: programmer, leading to some inefficiencies on register-starved
                   2742: machines. We use explicit register declarations (@pxref{Explicit Reg
                   2743: Vars, , Variables in Specified Registers, gcc.info, GNU C Manual}) to
                   2744: improve the speed on some machines. They are turned on by using the
                   2745: @code{gcc} switch @code{-DFORCE_REG}. Unfortunately, this feature not
                   2746: only depends on the machine, but also on the compiler version: On some
                   2747: machines some compiler versions produce incorrect code when certain
                   2748: explicit register declarations are used. So by default
                   2749: @code{-DFORCE_REG} is not used.
                   2750: 
1.4       anton    2751: @node Threading, Primitives, Portability, Internals
1.3       anton    2752: @section Threading
                   2753: 
                   2754: GNU C's labels as values extension (available since @code{gcc-2.0},
                   2755: @pxref{Labels as Values, , Labels as Values, gcc.info, GNU C Manual})
                   2756: makes it possible to take the address of @var{label} by writing
                   2757: @code{&&@var{label}}.  This address can then be used in a statement like
                   2758: @code{goto *@var{address}}. I.e., @code{goto *&&x} is the same as
                   2759: @code{goto x}.
                   2760: 
                   2761: With this feature an indirect threaded NEXT looks like:
                   2762: @example
                   2763: cfa = *ip++;
                   2764: ca = *cfa;
                   2765: goto *ca;
                   2766: @end example
                   2767: For those unfamiliar with the names: @code{ip} is the Forth instruction
                   2768: pointer; the @code{cfa} (code-field address) corresponds to ANS Forths
                   2769: execution token and points to the code field of the next word to be
                   2770: executed; The @code{ca} (code address) fetched from there points to some
                   2771: executable code, e.g., a primitive or the colon definition handler
                   2772: @code{docol}.
                   2773: 
                   2774: Direct threading is even simpler:
                   2775: @example
                   2776: ca = *ip++;
                   2777: goto *ca;
                   2778: @end example
                   2779: 
                   2780: Of course we have packaged the whole thing neatly in macros called
                   2781: @code{NEXT} and @code{NEXT1} (the part of NEXT after fetching the cfa).
                   2782: 
1.4       anton    2783: @menu
                   2784: * Scheduling::                  
                   2785: * Direct or Indirect Threaded?::  
                   2786: * DOES>::                       
                   2787: @end menu
                   2788: 
                   2789: @node Scheduling, Direct or Indirect Threaded?, Threading, Threading
1.3       anton    2790: @subsection Scheduling
                   2791: 
                   2792: There is a little complication: Pipelined and superscalar processors,
                   2793: i.e., RISC and some modern CISC machines can process independent
                   2794: instructions while waiting for the results of an instruction. The
                   2795: compiler usually reorders (schedules) the instructions in a way that
                   2796: achieves good usage of these delay slots. However, on our first tries
                   2797: the compiler did not do well on scheduling primitives. E.g., for
                   2798: @code{+} implemented as
                   2799: @example
                   2800: n=sp[0]+sp[1];
                   2801: sp++;
                   2802: sp[0]=n;
                   2803: NEXT;
                   2804: @end example
                   2805: the NEXT comes strictly after the other code, i.e., there is nearly no
                   2806: scheduling. After a little thought the problem becomes clear: The
                   2807: compiler cannot know that sp and ip point to different addresses (and
1.4       anton    2808: the version of @code{gcc} we used would not know it even if it was
                   2809: possible), so it could not move the load of the cfa above the store to
                   2810: the TOS. Indeed the pointers could be the same, if code on or very near
                   2811: the top of stack were executed. In the interest of speed we chose to
                   2812: forbid this probably unused ``feature'' and helped the compiler in
                   2813: scheduling: NEXT is divided into the loading part (@code{NEXT_P1}) and
                   2814: the goto part (@code{NEXT_P2}). @code{+} now looks like:
1.3       anton    2815: @example
                   2816: n=sp[0]+sp[1];
                   2817: sp++;
                   2818: NEXT_P1;
                   2819: sp[0]=n;
                   2820: NEXT_P2;
                   2821: @end example
1.4       anton    2822: This can be scheduled optimally by the compiler.
1.3       anton    2823: 
                   2824: This division can be turned off with the switch @code{-DCISC_NEXT}. This
                   2825: switch is on by default on machines that do not profit from scheduling
                   2826: (e.g., the 80386), in order to preserve registers.
                   2827: 
1.4       anton    2828: @node Direct or Indirect Threaded?, DOES>, Scheduling, Threading
1.3       anton    2829: @subsection Direct or Indirect Threaded?
                   2830: 
                   2831: Both! After packaging the nasty details in macro definitions we
                   2832: realized that we could switch between direct and indirect threading by
                   2833: simply setting a compilation flag (@code{-DDIRECT_THREADED}) and
                   2834: defining a few machine-specific macros for the direct-threading case.
                   2835: On the Forth level we also offer access words that hide the
                   2836: differences between the threading methods (@pxref{Threading Words}).
                   2837: 
                   2838: Indirect threading is implemented completely
                   2839: machine-independently. Direct threading needs routines for creating
                   2840: jumps to the executable code (e.g. to docol or dodoes). These routines
                   2841: are inherently machine-dependent, but they do not amount to many source
                   2842: lines. I.e., even porting direct threading to a new machine is a small
                   2843: effort.
                   2844: 
1.4       anton    2845: @node DOES>,  , Direct or Indirect Threaded?, Threading
1.3       anton    2846: @subsection DOES>
                   2847: One of the most complex parts of a Forth engine is @code{dodoes}, i.e.,
                   2848: the chunk of code executed by every word defined by a
                   2849: @code{CREATE}...@code{DOES>} pair. The main problem here is: How to find
                   2850: the Forth code to be executed, i.e. the code after the @code{DOES>} (the
                   2851: DOES-code)? There are two solutions:
                   2852: 
                   2853: In fig-Forth the code field points directly to the dodoes and the
                   2854: DOES-code address is stored in the cell after the code address
                   2855: (i.e. at cfa cell+). It may seem that this solution is illegal in the
                   2856: Forth-79 and all later standards, because in fig-Forth this address
                   2857: lies in the body (which is illegal in these standards). However, by
                   2858: making the code field larger for all words this solution becomes legal
                   2859: again. We use this approach for the indirect threaded version. Leaving
                   2860: a cell unused in most words is a bit wasteful, but on the machines we
                   2861: are targetting this is hardly a problem. The other reason for having a
                   2862: code field size of two cells is to avoid having different image files
1.4       anton    2863: for direct and indirect threaded systems (@pxref{System Architecture}).
1.3       anton    2864: 
                   2865: The other approach is that the code field points or jumps to the cell
                   2866: after @code{DOES}. In this variant there is a jump to @code{dodoes} at
                   2867: this address. @code{dodoes} can then get the DOES-code address by
                   2868: computing the code address, i.e., the address of the jump to dodoes,
                   2869: and add the length of that jump field. A variant of this is to have a
                   2870: call to @code{dodoes} after the @code{DOES>}; then the return address
                   2871: (which can be found in the return register on RISCs) is the DOES-code
                   2872: address. Since the two cells available in the code field are usually
                   2873: used up by the jump to the code address in direct threading, we use
                   2874: this approach for direct threading. We did not want to add another
                   2875: cell to the code field.
                   2876: 
1.4       anton    2877: @node Primitives, System Architecture, Threading, Internals
1.3       anton    2878: @section Primitives
                   2879: 
1.4       anton    2880: @menu
                   2881: * Automatic Generation::        
                   2882: * TOS Optimization::            
                   2883: * Produced code::               
                   2884: @end menu
                   2885: 
                   2886: @node Automatic Generation, TOS Optimization, Primitives, Primitives
1.3       anton    2887: @subsection Automatic Generation
                   2888: 
                   2889: Since the primitives are implemented in a portable language, there is no
                   2890: longer any need to minimize the number of primitives. On the contrary,
                   2891: having many primitives is an advantage: speed. In order to reduce the
                   2892: number of errors in primitives and to make programming them easier, we
                   2893: provide a tool, the primitive generator (@file{prims2x.fs}), that
                   2894: automatically generates most (and sometimes all) of the C code for a
                   2895: primitive from the stack effect notation.  The source for a primitive
                   2896: has the following form:
                   2897: 
                   2898: @format
                   2899: @var{Forth-name}       @var{stack-effect}      @var{category}  [@var{pronounc.}]
                   2900: [@code{""}@var{glossary entry}@code{""}]
                   2901: @var{C code}
                   2902: [@code{:}
                   2903: @var{Forth code}]
                   2904: @end format
                   2905: 
                   2906: The items in brackets are optional. The category and glossary fields
                   2907: are there for generating the documentation, the Forth code is there
                   2908: for manual implementations on machines without GNU C. E.g., the source
                   2909: for the primitive @code{+} is:
                   2910: @example
                   2911: +    n1 n2 -- n    core    plus
                   2912: n = n1+n2;
                   2913: @end example
                   2914: 
                   2915: This looks like a specification, but in fact @code{n = n1+n2} is C
                   2916: code. Our primitive generation tool extracts a lot of information from
                   2917: the stack effect notations@footnote{We use a one-stack notation, even
                   2918: though we have separate data and floating-point stacks; The separate
                   2919: notation can be generated easily from the unified notation.}: The number
                   2920: of items popped from and pushed on the stack, their type, and by what
                   2921: name they are referred to in the C code. It then generates a C code
                   2922: prelude and postlude for each primitive. The final C code for @code{+}
                   2923: looks like this:
                   2924: 
                   2925: @example
                   2926: I_plus:        /* + ( n1 n2 -- n ) */  /* label, stack effect */
                   2927: /*  */                          /* documentation */
1.4       anton    2928: @{
1.3       anton    2929: DEF_CA                          /* definition of variable ca (indirect threading) */
                   2930: Cell n1;                        /* definitions of variables */
                   2931: Cell n2;
                   2932: Cell n;
                   2933: n1 = (Cell) sp[1];              /* input */
                   2934: n2 = (Cell) TOS;
                   2935: sp += 1;                        /* stack adjustment */
                   2936: NAME("+")                       /* debugging output (with -DDEBUG) */
1.4       anton    2937: @{
1.3       anton    2938: n = n1+n2;                      /* C code taken from the source */
1.4       anton    2939: @}
1.3       anton    2940: NEXT_P1;                        /* NEXT part 1 */
                   2941: TOS = (Cell)n;                  /* output */
                   2942: NEXT_P2;                        /* NEXT part 2 */
1.4       anton    2943: @}
1.3       anton    2944: @end example
                   2945: 
                   2946: This looks long and inefficient, but the GNU C compiler optimizes quite
                   2947: well and produces optimal code for @code{+} on, e.g., the R3000 and the
                   2948: HP RISC machines: Defining the @code{n}s does not produce any code, and
                   2949: using them as intermediate storage also adds no cost.
                   2950: 
                   2951: There are also other optimizations, that are not illustrated by this
                   2952: example: Assignments between simple variables are usually for free (copy
                   2953: propagation). If one of the stack items is not used by the primitive
                   2954: (e.g.  in @code{drop}), the compiler eliminates the load from the stack
                   2955: (dead code elimination). On the other hand, there are some things that
                   2956: the compiler does not do, therefore they are performed by
                   2957: @file{prims2x.fs}: The compiler does not optimize code away that stores
                   2958: a stack item to the place where it just came from (e.g., @code{over}).
                   2959: 
                   2960: While programming a primitive is usually easy, there are a few cases
                   2961: where the programmer has to take the actions of the generator into
                   2962: account, most notably @code{?dup}, but also words that do not (always)
                   2963: fall through to NEXT.
                   2964: 
1.4       anton    2965: @node TOS Optimization, Produced code, Automatic Generation, Primitives
1.3       anton    2966: @subsection TOS Optimization
                   2967: 
                   2968: An important optimization for stack machine emulators, e.g., Forth
                   2969: engines, is keeping  one or more of the top stack items in
1.4       anton    2970: registers.  If a word has the stack effect @var{in1}...@var{inx} @code{--}
                   2971: @var{out1}...@var{outy}, keeping the top @var{n} items in registers
1.3       anton    2972: @itemize
                   2973: @item
                   2974: is better than keeping @var{n-1} items, if @var{x>=n} and @var{y>=n},
                   2975: due to fewer loads from and stores to the stack.
                   2976: @item is slower than keeping @var{n-1} items, if @var{x<>y} and @var{x<n} and
                   2977: @var{y<n}, due to additional moves between registers.
                   2978: @end itemize
                   2979: 
                   2980: In particular, keeping one item in a register is never a disadvantage,
                   2981: if there are enough registers. Keeping two items in registers is a
                   2982: disadvantage for frequent words like @code{?branch}, constants,
                   2983: variables, literals and @code{i}. Therefore our generator only produces
                   2984: code that keeps zero or one items in registers. The generated C code
                   2985: covers both cases; the selection between these alternatives is made at
                   2986: C-compile time using the switch @code{-DUSE_TOS}. @code{TOS} in the C
                   2987: code for @code{+} is just a simple variable name in the one-item case,
                   2988: otherwise it is a macro that expands into @code{sp[0]}. Note that the
                   2989: GNU C compiler tries to keep simple variables like @code{TOS} in
                   2990: registers, and it usually succeeds, if there are enough registers.
                   2991: 
                   2992: The primitive generator performs the TOS optimization for the
                   2993: floating-point stack, too (@code{-DUSE_FTOS}). For floating-point
                   2994: operations the benefit of this optimization is even larger:
                   2995: floating-point operations take quite long on most processors, but can be
                   2996: performed in parallel with other operations as long as their results are
                   2997: not used. If the FP-TOS is kept in a register, this works. If
                   2998: it is kept on the stack, i.e., in memory, the store into memory has to
                   2999: wait for the result of the floating-point operation, lengthening the
                   3000: execution time of the primitive considerably.
                   3001: 
                   3002: The TOS optimization makes the automatic generation of primitives a
                   3003: bit more complicated. Just replacing all occurrences of @code{sp[0]} by
                   3004: @code{TOS} is not sufficient. There are some special cases to
                   3005: consider:
                   3006: @itemize
                   3007: @item In the case of @code{dup ( w -- w w )} the generator must not
                   3008: eliminate the store to the original location of the item on the stack,
                   3009: if the TOS optimization is turned on.
1.4       anton    3010: @item Primitives with stack effects of the form @code{--}
                   3011: @var{out1}...@var{outy} must store the TOS to the stack at the start.
                   3012: Likewise, primitives with the stack effect @var{in1}...@var{inx} @code{--}
1.3       anton    3013: must load the TOS from the stack at the end. But for the null stack
                   3014: effect @code{--} no stores or loads should be generated.
                   3015: @end itemize
                   3016: 
1.4       anton    3017: @node Produced code,  , TOS Optimization, Primitives
1.3       anton    3018: @subsection Produced code
                   3019: 
                   3020: To see what assembly code is produced for the primitives on your machine
                   3021: with your compiler and your flag settings, type @code{make engine.s} and
1.4       anton    3022: look at the resulting file @file{engine.s}.
1.3       anton    3023: 
1.4       anton    3024: @node System Architecture,  , Primitives, Internals
1.3       anton    3025: @section System Architecture
                   3026: 
                   3027: Our Forth system consists not only of primitives, but also of
                   3028: definitions written in Forth. Since the Forth compiler itself belongs
                   3029: to those definitions, it is not possible to start the system with the
                   3030: primitives and the Forth source alone. Therefore we provide the Forth
                   3031: code as an image file in nearly executable form. At the start of the
                   3032: system a C routine loads the image file into memory, sets up the
                   3033: memory (stacks etc.) according to information in the image file, and
                   3034: starts executing Forth code.
                   3035: 
                   3036: The image file format is a compromise between the goals of making it
                   3037: easy to generate image files and making them portable. The easiest way
                   3038: to generate an image file is to just generate a memory dump. However,
                   3039: this kind of image file cannot be used on a different machine, or on
                   3040: the next version of the engine on the same machine, it even might not
                   3041: work with the same engine compiled by a different version of the C
                   3042: compiler. We would like to have as few versions of the image file as
                   3043: possible, because we do not want to distribute many versions of the
                   3044: same image file, and to make it easy for the users to use their image
                   3045: files on many machines. We currently need to create a different image
                   3046: file for machines with different cell sizes and different byte order
                   3047: (little- or big-endian)@footnote{We consider adding information to the
                   3048: image file that enables the loader to change the byte order.}.
                   3049: 
                   3050: Forth code that is going to end up in a portable image file has to
1.4       anton    3051: comply to some restrictions: addresses have to be stored in memory with
                   3052: special words (@code{A!}, @code{A,}, etc.) in order to make the code
                   3053: relocatable. Cells, floats, etc., have to be stored at the natural
                   3054: alignment boundaries@footnote{E.g., store floats (8 bytes) at an address
                   3055: dividable by~8. This happens automatically in our system when you use
                   3056: the ANS Forth alignment words.}, in order to avoid alignment faults on
                   3057: machines with stricter alignment. The image file is produced by a
                   3058: metacompiler (@file{cross.fs}).
1.3       anton    3059: 
                   3060: So, unlike the image file of Mitch Bradleys @code{cforth}, our image
                   3061: file is not directly executable, but has to undergo some manipulations
                   3062: during loading. Address relocation is performed at image load-time, not
                   3063: at run-time. The loader also has to replace tokens standing for
                   3064: primitive calls with the appropriate code-field addresses (or code
                   3065: addresses in the case of direct threading).
1.4       anton    3066: 
                   3067: @node Bugs, Pedigree, Internals, Top
                   3068: @chapter Bugs
                   3069: 
                   3070: @node Pedigree, Word Index, Bugs, Top
                   3071: @chapter Pedigree
                   3072: 
                   3073: @node Word Index, Node Index, Pedigree, Top
                   3074: @chapter Word Index
                   3075: 
                   3076: @node Node Index,  , Word Index, Top
                   3077: @chapter Node Index
1.1       anton    3078: 
                   3079: @contents
                   3080: @bye
                   3081: 

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