Annotation of gforth/gforth.ds, revision 1.3

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.)
                      4: @setfilename gforth-info
                      5: @settitle GNU Forth Manual
                      6: @setchapternewpage odd
                      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:      
                     18:      @ignore
                     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:      
                     24:      @end ignore
                     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
                     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
                     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: 
                    256: @node Words,  , Invocation, Top
                    257: @chapter Forth Words
                    258: 
                    259: @menu
                    260: * Notation::
                    261: * Arithmetic::
                    262: * Stack Manipulation::
                    263: * Memory access::
                    264: * Control Structures::
                    265: * Local Variables::
                    266: * Defining Words::
                    267: * Vocabularies::
                    268: * Files::
                    269: * Blocks::
                    270: * Other I/O::
                    271: * Programming Tools::
                    272: @end menu
                    273: 
                    274: @node Notation, Arithmetic, Words, Words
                    275: @section Notation
                    276: 
                    277: The Forth words are described in this section in the glossary notation
                    278: that has become a de-facto standard for Forth texts, i.e.
                    279: 
                    280: @quotation
                    281: @var{word}     @var{Stack effect}   @var{wordset}   @var{pronunciation}
                    282: @var{Description}
                    283: @end quotation
                    284: 
                    285: @table @var
                    286: @item word
                    287: The name of the word. BTW, GNU Forth is case insensitive, so you can
                    288: type the words in in lower case.
                    289: 
                    290: @item Stack effect
                    291: The stack effect is written in the notation @code{@var{before} --
                    292: @var{after}}, where @var{before} and @var{after} describe the top of
                    293: stack entries before and after the execution of the word. The rest of
                    294: the stack is not touched by the word. The top of stack is rightmost,
                    295: i.e., a stack sequence is written as it is typed in. Note that GNU Forth
                    296: uses a separate floating point stack, but a unified stack
                    297: notation. Also, return stack effects are not shown in @var{stack
                    298: effect}, but in @var{Description}. The name of a stack item describes
                    299: the type and/or the function of the item. See below for a discussion of
                    300: the types.
                    301: 
                    302: @item pronunciation
                    303: How the word is pronounced
                    304: 
                    305: @item wordset
                    306: The ANS Forth standard is divided into several wordsets. A standard
                    307: system need not support all of them. So, the fewer wordsets your program
                    308: uses the more portable it will be in theory. However, we suspect that
                    309: most ANS Forth systems on personal machines will feature all
                    310: wordsets. Words that are not defined in the ANS standard have
                    311: @code{gforth} as wordset.
                    312: 
                    313: @item Description
                    314: A description of the behaviour of the word.
                    315: @end table
                    316: 
                    317: The name of a stack item corresponds in the following way with its type:
                    318: 
                    319: @table @code
                    320: @item name starts with
                    321: Type
                    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: 
                    356: @node Arithmetic,  , Notation, Words
                    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
                    366: former).
                    367: 
                    368: @subsection Single precision
                    369: doc-+
                    370: doc--
                    371: doc-*
                    372: doc-/
                    373: doc-mod
                    374: doc-/mod
                    375: doc-negate
                    376: doc-abs
                    377: doc-min
                    378: doc-max
                    379: 
                    380: @subsection Bitwise operations
                    381: doc-and
                    382: doc-or
                    383: doc-xor
                    384: doc-invert
                    385: doc-2*
                    386: doc-2/
                    387: 
                    388: @subsection Mixed precision
                    389: doc-m+
                    390: doc-*/
                    391: doc-*/mod
                    392: doc-m*
                    393: doc-um*
                    394: doc-m*/
                    395: doc-um/mod
                    396: doc-fm/mod
                    397: doc-sm/rem
                    398: 
                    399: @subsection Double precision
                    400: doc-d+
                    401: doc-d-
                    402: doc-dnegate
                    403: doc-dabs
                    404: doc-dmin
                    405: doc-dmax
                    406: 
                    407: @node Stack Manipulation,,,
                    408: @section Stack Manipulation
                    409: 
                    410: gforth has a data stack (aka parameter stack) for characters, cells,
                    411: addresses, and double cells, a floating point stack for floating point
                    412: numbers, a return stack for storing the return addresses of colon
                    413: definitions and other data, and a locals stack for storing local
                    414: variables. Note that while every sane Forth has a separate floating
                    415: point stack, this is not strictly required; an ANS Forth system could
                    416: theoretically keep floating point numbers on the data stack. As an
                    417: additional difficulty, you don't know how many cells a floating point
                    418: number takes. It is reportedly possible to write words in a way that
                    419: they work also for a unified stack model, but we do not recommend trying
                    420: it. Also, a Forth system is allowed to keep the local variables on the
                    421: return stack. This is reasonable, as local variables usually eliminate
                    422: the need to use the return stack explicitly. So, if you want to produce
                    423: a standard complying program and if you are using local variables in a
                    424: word, forget about return stack manipulations in that word (see the
                    425: standard document for the exact rules).
                    426: 
                    427: @subsection Data stack
                    428: doc-drop
                    429: doc-nip
                    430: doc-dup
                    431: doc-over
                    432: doc-tuck
                    433: doc-swap
                    434: doc-rot
                    435: doc--rot
                    436: doc-?dup
                    437: doc-pick
                    438: doc-roll
                    439: doc-2drop
                    440: doc-2nip
                    441: doc-2dup
                    442: doc-2over
                    443: doc-2tuck
                    444: doc-2swap
                    445: doc-2rot
                    446: 
                    447: @subsection Floating point stack
                    448: doc-fdrop
                    449: doc-fnip
                    450: doc-fdup
                    451: doc-fover
                    452: doc-ftuck
                    453: doc-fswap
                    454: doc-frot
                    455: 
                    456: @subsection Return stack
                    457: doc->r
                    458: doc-r>
                    459: doc-r@
                    460: doc-rdrop
                    461: doc-2>r
                    462: doc-2r>
                    463: doc-2r@
                    464: doc-2rdrop
                    465: 
                    466: @subsection Locals stack
                    467: 
                    468: @subsection Stack pointer manipulation
                    469: doc-sp@
                    470: doc-sp!
                    471: doc-fp@
                    472: doc-fp!
                    473: doc-rp@
                    474: doc-rp!
                    475: doc-lp@
                    476: doc-lp!
                    477: 
                    478: @node Memory access
                    479: @section Memory access
                    480: 
                    481: @subsection Stack-Memory transfers
                    482: 
                    483: doc-@
                    484: doc-!
                    485: doc-+!
                    486: doc-c@
                    487: doc-c!
                    488: doc-2@
                    489: doc-2!
                    490: doc-f@
                    491: doc-f!
                    492: doc-sf@
                    493: doc-sf!
                    494: doc-df@
                    495: doc-df!
                    496: 
                    497: @subsection Address arithmetic
                    498: 
                    499: ANS Forth does not specify the sizes of the data types. Instead, it
                    500: offers a number of words for computing sizes and doing address
                    501: arithmetic. Basically, address arithmetic is performed in terms of
                    502: address units (aus); on most systems the address unit is one byte. Note
                    503: that a character may have more than one au, so @code{chars} is no noop
                    504: (on systems where it is a noop, it compiles to nothing).
                    505: 
                    506: ANS Forth also defines words for aligning addresses for specific
                    507: addresses. Many computers require that accesses to specific data types
                    508: must only occur at specific addresses; e.g., that cells may only be
                    509: accessed at addresses divisible by 4. Even if a machine allows unaligned
                    510: accesses, it can usually perform aligned accesses faster. 
                    511: 
                    512: For the performance-concious: alignment operations are usually only
                    513: necessary during the definition of a data structure, not during the
                    514: (more frequent) accesses to it.
                    515: 
                    516: ANS Forth defines no words for character-aligning addresses. This is not
                    517: an oversight, but reflects the fact that addresses that are not
                    518: char-aligned have no use in the standard and therefore will not be
                    519: created.
                    520: 
                    521: The standard guarantees that addresses returned by @code{CREATE}d words
                    522: are cell-aligned; in addition, gforth guarantees that these addresses
                    523: are aligned for all purposes.
                    524: 
                    525: doc-chars
                    526: doc-char+
                    527: doc-cells
                    528: doc-cell+
                    529: doc-align
                    530: doc-aligned
                    531: doc-floats
                    532: doc-float+
                    533: doc-falign
                    534: doc-faligned
                    535: doc-sfloats
                    536: doc-sfloat+
                    537: doc-sfalign
                    538: doc-sfaligned
                    539: doc-dfloats
                    540: doc-dfloat+
                    541: doc-dfalign
                    542: doc-dfaligned
                    543: doc-address-unit-bits
                    544: 
                    545: @subsection Memory block access
                    546: 
                    547: doc-move
                    548: doc-erase
                    549: 
                    550: While the previous words work on address units, the rest works on
                    551: characters.
                    552: 
                    553: doc-cmove
                    554: doc-cmove>
                    555: doc-fill
                    556: doc-blank
                    557: 
                    558: @node Control Structures
                    559: @section Control Structures
                    560: 
                    561: Control structures in Forth cannot be used in interpret state, only in
                    562: compile state, i.e., in a colon definition. We do not like this
                    563: limitation, but have not seen a satisfying way around it yet, although
                    564: many schemes have been proposed.
                    565: 
                    566: @subsection Selection
                    567: 
                    568: @example
                    569: @var{flag}
                    570: IF
                    571:   @var{code}
                    572: ENDIF
                    573: @end example
                    574: or
                    575: @example
                    576: @var{flag}
                    577: IF
                    578:   @var{code1}
                    579: ELSE
                    580:   @var{code2}
                    581: ENDIF
                    582: @end example
                    583: 
                    584: You can use @code{THEN} instead of {ENDIF}. Indeed, @code{THEN} is
                    585: standard, and @code{ENDIF} is not, although it is quite popular. We
                    586: recommend using @code{ENDIF}, because it is less confusing for people
                    587: who also know other languages (and is not prone to reinforcing negative
                    588: prejudices against Forth in these people). Adding @code{ENDIF} to a
                    589: system that only supplies @code{THEN} is simple:
                    590: @example
                    591: : endif   POSTPONE then ; immediate
                    592: @end example
                    593: 
                    594: [According to @cite{Webster's New Encyclopedic Dictionary}, @dfn{then
                    595: (adv.)}  has the following meanings:
                    596: @quotation
                    597: ... 2b: following next after in order ... 3d: as a necessary consequence
                    598: (if you were there, then you saw them).
                    599: @end quotation
                    600: Forth's @code{THEN} has the meaning 2b, whereas @code{THEN} in Pascal
                    601: and many other programming languages has the meaning 3d.]
                    602: 
                    603: We also provide the words @code{?dup-if} and @code{?dup-0=-if}, so you
                    604: can avoid using @code{?dup}.
                    605: 
                    606: @example
                    607: @var{n}
                    608: CASE
                    609:   @var{n1} OF @var{code1} ENDOF
                    610:   @var{n2} OF @var{code2} ENDOF
                    611:   @dots
                    612: ENDCASE
                    613: @end example
                    614: 
                    615: Executes the first @var{codei}, where the @var{ni} is equal to
                    616: @var{n}. A default case can be added by simply writing the code after
                    617: the last @code{ENDOF}. It may use @var{n}, which is on top of the stack,
                    618: but must not consume it.
                    619: 
                    620: @subsection Simple Loops
                    621: 
                    622: @example
                    623: BEGIN
                    624:   @var{code1}
                    625:   @var{flag}
                    626: WHILE
                    627:   @var{code2}
                    628: REPEAT
                    629: @end example
                    630: 
                    631: @var{code1} is executed and @var{flag} is computed. If it is true,
                    632: @var{code2} is executed and the loop is restarted; If @var{flag} is false, execution continues after the @code{REPEAT}.
                    633: 
                    634: @example
                    635: BEGIN
                    636:   @var{code}
                    637:   @var{flag}
                    638: UNTIL
                    639: @end example
                    640: 
                    641: @var{code} is executed. The loop is restarted if @code{flag} is false.
                    642: 
                    643: @example
                    644: BEGIN
                    645:   @var{code}
                    646: AGAIN
                    647: @end example
                    648: 
                    649: This is an endless loop.
                    650: 
                    651: @subsection Counted Loops
                    652: 
                    653: The basic counted loop is:
                    654: @example
                    655: @var{limit} @var{start}
                    656: ?DO
                    657:   @var{body}
                    658: LOOP
                    659: @end example
                    660: 
                    661: This performs one iteration for every integer, starting from @var{start}
                    662: and up to, but excluding @var{limit}. The counter, aka index, can be
                    663: accessed with @code{i}. E.g., the loop
                    664: @example
                    665: 10 0 ?DO
                    666:   i .
                    667: LOOP
                    668: @end example
                    669: prints
                    670: @example
                    671: 0 1 2 3 4 5 6 7 8 9
                    672: @end example
                    673: The index of the innermost loop can be accessed with @code{i}, the index
                    674: of the next loop with @code{j}, and the index of the third loop with
                    675: @code{k}.
                    676: 
                    677: The loop control data are kept on the return stack, so there are some
                    678: restrictions on mixing return stack accesses and counted loop
                    679: words. E.g., if you put values on the return stack outside the loop, you
                    680: cannot read them inside the loop. If you put values on the return stack
                    681: within a loop, you have to remove them before the end of the loop and
                    682: before accessing the index of the loop.
                    683: 
                    684: There are several variations on the counted loop:
                    685: 
                    686: @code{LEAVE} leaves the innermost counted loop immediately.
                    687: 
                    688: @code{LOOP} can be replaced with @code{@var{n} +LOOP}; this updates the
                    689: index by @var{n} instead of by 1. The loop is terminated when the border
                    690: between @var{limit-1} and @var{limit} is crossed. E.g.:
                    691: 
1.2       anton     692: @code{4 0 ?DO  i .  2 +LOOP}   prints @code{0 2}
1.1       anton     693: 
1.2       anton     694: @code{4 1 ?DO  i .  2 +LOOP}   prints @code{1 3}
1.1       anton     695: 
                    696: The behaviour of @code{@var{n} +LOOP} is peculiar when @var{n} is negative:
                    697: 
1.2       anton     698: @code{-1 0 ?DO  i .  -1 +LOOP}  prints @code{0 -1}
1.1       anton     699: 
1.2       anton     700: @code{ 0 0 ?DO  i .  -1 +LOOP}  prints nothing
1.1       anton     701: 
                    702: Therefore we recommend avoiding using @code{@var{n} +LOOP} with negative
                    703: @var{n}. One alternative is @code{@var{n} S+LOOP}, where the negative
                    704: case behaves symmetrical to the positive case:
                    705: 
1.2       anton     706: @code{-2 0 ?DO  i .  -1 +LOOP}  prints @code{0 -1}
1.1       anton     707: 
1.2       anton     708: @code{-1 0 ?DO  i .  -1 +LOOP}  prints @code{0}
1.1       anton     709: 
1.2       anton     710: @code{ 0 0 ?DO  i .  -1 +LOOP}  prints nothing
1.1       anton     711: 
1.2       anton     712: The loop is terminated when the border between @var{limit@minus{}sgn(n)} and
1.1       anton     713: @var{limit} is crossed. However, @code{S+LOOP} is not part of the ANS
                    714: Forth standard.
                    715: 
                    716: @code{?DO} can be replaced by @code{DO}. @code{DO} enters the loop even
                    717: when the start and the limit value are equal. We do not recommend using
                    718: @code{DO}. It will just give you maintenance troubles.
                    719: 
                    720: @code{UNLOOP} is used to prepare for an abnormal loop exit, e.g., via
                    721: @code{EXIT}. @code{UNLOOP} removes the loop control parameters from the
                    722: return stack so @code{EXIT} can get to its return address.
                    723: 
                    724: Another counted loop is
                    725: @example
                    726: @var{n}
                    727: FOR
                    728:   @var{body}
                    729: NEXT
                    730: @end example
                    731: This is the preferred loop of native code compiler writers who are too
                    732: lazy to optimize @code{?DO} loops properly. In GNU Forth, this loop
                    733: iterates @var{n+1} times; @code{i} produces values starting with @var{n}
                    734: and ending with 0. Other Forth systems may behave differently, even if
                    735: they support @code{FOR} loops.
                    736: 
1.2       anton     737: @subsection Arbitrary control structures
                    738: 
                    739: ANS Forth permits and supports using control structures in a non-nested
                    740: way. Information about incomplete control structures is stored on the
                    741: control-flow stack. This stack may be implemented on the Forth data
                    742: stack, and this is what we have done in gforth.
                    743: 
                    744: An @i{orig} entry represents an unresolved forward branch, a @i{dest}
                    745: entry represents a backward branch target. A few words are the basis for
                    746: building any control structure possible (except control structures that
                    747: need storage, like calls, coroutines, and backtracking).
                    748: 
1.3     ! anton     749: doc-if
        !           750: doc-ahead
        !           751: doc-then
        !           752: doc-begin
        !           753: doc-until
        !           754: doc-again
        !           755: doc-cs-pick
        !           756: doc-cs-roll
1.2       anton     757: 
                    758: On many systems control-flow stack items take one word, in gforth they
                    759: currently take three (this may change in the future). Therefore it is a
                    760: really good idea to manipulate the control flow stack with
                    761: @code{cs-pick} and @code{cs-roll}, not with data stack manipulation
                    762: words.
                    763: 
                    764: Some standard control structure words are built from these words:
                    765: 
1.3     ! anton     766: doc-else
        !           767: doc-while
        !           768: doc-repeat
1.2       anton     769: 
                    770: Counted loop words constitute a separate group of words:
                    771: 
1.3     ! anton     772: doc-?do
        !           773: doc-do
        !           774: doc-for
        !           775: doc-loop
        !           776: doc-s+loop
        !           777: doc-+loop
        !           778: doc-next
        !           779: doc-leave
        !           780: doc-?leave
        !           781: doc-unloop
        !           782: doc-undo
1.2       anton     783: 
                    784: The standard does not allow using @code{cs-pick} and @code{cs-roll} on
                    785: @i{do-sys}. Our system allows it, but it's your job to ensure that for
                    786: every @code{?DO} etc. there is exactly one @code{UNLOOP} on any path
1.3     ! anton     787: through the definition (@code{LOOP} etc. compile an @code{UNLOOP} on the
        !           788: fall-through path). Also, you have to ensure that all @code{LEAVE}s are
        !           789: resolved (by using one of the loop-ending words or @code{UNDO}).
1.2       anton     790: 
                    791: Another group of control structure words are
                    792: 
1.3     ! anton     793: doc-case
        !           794: doc-endcase
        !           795: doc-of
        !           796: doc-endof
1.2       anton     797: 
                    798: @i{case-sys} and @i{of-sys} cannot be processed using @code{cs-pick} and
                    799: @code{cs-roll}.
                    800: 
1.3     ! anton     801: @subsubsection Programming Style
        !           802: 
        !           803: In order to ensure readability we recommend that you do not create
        !           804: arbitrary control structures directly, but define new control structure
        !           805: words for the control structure you want and use these words in your
        !           806: program.
        !           807: 
        !           808: E.g., instead of writing
        !           809: 
        !           810: @example
        !           811: begin
        !           812:   ...
        !           813: if [ 1 cs-roll ]
        !           814:   ...
        !           815: again then
        !           816: @end example
        !           817: 
        !           818: we recommend defining control structure words, e.g.,
        !           819: 
        !           820: @example
        !           821: : while ( dest -- orig dest )
        !           822:  POSTPONE if
        !           823:  1 cs-roll ; immediate
        !           824: 
        !           825: : repeat ( orig dest -- )
        !           826:  POSTPONE again
        !           827:  POSTPONE then ; immediate
        !           828: @end example
        !           829: 
        !           830: and then using these to create the control structure:
        !           831: 
        !           832: @example
        !           833: begin
        !           834:   ...
        !           835: while
        !           836:   ...
        !           837: repeat
        !           838: @end example
        !           839: 
        !           840: That's much easier to read, isn't it? Of course, @code{BEGIN} and
        !           841: @code{WHILE} are predefined, so in this example it would not be
        !           842: necessary to define them.
        !           843: 
        !           844: @subsection Calls and returns
        !           845: 
        !           846: A definition can be called simply be writing the name of the
        !           847: definition. When the end of the definition is reached, it returns. An earlier return can be forced using
        !           848: 
        !           849: doc-exit
        !           850: 
        !           851: Don't forget to clean up the return stack and @code{UNLOOP} any
        !           852: outstanding @code{?DO}...@code{LOOP}s before @code{EXIT}ing. The
        !           853: primitive compiled by @code{EXIT} is
        !           854: 
        !           855: doc-;s
        !           856: 
        !           857: @subsection Exception Handling
        !           858: 
        !           859: doc-catch
        !           860: doc-throw
        !           861: 
1.1       anton     862: @node Locals
                    863: @section Locals
                    864: 
1.2       anton     865: Local variables can make Forth programming more enjoyable and Forth
                    866: programs easier to read. Unfortunately, the locals of ANS Forth are
                    867: laden with restrictions. Therefore, we provide not only the ANS Forth
                    868: locals wordset, but also our own, more powerful locals wordset (we
                    869: implemented the ANS Forth locals wordset through our locals wordset).
                    870: 
                    871: @menu
                    872: @end menu
                    873: 
                    874: @subsection gforth locals
                    875: 
                    876: Locals can be defined with
                    877: 
                    878: @example
                    879: @{ local1 local2 ... -- comment @}
                    880: @end example
                    881: or
                    882: @example
                    883: @{ local1 local2 ... @}
                    884: @end example
                    885: 
                    886: E.g.,
                    887: @example
                    888: : max @{ n1 n2 -- n3 @}
                    889:  n1 n2 > if
                    890:    n1
                    891:  else
                    892:    n2
                    893:  endif ;
                    894: @end example
                    895: 
                    896: The similarity of locals definitions with stack comments is intended. A
                    897: locals definition often replaces the stack comment of a word. The order
                    898: of the locals corresponds to the order in a stack comment and everything
                    899: after the @code{--} is really a comment.
                    900: 
                    901: This similarity has one disadvantage: It is too easy to confuse locals
                    902: declarations with stack comments, causing bugs and making them hard to
                    903: find. However, this problem can be avoided by appropriate coding
                    904: conventions: Do not use both notations in the same program. If you do,
                    905: they should be distinguished using additional means, e.g. by position.
                    906: 
                    907: The name of the local may be preceded by a type specifier, e.g.,
                    908: @code{F:} for a floating point value:
                    909: 
                    910: @example
                    911: : CX* @{ F: Ar F: Ai F: Br F: Bi -- Cr Ci @}
                    912: \ complex multiplication
                    913:  Ar Br f* Ai Bi f* f-
                    914:  Ar Bi f* Ai Br f* f+ ;
                    915: @end example
                    916: 
                    917: GNU Forth currently supports cells (@code{W:}, @code{W^}), doubles
                    918: (@code{D:}, @code{D^}), floats (@code{F:}, @code{F^}) and characters
                    919: (@code{C:}, @code{C^}) in two flavours: a value-flavoured local (defined
                    920: with @code{W:}, @code{D:} etc.) produces its value and can be changed
                    921: with @code{TO}. A variable-flavoured local (defined with @code{W^} etc.)
                    922: produces its address (which becomes invalid when the variable's scope is
                    923: left). E.g., the standard word @code{emit} can be defined in therms of
                    924: @code{type} like this:
                    925: 
                    926: @example
                    927: : emit @{ C^ char* -- @}
                    928:     char* 1 type ;
                    929: @end example
                    930: 
                    931: A local without type specifier is a @code{W:} local. Both flavours of
                    932: locals are initialized with values from the data or FP stack.
                    933: 
                    934: Currently there is no way to define locals with user-defined data
                    935: structures, but we are working on it.
                    936: 
                    937: GNU Forth allows defining locals everywhere in a colon definition. This poses the following questions:
                    938: 
                    939: @subsubsection Where are locals visible by name?
                    940: 
                    941: Basically, the answer is that locals are visible where you would expect
                    942: it in block-structured languages, and sometimes a little longer. If you
                    943: want to restrict the scope of a local, enclose its definition in
                    944: @code{SCOPE}...@code{ENDSCOPE}.
                    945: 
                    946: doc-scope
                    947: doc-endscope
                    948: 
                    949: These words behave like control structure words, so you can use them
                    950: with @code{CS-PICK} and @code{CS-ROLL} to restrict the scope in
                    951: arbitrary ways.
                    952: 
                    953: If you want a more exact answer to the visibility question, here's the
                    954: basic principle: A local is visible in all places that can only be
                    955: reached through the definition of the local@footnote{In compiler
                    956: construction terminology, all places dominated by the definition of the
                    957: local.}. In other words, it is not visible in places that can be reached
                    958: without going through the definition of the local. E.g., locals defined
                    959: in @code{IF}...@code{ENDIF} are visible until the @code{ENDIF}, locals
                    960: defined in @code{BEGIN}...@code{UNTIL} are visible after the
                    961: @code{UNTIL} (until, e.g., a subsequent @code{ENDSCOPE}).
                    962: 
                    963: The reasoning behind this solution is: We want to have the locals
                    964: visible as long as it is meaningful. The user can always make the
                    965: visibility shorter by using explicit scoping. In a place that can
                    966: only be reached through the definition of a local, the meaning of a
                    967: local name is clear. In other places it is not: How is the local
                    968: initialized at the control flow path that does not contain the
                    969: definition? Which local is meant, if the same name is defined twice in
                    970: two independent control flow paths?
                    971: 
                    972: This should be enough detail for nearly all users, so you can skip the
                    973: rest of this section. If you relly must know all the gory details and
                    974: options, read on.
                    975: 
                    976: In order to implement this rule, the compiler has to know which places
                    977: are unreachable. It knows this automatically after @code{AHEAD},
                    978: @code{AGAIN}, @code{EXIT} and @code{LEAVE}; in other cases (e.g., after
                    979: most @code{THROW}s), you can use the word @code{UNREACHABLE} to tell the
                    980: compiler that the control flow never reaches that place. If
                    981: @code{UNREACHABLE} is not used where it could, the only consequence is
                    982: that the visibility of some locals is more limited than the rule above
                    983: says. If @code{UNREACHABLE} is used where it should not (i.e., if you
                    984: lie to the compiler), buggy code will be produced.
                    985: 
                    986: Another problem with this rule is that at @code{BEGIN}, the compiler
1.3     ! anton     987: does not know which locals will be visible on the incoming
        !           988: back-edge. All problems discussed in the following are due to this
        !           989: ignorance of the compiler (we discuss the problems using @code{BEGIN}
        !           990: loops as examples; the discussion also applies to @code{?DO} and other
1.2       anton     991: loops). Perhaps the most insidious example is:
                    992: @example
                    993: AHEAD
                    994: BEGIN
                    995:   x
                    996: [ 1 CS-ROLL ] THEN
                    997:   { x }
                    998:   ...
                    999: UNTIL
                   1000: @end example
                   1001: 
                   1002: This should be legal according to the visibility rule. The use of
                   1003: @code{x} can only be reached through the definition; but that appears
                   1004: textually below the use.
                   1005: 
                   1006: From this example it is clear that the visibility rules cannot be fully
                   1007: implemented without major headaches. Our implementation treats common
                   1008: cases as advertised and the exceptions are treated in a safe way: The
                   1009: compiler makes a reasonable guess about the locals visible after a
                   1010: @code{BEGIN}; if it is too pessimistic, the
                   1011: user will get a spurious error about the local not being defined; if the
                   1012: compiler is too optimistic, it will notice this later and issue a
                   1013: warning. In the case above the compiler would complain about @code{x}
                   1014: being undefined at its use. You can see from the obscure examples in
                   1015: this section that it takes quite unusual control structures to get the
                   1016: compiler into trouble, and even then it will often do fine.
                   1017: 
                   1018: If the @code{BEGIN} is reachable from above, the most optimistic guess
                   1019: is that all locals visible before the @code{BEGIN} will also be
                   1020: visible after the @code{BEGIN}. This guess is valid for all loops that
                   1021: are entered only through the @code{BEGIN}, in particular, for normal
                   1022: @code{BEGIN}...@code{WHILE}...@code{REPEAT} and
                   1023: @code{BEGIN}...@code{UNTIL} loops and it is implemented in our
                   1024: compiler. When the branch to the @code{BEGIN} is finally generated by
                   1025: @code{AGAIN} or @code{UNTIL}, the compiler checks the guess and
                   1026: warns the user if it was too optimisitic:
                   1027: @example
                   1028: IF
                   1029:   { x }
                   1030: BEGIN
                   1031:   \ x ? 
                   1032: [ 1 cs-roll ] THEN
                   1033:   ...
                   1034: UNTIL
                   1035: @end example
                   1036: 
                   1037: Here, @code{x} lives only until the @code{BEGIN}, but the compiler
                   1038: optimistically assumes that it lives until the @code{THEN}. It notices
                   1039: this difference when it compiles the @code{UNTIL} and issues a
                   1040: warning. The user can avoid the warning, and make sure that @code{x}
                   1041: is not used in the wrong area by using explicit scoping:
                   1042: @example
                   1043: IF
                   1044:   SCOPE
                   1045:   { x }
                   1046:   ENDSCOPE
                   1047: BEGIN
                   1048: [ 1 cs-roll ] THEN
                   1049:   ...
                   1050: UNTIL
                   1051: @end example
                   1052: 
                   1053: Since the guess is optimistic, there will be no spurious error messages
                   1054: about undefined locals.
                   1055: 
                   1056: If the @code{BEGIN} is not reachable from above (e.g., after
                   1057: @code{AHEAD} or @code{EXIT}), the compiler cannot even make an
                   1058: optimistic guess, as the locals visible after the @code{BEGIN} may be
                   1059: defined later. Therefore, the compiler assumes that no locals are
                   1060: visible after the @code{BEGIN}. However, the useer can use
                   1061: @code{ASSUME-LIVE} to make the compiler assume that the same locals are
                   1062: visible at the BEGIN as at the point where the item was created.
                   1063: 
                   1064: doc-assume-live
                   1065: 
                   1066: E.g.,
                   1067: @example
                   1068: { x }
                   1069: AHEAD
                   1070: ASSUME-LIVE
                   1071: BEGIN
                   1072:   x
                   1073: [ 1 CS-ROLL ] THEN
                   1074:   ...
                   1075: UNTIL
                   1076: @end example
                   1077: 
                   1078: Other cases where the locals are defined before the @code{BEGIN} can be
                   1079: handled by inserting an appropriate @code{CS-ROLL} before the
                   1080: @code{ASSUME-LIVE} (and changing the control-flow stack manipulation
                   1081: behind the @code{ASSUME-LIVE}).
                   1082: 
                   1083: Cases where locals are defined after the @code{BEGIN} (but should be
                   1084: visible immediately after the @code{BEGIN}) can only be handled by
                   1085: rearranging the loop. E.g., the ``most insidious'' example above can be
                   1086: arranged into:
                   1087: @example
                   1088: BEGIN
                   1089:   { x }
                   1090:   ... 0=
                   1091: WHILE
                   1092:   x
                   1093: REPEAT
                   1094: @end example
                   1095: 
                   1096: @subsubsection How long do locals live?
                   1097: 
                   1098: The right answer for the lifetime question would be: A local lives at
                   1099: least as long as it can be accessed. For a value-flavoured local this
                   1100: means: until the end of its visibility. However, a variable-flavoured
                   1101: local could be accessed through its address far beyond its visibility
                   1102: scope. Ultimately, this would mean that such locals would have to be
                   1103: garbage collected. Since this entails un-Forth-like implementation
                   1104: complexities, I adopted the same cowardly solution as some other
                   1105: languages (e.g., C): The local lives only as long as it is visible;
                   1106: afterwards its address is invalid (and programs that access it
                   1107: afterwards are erroneous).
                   1108: 
                   1109: @subsubsection Programming Style
                   1110: 
                   1111: The freedom to define locals anywhere has the potential to change
                   1112: programming styles dramatically. In particular, the need to use the
                   1113: return stack for intermediate storage vanishes. Moreover, all stack
                   1114: manipulations (except @code{PICK}s and @code{ROLL}s with run-time
                   1115: determined arguments) can be eliminated: If the stack items are in the
                   1116: wrong order, just write a locals definition for all of them; then
                   1117: write the items in the order you want.
                   1118: 
                   1119: This seems a little far-fetched and eliminating stack manipulations is
                   1120: unlikely to become a conscious programming objective. Still, the
                   1121: number of stack manipulations will be reduced dramatically if local
                   1122: variables are used liberally (e.g., compare @code{max} in \sect{misc}
                   1123: with a traditional implementation of @code{max}).
                   1124: 
                   1125: This shows one potential benefit of locals: making Forth programs more
                   1126: readable. Of course, this benefit will only be realized if the
                   1127: programmers continue to honour the principle of factoring instead of
                   1128: using the added latitude to make the words longer.
                   1129: 
                   1130: Using @code{TO} can and should be avoided.  Without @code{TO},
                   1131: every value-flavoured local has only a single assignment and many
                   1132: advantages of functional languages apply to Forth. I.e., programs are
                   1133: easier to analyse, to optimize and to read: It is clear from the
                   1134: definition what the local stands for, it does not turn into something
                   1135: different later.
                   1136: 
                   1137: E.g., a definition using @code{TO} might look like this:
                   1138: @example
                   1139: : strcmp @{ addr1 u1 addr2 u2 -- n @}
                   1140:  u1 u2 min 0
                   1141:  ?do
                   1142:    addr1 c@ addr2 c@ - ?dup
                   1143:    if
                   1144:      unloop exit
                   1145:    then
                   1146:    addr1 char+ TO addr1
                   1147:    addr2 char+ TO addr2
                   1148:  loop
                   1149:  u1 u2 - ;
                   1150: @end example
                   1151: Here, @code{TO} is used to update @code{addr1} and @code{addr2} at
                   1152: every loop iteration. @code{strcmp} is a typical example of the
                   1153: readability problems of using @code{TO}. When you start reading
                   1154: @code{strcmp}, you think that @code{addr1} refers to the start of the
                   1155: string. Only near the end of the loop you realize that it is something
                   1156: else.
                   1157: 
                   1158: This can be avoided by defining two locals at the start of the loop that
                   1159: are initialized with the right value for the current iteration.
                   1160: @example
                   1161: : strcmp @{ addr1 u1 addr2 u2 -- n @}
                   1162:  addr1 addr2
                   1163:  u1 u2 min 0 
                   1164:  ?do @{ s1 s2 @}
                   1165:    s1 c@ s2 c@ - ?dup 
                   1166:    if
                   1167:      unloop exit
                   1168:    then
                   1169:    s1 char+ s2 char+
                   1170:  loop
                   1171:  2drop
                   1172:  u1 u2 - ;
                   1173: @end example
                   1174: Here it is clear from the start that @code{s1} has a different value
                   1175: in every loop iteration.
                   1176: 
                   1177: @subsubsection Implementation
                   1178: 
                   1179: GNU Forth uses an extra locals stack. The most compelling reason for
                   1180: this is that the return stack is not float-aligned; using an extra stack
                   1181: also eliminates the problems and restrictions of using the return stack
                   1182: as locals stack. Like the other stacks, the locals stack grows toward
                   1183: lower addresses. A few primitives allow an efficient implementation:
                   1184: 
                   1185: doc-@local#
                   1186: doc-f@local#
                   1187: doc-laddr#
                   1188: doc-lp+!#
                   1189: doc-lp!
                   1190: doc->l
                   1191: doc-f>l
                   1192: 
                   1193: In addition to these primitives, some specializations of these
                   1194: primitives for commonly occurring inline arguments are provided for
                   1195: efficiency reasons, e.g., @code{@@local0} as specialization of
                   1196: @code{@@local#} for the inline argument 0. The following compiling words
                   1197: compile the right specialized version, or the general version, as
                   1198: appropriate:
                   1199: 
                   1200: doc-compile-@@local
                   1201: doc-compile-f@@local
                   1202: doc-compile-lp+!
                   1203: 
                   1204: Combinations of conditional branches and @code{lp+!#} like
                   1205: @code{?branch-lp+!#} (the locals pointer is only changed if the branch
                   1206: is taken) are provided for efficiency and correctness in loops.
                   1207: 
                   1208: A special area in the dictionary space is reserved for keeping the
                   1209: local variable names. @code{@{} switches the dictionary pointer to this
                   1210: area and @code{@}} switches it back and generates the locals
                   1211: initializing code. @code{W:} etc.@ are normal defining words. This
                   1212: special area is cleared at the start of every colon definition.
                   1213: 
                   1214: A special feature of GNU Forths dictionary is used to implement the
                   1215: definition of locals without type specifiers: every wordlist (aka
                   1216: vocabulary) has its own methods for searching
                   1217: etc. (@xref{dictionary}). For the present purpose we defined a wordlist
                   1218: with a special search method: When it is searched for a word, it
                   1219: actually creates that word using @code{W:}. @code{@{} changes the search
                   1220: order to first search the wordlist containing @code{@}}, @code{W:} etc.,
                   1221: and then the wordlist for defining locals without type specifiers.
                   1222: 
                   1223: The lifetime rules support a stack discipline within a colon
                   1224: definition: The lifetime of a local is either nested with other locals
                   1225: lifetimes or it does not overlap them.
                   1226: 
                   1227: At @code{BEGIN}, @code{IF}, and @code{AHEAD} no code for locals stack
                   1228: pointer manipulation is generated. Between control structure words
                   1229: locals definitions can push locals onto the locals stack. @code{AGAIN}
                   1230: is the simplest of the other three control flow words. It has to
                   1231: restore the locals stack depth of the corresponding @code{BEGIN}
                   1232: before branching. The code looks like this:
                   1233: @format
                   1234: @code{lp+!#} current-locals-size @minus{} dest-locals-size
                   1235: @code{branch} <begin>
                   1236: @end format
                   1237: 
                   1238: @code{UNTIL} is a little more complicated: If it branches back, it
                   1239: must adjust the stack just like @code{AGAIN}. But if it falls through,
                   1240: the locals stack must not be changed. The compiler generates the
                   1241: following code:
                   1242: @format
                   1243: @code{?branch-lp+!#} <begin> current-locals-size @minus{} dest-locals-size
                   1244: @end format
                   1245: The locals stack pointer is only adjusted if the branch is taken.
                   1246: 
                   1247: @code{THEN} can produce somewhat inefficient code:
                   1248: @format
                   1249: @code{lp+!#} current-locals-size @minus{} orig-locals-size
                   1250: <orig target>:
                   1251: @code{lp+!#} orig-locals-size @minus{} new-locals-size
                   1252: @end format
                   1253: The second @code{lp+!#} adjusts the locals stack pointer from the
                   1254: level at the {\em orig} point to the level after the @code{THEN}. The
                   1255: first @code{lp+!#} adjusts the locals stack pointer from the current
                   1256: level to the level at the orig point, so the complete effect is an
                   1257: adjustment from the current level to the right level after the
                   1258: @code{THEN}.
                   1259: 
                   1260: In a conventional Forth implementation a dest control-flow stack entry
                   1261: is just the target address and an orig entry is just the address to be
                   1262: patched. Our locals implementation adds a wordlist to every orig or dest
                   1263: item. It is the list of locals visible (or assumed visible) at the point
                   1264: described by the entry. Our implementation also adds a tag to identify
                   1265: the kind of entry, in particular to differentiate between live and dead
                   1266: (reachable and unreachable) orig entries.
                   1267: 
                   1268: A few unusual operations have to be performed on locals wordlists:
                   1269: 
                   1270: doc-common-list
                   1271: doc-sub-list?
                   1272: doc-list-size
                   1273: 
                   1274: Several features of our locals wordlist implementation make these
                   1275: operations easy to implement: The locals wordlists are organised as
                   1276: linked lists; the tails of these lists are shared, if the lists
                   1277: contain some of the same locals; and the address of a name is greater
                   1278: than the address of the names behind it in the list.
                   1279: 
                   1280: Another important implementation detail is the variable
                   1281: @code{dead-code}. It is used by @code{BEGIN} and @code{THEN} to
                   1282: determine if they can be reached directly or only through the branch
                   1283: that they resolve. @code{dead-code} is set by @code{UNREACHABLE},
                   1284: @code{AHEAD}, @code{EXIT} etc., and cleared at the start of a colon
                   1285: definition, by @code{BEGIN} and usually by @code{THEN}.
                   1286: 
                   1287: Counted loops are similar to other loops in most respects, but
                   1288: @code{LEAVE} requires special attention: It performs basically the same
                   1289: service as @code{AHEAD}, but it does not create a control-flow stack
                   1290: entry. Therefore the information has to be stored elsewhere;
                   1291: traditionally, the information was stored in the target fields of the
                   1292: branches created by the @code{LEAVE}s, by organizing these fields into a
                   1293: linked list. Unfortunately, this clever trick does not provide enough
                   1294: space for storing our extended control flow information. Therefore, we
                   1295: introduce another stack, the leave stack. It contains the control-flow
                   1296: stack entries for all unresolved @code{LEAVE}s.
                   1297: 
                   1298: Local names are kept until the end of the colon definition, even if
                   1299: they are no longer visible in any control-flow path. In a few cases
                   1300: this may lead to increased space needs for the locals name area, but
                   1301: usually less than reclaiming this space would cost in code size.
                   1302: 
                   1303: 
                   1304: @subsection ANS Forth locals
                   1305: 
                   1306: The ANS Forth locals wordset does not define a syntax for locals, but
                   1307: words that make it possible to define various syntaxes. One of the
                   1308: possible syntaxes is a subset of the syntax we used in the gforth locals
                   1309: wordset, i.e.:
                   1310: 
                   1311: @example
                   1312: @{ local1 local2 ... -- comment @}
                   1313: @end example
                   1314: or
                   1315: @example
                   1316: @{ local1 local2 ... @}
                   1317: @end example
                   1318: 
                   1319: The order of the locals corresponds to the order in a stack comment. The
                   1320: restrictions are:
1.1       anton    1321: 
1.2       anton    1322: @itemize @bullet
                   1323: @item
                   1324: Locals can only be cell-sized values (no type specifers are allowed).
                   1325: @item
                   1326: Locals can be defined only outside control structures.
                   1327: @item
                   1328: Locals can interfere with explicit usage of the return stack. For the
                   1329: exact (and long) rules, see the standard. If you don't use return stack
                   1330: accessing words in a definition using locals, you will we all right. The
                   1331: purpose of this rule is to make locals implementation on the return
                   1332: stack easier.
                   1333: @item
                   1334: The whole definition must be in one line.
                   1335: @end itemize
                   1336: 
                   1337: Locals defined in this way behave like @code{VALUE}s
                   1338: (@xref{values}). I.e., they are initialized from the stack. Using their
                   1339: name produces their value. Their value can be changed using @code{TO}.
                   1340: 
                   1341: Since this syntax is supported by gforth directly, you need not do
                   1342: anything to use it. If you want to port a program using this syntax to
                   1343: another ANS Forth system, use @file{anslocal.fs} to implement the syntax
                   1344: on the other system.
                   1345: 
                   1346: Note that a syntax shown in the standard, section A.13 looks
                   1347: similar, but is quite different in having the order of locals
                   1348: reversed. Beware!
                   1349: 
                   1350: The ANS Forth locals wordset itself consists of the following word
                   1351: 
                   1352: doc-(local)
                   1353: 
                   1354: The ANS Forth locals extension wordset defines a syntax, but it is so
                   1355: awful that we strongly recommend not to use it. We have implemented this
                   1356: syntax to make porting to gforth easy, but do not document it here. The
                   1357: problem with this syntax is that the locals are defined in an order
                   1358: reversed with respect to the standard stack comment notation, making
                   1359: programs harder to read, and easier to misread and miswrite. The only
                   1360: merit of this syntax is that it is easy to implement using the ANS Forth
                   1361: locals wordset.
1.3     ! anton    1362: 
        !          1363: @node Internals
        !          1364: @chapter Internals
        !          1365: 
        !          1366: Reading this section is not necessary for programming with gforth. It
        !          1367: should be helpful for finding your way in the gforth sources.
        !          1368: 
        !          1369: @section Portability
        !          1370: 
        !          1371: One of the main goals of the effort is availability across a wide range
        !          1372: of personal machines. fig-Forth, and, to a lesser extent, F83, achieved
        !          1373: this goal by manually coding the engine in assembly language for several
        !          1374: then-popular processors. This approach is very labor-intensive and the
        !          1375: results are short-lived due to progress in computer architecture.
        !          1376: 
        !          1377: Others have avoided this problem by coding in C, e.g., Mitch Bradley
        !          1378: (cforth), Mikael Patel (TILE) and Dirk Zoller (pfe). This approach is
        !          1379: particularly popular for UNIX-based Forths due to the large variety of
        !          1380: architectures of UNIX machines. Unfortunately an implementation in C
        !          1381: does not mix well with the goals of efficiency and with using
        !          1382: traditional techniques: Indirect or direct threading cannot be expressed
        !          1383: in C, and switch threading, the fastest technique available in C, is
        !          1384: significantly slower. Another problem with C is that it's very
        !          1385: cumbersome to express double integer arithmetic.
        !          1386: 
        !          1387: Fortunately, there is a portable language that does not have these
        !          1388: limitations: GNU C, the version of C processed by the GNU C compiler
        !          1389: (@pxref{C Extensions, , Extensions to the C Language Family, gcc.info,
        !          1390: GNU C Manual}). Its labels as values feature (@pxref{Labels as Values, ,
        !          1391: Labels as Values, gcc.info, GNU C Manual}) makes direct and indirect
        !          1392: threading possible, its @code{long long} type (@pxref{Long Long, ,
        !          1393: Double-Word Integers, gcc.info, GNU C Manual}) corresponds to Forths
        !          1394: double numbers. GNU C is available for free on all important (and many
        !          1395: unimportant) UNIX machines, VMS, 80386s running MS-DOS, the Amiga, and
        !          1396: the Atari ST, so a Forth written in GNU C can run on all these
        !          1397: machines@footnote{Due to Apple's look-and-feel lawsuit it is not
        !          1398: available on the Mac (@pxref{Boycott, , Protect Your Freedom--Fight
        !          1399: ``Look And Feel'', gcc.info, GNU C Manual}).}.
        !          1400: 
        !          1401: Writing in a portable language has the reputation of producing code that
        !          1402: is slower than assembly. For our Forth engine we repeatedly looked at
        !          1403: the code produced by the compiler and eliminated most compiler-induced
        !          1404: inefficiencies by appropriate changes in the source-code.
        !          1405: 
        !          1406: However, register allocation cannot be portably influenced by the
        !          1407: programmer, leading to some inefficiencies on register-starved
        !          1408: machines. We use explicit register declarations (@pxref{Explicit Reg
        !          1409: Vars, , Variables in Specified Registers, gcc.info, GNU C Manual}) to
        !          1410: improve the speed on some machines. They are turned on by using the
        !          1411: @code{gcc} switch @code{-DFORCE_REG}. Unfortunately, this feature not
        !          1412: only depends on the machine, but also on the compiler version: On some
        !          1413: machines some compiler versions produce incorrect code when certain
        !          1414: explicit register declarations are used. So by default
        !          1415: @code{-DFORCE_REG} is not used.
        !          1416: 
        !          1417: @section Threading
        !          1418: 
        !          1419: GNU C's labels as values extension (available since @code{gcc-2.0},
        !          1420: @pxref{Labels as Values, , Labels as Values, gcc.info, GNU C Manual})
        !          1421: makes it possible to take the address of @var{label} by writing
        !          1422: @code{&&@var{label}}.  This address can then be used in a statement like
        !          1423: @code{goto *@var{address}}. I.e., @code{goto *&&x} is the same as
        !          1424: @code{goto x}.
        !          1425: 
        !          1426: With this feature an indirect threaded NEXT looks like:
        !          1427: @example
        !          1428: cfa = *ip++;
        !          1429: ca = *cfa;
        !          1430: goto *ca;
        !          1431: @end example
        !          1432: For those unfamiliar with the names: @code{ip} is the Forth instruction
        !          1433: pointer; the @code{cfa} (code-field address) corresponds to ANS Forths
        !          1434: execution token and points to the code field of the next word to be
        !          1435: executed; The @code{ca} (code address) fetched from there points to some
        !          1436: executable code, e.g., a primitive or the colon definition handler
        !          1437: @code{docol}.
        !          1438: 
        !          1439: Direct threading is even simpler:
        !          1440: @example
        !          1441: ca = *ip++;
        !          1442: goto *ca;
        !          1443: @end example
        !          1444: 
        !          1445: Of course we have packaged the whole thing neatly in macros called
        !          1446: @code{NEXT} and @code{NEXT1} (the part of NEXT after fetching the cfa).
        !          1447: 
        !          1448: @subsection Scheduling
        !          1449: 
        !          1450: There is a little complication: Pipelined and superscalar processors,
        !          1451: i.e., RISC and some modern CISC machines can process independent
        !          1452: instructions while waiting for the results of an instruction. The
        !          1453: compiler usually reorders (schedules) the instructions in a way that
        !          1454: achieves good usage of these delay slots. However, on our first tries
        !          1455: the compiler did not do well on scheduling primitives. E.g., for
        !          1456: @code{+} implemented as
        !          1457: @example
        !          1458: n=sp[0]+sp[1];
        !          1459: sp++;
        !          1460: sp[0]=n;
        !          1461: NEXT;
        !          1462: @end example
        !          1463: the NEXT comes strictly after the other code, i.e., there is nearly no
        !          1464: scheduling. After a little thought the problem becomes clear: The
        !          1465: compiler cannot know that sp and ip point to different addresses (and
        !          1466: the version of @code{gcc} we used would not know it even if it could),
        !          1467: so it could not move the load of the cfa above the store to the
        !          1468: TOS. Indeed the pointers could be the same, if code on or very near the
        !          1469: top of stack were executed. In the interest of speed we chose to forbid
        !          1470: this probably unused ``feature'' and helped the compiler in scheduling:
        !          1471: NEXT is divided into the loading part (@code{NEXT_P1}) and the goto part
        !          1472: (@code{NEXT_P2}). @code{+} now looks like:
        !          1473: @example
        !          1474: n=sp[0]+sp[1];
        !          1475: sp++;
        !          1476: NEXT_P1;
        !          1477: sp[0]=n;
        !          1478: NEXT_P2;
        !          1479: @end example
        !          1480: This can be scheduled optimally by the compiler (see \sect{TOS}).
        !          1481: 
        !          1482: This division can be turned off with the switch @code{-DCISC_NEXT}. This
        !          1483: switch is on by default on machines that do not profit from scheduling
        !          1484: (e.g., the 80386), in order to preserve registers.
        !          1485: 
        !          1486: @subsection Direct or Indirect Threaded?
        !          1487: 
        !          1488: Both! After packaging the nasty details in macro definitions we
        !          1489: realized that we could switch between direct and indirect threading by
        !          1490: simply setting a compilation flag (@code{-DDIRECT_THREADED}) and
        !          1491: defining a few machine-specific macros for the direct-threading case.
        !          1492: On the Forth level we also offer access words that hide the
        !          1493: differences between the threading methods (@pxref{Threading Words}).
        !          1494: 
        !          1495: Indirect threading is implemented completely
        !          1496: machine-independently. Direct threading needs routines for creating
        !          1497: jumps to the executable code (e.g. to docol or dodoes). These routines
        !          1498: are inherently machine-dependent, but they do not amount to many source
        !          1499: lines. I.e., even porting direct threading to a new machine is a small
        !          1500: effort.
        !          1501: 
        !          1502: @subsection DOES>
        !          1503: One of the most complex parts of a Forth engine is @code{dodoes}, i.e.,
        !          1504: the chunk of code executed by every word defined by a
        !          1505: @code{CREATE}...@code{DOES>} pair. The main problem here is: How to find
        !          1506: the Forth code to be executed, i.e. the code after the @code{DOES>} (the
        !          1507: DOES-code)? There are two solutions:
        !          1508: 
        !          1509: In fig-Forth the code field points directly to the dodoes and the
        !          1510: DOES-code address is stored in the cell after the code address
        !          1511: (i.e. at cfa cell+). It may seem that this solution is illegal in the
        !          1512: Forth-79 and all later standards, because in fig-Forth this address
        !          1513: lies in the body (which is illegal in these standards). However, by
        !          1514: making the code field larger for all words this solution becomes legal
        !          1515: again. We use this approach for the indirect threaded version. Leaving
        !          1516: a cell unused in most words is a bit wasteful, but on the machines we
        !          1517: are targetting this is hardly a problem. The other reason for having a
        !          1518: code field size of two cells is to avoid having different image files
        !          1519: for direct and indirect threaded systems (@pxref{image-format}).
        !          1520: 
        !          1521: The other approach is that the code field points or jumps to the cell
        !          1522: after @code{DOES}. In this variant there is a jump to @code{dodoes} at
        !          1523: this address. @code{dodoes} can then get the DOES-code address by
        !          1524: computing the code address, i.e., the address of the jump to dodoes,
        !          1525: and add the length of that jump field. A variant of this is to have a
        !          1526: call to @code{dodoes} after the @code{DOES>}; then the return address
        !          1527: (which can be found in the return register on RISCs) is the DOES-code
        !          1528: address. Since the two cells available in the code field are usually
        !          1529: used up by the jump to the code address in direct threading, we use
        !          1530: this approach for direct threading. We did not want to add another
        !          1531: cell to the code field.
        !          1532: 
        !          1533: @section Primitives
        !          1534: 
        !          1535: @subsection Automatic Generation
        !          1536: 
        !          1537: Since the primitives are implemented in a portable language, there is no
        !          1538: longer any need to minimize the number of primitives. On the contrary,
        !          1539: having many primitives is an advantage: speed. In order to reduce the
        !          1540: number of errors in primitives and to make programming them easier, we
        !          1541: provide a tool, the primitive generator (@file{prims2x.fs}), that
        !          1542: automatically generates most (and sometimes all) of the C code for a
        !          1543: primitive from the stack effect notation.  The source for a primitive
        !          1544: has the following form:
        !          1545: 
        !          1546: @format
        !          1547: @var{Forth-name}       @var{stack-effect}      @var{category}  [@var{pronounc.}]
        !          1548: [@code{""}@var{glossary entry}@code{""}]
        !          1549: @var{C code}
        !          1550: [@code{:}
        !          1551: @var{Forth code}]
        !          1552: @end format
        !          1553: 
        !          1554: The items in brackets are optional. The category and glossary fields
        !          1555: are there for generating the documentation, the Forth code is there
        !          1556: for manual implementations on machines without GNU C. E.g., the source
        !          1557: for the primitive @code{+} is:
        !          1558: @example
        !          1559: +    n1 n2 -- n    core    plus
        !          1560: n = n1+n2;
        !          1561: @end example
        !          1562: 
        !          1563: This looks like a specification, but in fact @code{n = n1+n2} is C
        !          1564: code. Our primitive generation tool extracts a lot of information from
        !          1565: the stack effect notations@footnote{We use a one-stack notation, even
        !          1566: though we have separate data and floating-point stacks; The separate
        !          1567: notation can be generated easily from the unified notation.}: The number
        !          1568: of items popped from and pushed on the stack, their type, and by what
        !          1569: name they are referred to in the C code. It then generates a C code
        !          1570: prelude and postlude for each primitive. The final C code for @code{+}
        !          1571: looks like this:
        !          1572: 
        !          1573: @example
        !          1574: I_plus:        /* + ( n1 n2 -- n ) */  /* label, stack effect */
        !          1575: /*  */                          /* documentation */
        !          1576: {
        !          1577: DEF_CA                          /* definition of variable ca (indirect threading) */
        !          1578: Cell n1;                        /* definitions of variables */
        !          1579: Cell n2;
        !          1580: Cell n;
        !          1581: n1 = (Cell) sp[1];              /* input */
        !          1582: n2 = (Cell) TOS;
        !          1583: sp += 1;                        /* stack adjustment */
        !          1584: NAME("+")                       /* debugging output (with -DDEBUG) */
        !          1585: {
        !          1586: n = n1+n2;                      /* C code taken from the source */
        !          1587: }
        !          1588: NEXT_P1;                        /* NEXT part 1 */
        !          1589: TOS = (Cell)n;                  /* output */
        !          1590: NEXT_P2;                        /* NEXT part 2 */
        !          1591: }
        !          1592: @end example
        !          1593: 
        !          1594: This looks long and inefficient, but the GNU C compiler optimizes quite
        !          1595: well and produces optimal code for @code{+} on, e.g., the R3000 and the
        !          1596: HP RISC machines: Defining the @code{n}s does not produce any code, and
        !          1597: using them as intermediate storage also adds no cost.
        !          1598: 
        !          1599: There are also other optimizations, that are not illustrated by this
        !          1600: example: Assignments between simple variables are usually for free (copy
        !          1601: propagation). If one of the stack items is not used by the primitive
        !          1602: (e.g.  in @code{drop}), the compiler eliminates the load from the stack
        !          1603: (dead code elimination). On the other hand, there are some things that
        !          1604: the compiler does not do, therefore they are performed by
        !          1605: @file{prims2x.fs}: The compiler does not optimize code away that stores
        !          1606: a stack item to the place where it just came from (e.g., @code{over}).
        !          1607: 
        !          1608: While programming a primitive is usually easy, there are a few cases
        !          1609: where the programmer has to take the actions of the generator into
        !          1610: account, most notably @code{?dup}, but also words that do not (always)
        !          1611: fall through to NEXT.
        !          1612: 
        !          1613: @subsection TOS Optimization
        !          1614: 
        !          1615: An important optimization for stack machine emulators, e.g., Forth
        !          1616: engines, is keeping  one or more of the top stack items in
        !          1617: registers.  If a word has the stack effect {@var{in1}...@var{inx} @code{--}
        !          1618: @var{out1}...@var{outy}}, keeping the top @var{n} items in registers
        !          1619: @itemize
        !          1620: @item
        !          1621: is better than keeping @var{n-1} items, if @var{x>=n} and @var{y>=n},
        !          1622: due to fewer loads from and stores to the stack.
        !          1623: @item is slower than keeping @var{n-1} items, if @var{x<>y} and @var{x<n} and
        !          1624: @var{y<n}, due to additional moves between registers.
        !          1625: @end itemize
        !          1626: 
        !          1627: In particular, keeping one item in a register is never a disadvantage,
        !          1628: if there are enough registers. Keeping two items in registers is a
        !          1629: disadvantage for frequent words like @code{?branch}, constants,
        !          1630: variables, literals and @code{i}. Therefore our generator only produces
        !          1631: code that keeps zero or one items in registers. The generated C code
        !          1632: covers both cases; the selection between these alternatives is made at
        !          1633: C-compile time using the switch @code{-DUSE_TOS}. @code{TOS} in the C
        !          1634: code for @code{+} is just a simple variable name in the one-item case,
        !          1635: otherwise it is a macro that expands into @code{sp[0]}. Note that the
        !          1636: GNU C compiler tries to keep simple variables like @code{TOS} in
        !          1637: registers, and it usually succeeds, if there are enough registers.
        !          1638: 
        !          1639: The primitive generator performs the TOS optimization for the
        !          1640: floating-point stack, too (@code{-DUSE_FTOS}). For floating-point
        !          1641: operations the benefit of this optimization is even larger:
        !          1642: floating-point operations take quite long on most processors, but can be
        !          1643: performed in parallel with other operations as long as their results are
        !          1644: not used. If the FP-TOS is kept in a register, this works. If
        !          1645: it is kept on the stack, i.e., in memory, the store into memory has to
        !          1646: wait for the result of the floating-point operation, lengthening the
        !          1647: execution time of the primitive considerably.
        !          1648: 
        !          1649: The TOS optimization makes the automatic generation of primitives a
        !          1650: bit more complicated. Just replacing all occurrences of @code{sp[0]} by
        !          1651: @code{TOS} is not sufficient. There are some special cases to
        !          1652: consider:
        !          1653: @itemize
        !          1654: @item In the case of @code{dup ( w -- w w )} the generator must not
        !          1655: eliminate the store to the original location of the item on the stack,
        !          1656: if the TOS optimization is turned on.
        !          1657: @item Primitives with stack effects of the form {@code{--}
        !          1658: @var{out1}...@var{outy}} must store the TOS to the stack at the start.
        !          1659: Likewise, primitives with the stack effect {@var{in1}...@var{inx} @code{--}}
        !          1660: must load the TOS from the stack at the end. But for the null stack
        !          1661: effect @code{--} no stores or loads should be generated.
        !          1662: @end itemize
        !          1663: 
        !          1664: @subsection Produced code
        !          1665: 
        !          1666: To see what assembly code is produced for the primitives on your machine
        !          1667: with your compiler and your flag settings, type @code{make engine.s} and
        !          1668: look at the resulting file @file{engine.c}.
        !          1669: 
        !          1670: @section System Architecture
        !          1671: 
        !          1672: Our Forth system consists not only of primitives, but also of
        !          1673: definitions written in Forth. Since the Forth compiler itself belongs
        !          1674: to those definitions, it is not possible to start the system with the
        !          1675: primitives and the Forth source alone. Therefore we provide the Forth
        !          1676: code as an image file in nearly executable form. At the start of the
        !          1677: system a C routine loads the image file into memory, sets up the
        !          1678: memory (stacks etc.) according to information in the image file, and
        !          1679: starts executing Forth code.
        !          1680: 
        !          1681: The image file format is a compromise between the goals of making it
        !          1682: easy to generate image files and making them portable. The easiest way
        !          1683: to generate an image file is to just generate a memory dump. However,
        !          1684: this kind of image file cannot be used on a different machine, or on
        !          1685: the next version of the engine on the same machine, it even might not
        !          1686: work with the same engine compiled by a different version of the C
        !          1687: compiler. We would like to have as few versions of the image file as
        !          1688: possible, because we do not want to distribute many versions of the
        !          1689: same image file, and to make it easy for the users to use their image
        !          1690: files on many machines. We currently need to create a different image
        !          1691: file for machines with different cell sizes and different byte order
        !          1692: (little- or big-endian)@footnote{We consider adding information to the
        !          1693: image file that enables the loader to change the byte order.}.
        !          1694: 
        !          1695: Forth code that is going to end up in a portable image file has to
        !          1696: comply to some restrictions: addresses have to be stored in memory
        !          1697: with special words (@code{A!}, @code{A,}, etc.) in order to make the
        !          1698: code relocatable. Cells, floats, etc., have to be stored at the
        !          1699: natural alignment boundaries@footnote{E.g., store floats (8 bytes) at
        !          1700: an address dividable by~8. This happens automatically in our system
        !          1701: when you use the ANSI alignment words.}, in order to avoid alignment
        !          1702: faults on machines with stricter alignment. The image file is produced
        !          1703: by a metacompiler (@file{cross.fs}).
        !          1704: 
        !          1705: So, unlike the image file of Mitch Bradleys @code{cforth}, our image
        !          1706: file is not directly executable, but has to undergo some manipulations
        !          1707: during loading. Address relocation is performed at image load-time, not
        !          1708: at run-time. The loader also has to replace tokens standing for
        !          1709: primitive calls with the appropriate code-field addresses (or code
        !          1710: addresses in the case of direct threading).
1.1       anton    1711: 
                   1712: @contents
                   1713: @bye
                   1714: 

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