Annotation of gforth/prims2x.fs, revision 1.145

1.16      anton       1: \ converts primitives to, e.g., C code 
                      2: 
1.134     anton       3: \ Copyright (C) 1995,1996,1997,1998,2000,2003 Free Software Foundation, Inc.
1.16      anton       4: 
                      5: \ This file is part of Gforth.
                      6: 
                      7: \ Gforth is free software; you can redistribute it and/or
                      8: \ modify it under the terms of the GNU General Public License
                      9: \ as published by the Free Software Foundation; either version 2
                     10: \ of the License, or (at your option) any later version.
                     11: 
                     12: \ This program is distributed in the hope that it will be useful,
                     13: \ but WITHOUT ANY WARRANTY; without even the implied warranty of
                     14: \ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
                     15: \ GNU General Public License for more details.
                     16: 
                     17: \ You should have received a copy of the GNU General Public License
                     18: \ along with this program; if not, write to the Free Software
1.48      anton      19: \ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111, USA.
1.16      anton      20: 
                     21: 
1.71      anton      22: \ This is not very nice (hard limits, no checking, assumes 1 chars = 1).
                     23: \ And it grew even worse when it aged.
1.1       anton      24: 
                     25: \ Optimizations:
                     26: \ superfluous stores are removed. GCC removes the superfluous loads by itself
                     27: \ TOS and FTOS can be kept in register( variable)s.
                     28: \ 
                     29: \ Problems:
                     30: \ The TOS optimization is somewhat hairy. The problems by example:
                     31: \ 1) dup ( w -- w w ): w=TOS; sp-=1; sp[1]=w; TOS=w;
                     32: \    The store is not superfluous although the earlier opt. would think so
                     33: \    Alternatively:    sp[0]=TOS; w=TOS; sp-=1; TOS=w;
                     34: \ 2) ( -- .. ): sp[0] = TOS; ... /* This additional store is necessary */
                     35: \ 3) ( .. -- ): ... TOS = sp[0]; /* as well as this load */
                     36: \ 4) ( -- ): /* but here they are unnecessary */
                     37: \ 5) Words that call NEXT themselves have to be done very carefully.
                     38: \
                     39: \ To do:
1.8       pazsan     40: \ add the store optimization for doubles
1.1       anton      41: \ regarding problem 1 above: It would be better (for over) to implement
                     42: \      the alternative
1.80      anton      43: \ store optimization for combined instructions.
                     44: 
                     45: \ Design Uglyness:
                     46: 
                     47: \ - global state (values, variables) in connection with combined instructions.
                     48: 
                     49: \ - index computation is different for instruction-stream and the
                     50: \ stacks; there are two mechanisms for dealing with that
                     51: \ (stack-in-index-xt and a test for stack==instruction-stream); there
                     52: \ should be only one.
1.1       anton      53: 
1.137     jwilke     54: \ for backwards compatibility, jaw
                     55: require compat/strcomp.fs
                     56: 
1.3       pazsan     57: warnings off
                     58: 
1.136     anton      59: \ redefinitions of kernel words not present in gforth-0.6.1
                     60: : latestxt lastcfa @ ;
                     61: : latest last @ ;
                     62: 
1.97      jwilke     63: [IFUNDEF] try
                     64: include startup.fs
                     65: [THEN]
                     66: 
1.49      anton      67: : struct% struct ; \ struct is redefined in gray
                     68: 
1.98      pazsan     69: warnings off
1.110     anton      70: \ warnings on
1.98      pazsan     71: 
1.39      jwilke     72: include ./gray.fs
1.133     anton      73: 128 constant max-effect \ number of things on one side of a stack effect
1.71      anton      74: 4 constant max-stacks  \ the max. number of stacks (including inst-stream).
1.1       anton      75: 255 constant maxchar
                     76: maxchar 1+ constant eof-char
1.17      anton      77: #tab constant tab-char
                     78: #lf constant nl-char
1.1       anton      79: 
1.18      anton      80: variable rawinput \ pointer to next character to be scanned
                     81: variable endrawinput \ pointer to the end of the input (the char after the last)
                     82: variable cookedinput \ pointer to the next char to be parsed
1.17      anton      83: variable line \ line number of char pointed to by input
1.65      anton      84: variable line-start \ pointer to start of current line (for error messages)
                     85: 0 line !
1.17      anton      86: 2variable filename \ filename of original input file
                     87: 0 0 filename 2!
1.111     anton      88: 2variable out-filename \ filename of the output file (for sync lines)
                     89: 0 0 out-filename 2!
1.25      pazsan     90: 2variable f-comment
                     91: 0 0 f-comment 2!
1.17      anton      92: variable skipsynclines \ are sync lines ("#line ...") invisible to the parser?
1.111     anton      93: skipsynclines on
                     94: variable out-nls \ newlines in output (for output sync lines)
                     95: 0 out-nls !
1.112     anton      96: variable store-optimization \ use store optimization?
                     97: store-optimization off
                     98: 
1.116     anton      99: variable include-skipped-insts
                    100: \ does the threaded code for a combined instruction include the cells
                    101: \ for the component instructions (true) or only the cells for the
                    102: \ inline arguments (false)
                    103: include-skipped-insts off
1.1       anton     104: 
1.121     anton     105: variable immarg \ values for immediate arguments (to be used in IMM_ARG macros)
                    106: $12340000 immarg !
                    107: 
1.72      anton     108: : th ( addr1 n -- addr2 )
                    109:     cells + ;
                    110: 
                    111: : holds ( addr u -- )
                    112:     \ like HOLD, but for a string
                    113:     tuck + swap 0 +do
                    114:        1- dup c@ hold
                    115:     loop
                    116:     drop ;
1.71      anton     117: 
1.82      anton     118: : insert-wordlist { c-addr u wordlist xt -- }
1.81      anton     119:     \ adds name "addr u" to wordlist using defining word xt
                    120:     \ xt may cause additional stack effects
                    121:     get-current >r wordlist set-current
                    122:     c-addr u nextname xt execute
                    123:     r> set-current ;
                    124: 
1.1       anton     125: : start ( -- addr )
1.18      anton     126:  cookedinput @ ;
1.1       anton     127: 
                    128: : end ( addr -- addr u )
1.18      anton     129:  cookedinput @ over - ;
1.1       anton     130: 
1.71      anton     131: : print-error-line ( -- )
                    132:     \ print the current line and position
                    133:     line-start @ endrawinput @ over - 2dup nl-char scan drop nip ( start end )
                    134:     over - type cr
                    135:     line-start @ rawinput @ over - typewhite ." ^" cr ;
                    136: 
                    137: : ?print-error { f addr u -- }
                    138:     f ?not? if
                    139:        outfile-id >r try
                    140:            stderr to outfile-id
                    141:            filename 2@ type ." :" line @ 0 .r ." : " addr u type cr
                    142:            print-error-line
                    143:            0
                    144:        recover endtry
                    145:        r> to outfile-id throw
1.111     anton     146:        1 (bye) \ abort
1.71      anton     147:     endif ;
                    148: 
1.63      anton     149: : quote ( -- )
                    150:     [char] " emit ;
                    151: 
1.111     anton     152: \ count output lines to generate sync lines for output
                    153: 
                    154: : count-nls ( addr u -- )
                    155:     bounds u+do
                    156:        i c@ nl-char = negate out-nls +!
                    157:     loop ;
                    158: 
                    159: :noname ( addr u -- )
                    160:     2dup count-nls
                    161:     defers type ;
                    162: is type
                    163: 
1.72      anton     164: variable output          \ xt ( -- ) of output word for simple primitives
                    165: variable output-combined \ xt ( -- ) of output word for combined primitives
1.1       anton     166: 
1.49      anton     167: struct%
1.71      anton     168:     cell%    field stack-number \ the number of this stack
1.49      anton     169:     cell% 2* field stack-pointer \ stackpointer name
1.74      anton     170:     cell%    field stack-type \ name for default type of stack items
1.53      anton     171:     cell%    field stack-in-index-xt \ ( in-size item -- in-index )
1.126     anton     172:     cell%    field stack-access-transform \ ( nitem -- index )
1.49      anton     173: end-struct stack%
                    174: 
1.53      anton     175: struct%
                    176:  cell% 2* field item-name   \ name, excluding stack prefixes
                    177:  cell%    field item-stack  \ descriptor for the stack used, 0 is default
                    178:  cell%    field item-type   \ descriptor for the item type
                    179:  cell%    field item-offset \ offset in stack items, 0 for the deepest element
1.66      anton     180:  cell%   field item-first  \ true if this is the first occurence of the item
1.53      anton     181: end-struct item%
                    182: 
                    183: struct%
                    184:     cell% 2* field type-c-name
                    185:     cell%    field type-stack \ default stack
                    186:     cell%    field type-size  \ size of type in stack items
                    187:     cell%    field type-fetch \ xt of fetch code generator ( item -- )
                    188:     cell%    field type-store \ xt of store code generator ( item -- )
                    189: end-struct type%
                    190: 
1.144     anton     191: struct%
                    192:     cell%    field register-number
                    193:     cell%    field register-type \ pointer to type
                    194:     cell% 2* field register-name \ c name
                    195: end-struct register%
                    196: 
                    197: struct%
                    198:     cell% 2* field ss-registers  \ addr u; ss-registers[0] is TOS
                    199:                                  \ 0 means: use memory
                    200:     cell%    field ss-offset     \ stack pointer offset: sp[-offset] is TOS
                    201: end-struct ss% \ stack-state
                    202: 
                    203: struct%
                    204:     cell%              field state-number
                    205:     cell% max-stacks * field state-sss
                    206: end-struct state%
                    207: 
1.72      anton     208: variable next-stack-number 0 next-stack-number !
                    209: create stacks max-stacks cells allot \ array of stacks
1.144     anton     210: 256 constant max-registers
                    211: create registers max-registers cells allot \ array of registers
                    212: variable nregisters 0 nregisters ! \ number of registers
                    213: variable next-state-number 0 next-state-number ! \ next state number
1.72      anton     214: 
1.53      anton     215: : stack-in-index ( in-size item -- in-index )
                    216:     item-offset @ - 1- ;
                    217: 
                    218: : inst-in-index ( in-size item -- in-index )
                    219:     nip dup item-offset @ swap item-type @ type-size @ + 1- ;
                    220: 
1.92      anton     221: : make-stack ( addr-ptr u1 type "stack-name" -- )
                    222:     next-stack-number @ max-stacks < s" too many stacks" ?print-error
1.49      anton     223:     create stack% %allot >r
1.72      anton     224:     r@ stacks next-stack-number @ th !
1.92      anton     225:     next-stack-number @ r@ stack-number !
                    226:     1 next-stack-number +!
1.74      anton     227:     r@ stack-type !
1.53      anton     228:     save-mem r@ stack-pointer 2! 
1.126     anton     229:     ['] stack-in-index r@ stack-in-index-xt !
                    230:     ['] noop r@ stack-access-transform !
                    231:     rdrop ;
1.49      anton     232: 
1.92      anton     233: : map-stacks { xt -- }
1.118     anton     234:     \ perform xt for all stacks
                    235:     next-stack-number @ 0 +do
                    236:        stacks i th @ xt execute
                    237:     loop ;
                    238: 
                    239: : map-stacks1 { xt -- }
1.92      anton     240:     \ perform xt for all stacks except inst-stream
                    241:     next-stack-number @ 1 +do
                    242:        stacks i th @ xt execute
                    243:     loop ;
                    244: 
1.49      anton     245: \ stack items
                    246: 
                    247: : init-item ( addr u addr1 -- )
                    248:     \ initialize item at addr1 with name addr u
                    249:     \ !! remove stack prefix
                    250:     dup item% %size erase
                    251:     item-name 2! ;
                    252: 
1.64      anton     253: : map-items { addr end xt -- }
                    254:     \ perform xt for all items in array addr...end
                    255:     end addr ?do
                    256:        i xt execute
                    257:     item% %size +loop ;
                    258: 
1.77      anton     259: \ types
                    260: 
                    261: : print-type-prefix ( type -- )
                    262:     body> >head name>string type ;
                    263: 
1.49      anton     264: \ various variables for storing stuff of one primitive
1.1       anton     265: 
1.69      anton     266: struct%
                    267:     cell% 2* field prim-name
                    268:     cell% 2* field prim-wordset
                    269:     cell% 2* field prim-c-name
                    270:     cell% 2* field prim-doc
                    271:     cell% 2* field prim-c-code
                    272:     cell% 2* field prim-forth-code
                    273:     cell% 2* field prim-stack-string
1.82      anton     274:     cell%    field prim-num            \ ordinal number
1.75      anton     275:     cell%    field prim-items-wordlist \ unique items
1.69      anton     276:     item% max-effect * field prim-effect-in
                    277:     item% max-effect * field prim-effect-out
                    278:     cell%    field prim-effect-in-end
                    279:     cell%    field prim-effect-out-end
1.71      anton     280:     cell% max-stacks * field prim-stacks-in  \ number of in items per stack
                    281:     cell% max-stacks * field prim-stacks-out \ number of out items per stack
1.69      anton     282: end-struct prim%
                    283: 
1.70      anton     284: : make-prim ( -- prim )
                    285:     prim% %alloc { p }
                    286:     s" " p prim-doc 2! s" " p prim-forth-code 2! s" " p prim-wordset 2!
                    287:     p ;
                    288: 
1.79      anton     289: 0 value prim     \ in combined prims either combined or a part
                    290: 0 value combined \ in combined prims the combined prim
                    291: variable in-part \ true if processing a part
                    292:  in-part off
1.144     anton     293: 0 value state-in  \ state on entering prim
                    294: 0 value state-out \ state on exiting prim
                    295: 0 value state-default  \ canonical state at bb boundaries
1.79      anton     296: 
1.118     anton     297: : prim-context ( ... p xt -- ... )
                    298:     \ execute xt with prim set to p
                    299:     prim >r
                    300:     swap to prim
                    301:     catch
                    302:     r> to prim
                    303:     throw ;
                    304: 
1.79      anton     305: 1000 constant max-combined
                    306: create combined-prims max-combined cells allot
                    307: variable num-combined
1.118     anton     308: variable part-num \ current part number during process-combined
1.79      anton     309: 
1.114     anton     310: : map-combined { xt -- }
                    311:     \ perform xt for all components of the current combined instruction
                    312:     num-combined @ 0 +do
                    313:        combined-prims i th @ xt execute
                    314:     loop ;
                    315: 
1.81      anton     316: table constant combinations
                    317:   \ the keys are the sequences of pointers to primitives
                    318: 
1.79      anton     319: create current-depth max-stacks cells allot
                    320: create max-depth     max-stacks cells allot
                    321: create min-depth     max-stacks cells allot
1.69      anton     322: 
1.118     anton     323: create sp-update-in max-stacks cells allot
                    324: \ where max-depth occured the first time
                    325: create max-depths max-stacks max-combined 1+ * cells allot
1.119     anton     326: \ maximum depth at start of each part: array[parts] of array[stack]
                    327: create max-back-depths max-stacks max-combined 1+ * cells allot
                    328: \ maximun depth from end of the combination to the start of the each part
1.118     anton     329: 
                    330: : s-c-max-depth ( nstack ncomponent -- addr )
                    331:     max-stacks * + cells max-depths + ;
                    332: 
1.119     anton     333: : s-c-max-back-depth ( nstack ncomponent -- addr )
                    334:     max-stacks * + cells max-back-depths + ;
                    335: 
1.71      anton     336: wordlist constant primitives
                    337: 
                    338: : create-prim ( prim -- )
1.82      anton     339:     dup prim-name 2@ primitives ['] constant insert-wordlist ;
1.71      anton     340: 
                    341: : stack-in ( stack -- addr )
                    342:     \ address of number of stack items in effect in
                    343:     stack-number @ cells prim prim-stacks-in + ;
                    344: 
                    345: : stack-out ( stack -- addr )
                    346:     \ address of number of stack items in effect out
                    347:     stack-number @ cells prim prim-stacks-out + ;
                    348: 
1.69      anton     349: \ global vars
1.17      anton     350: variable c-line
                    351: 2variable c-filename
                    352: variable name-line
                    353: 2variable name-filename
                    354: 2variable last-name-filename
1.30      pazsan    355: Variable function-number 0 function-number !
1.140     pazsan    356: Variable function-old 0 function-old !
                    357: : function-diff ( n -- )
                    358:     ." GROUPADD(" function-number @ function-old @ - 0 .r ." )" cr
                    359:     function-number @ function-old ! ;
                    360: : forth-fdiff ( -- )
                    361:     function-number @ function-old @ - 0 .r ."  groupadd" cr
                    362:     function-number @ function-old ! ;
1.1       anton     363: 
                    364: \ a few more set ops
                    365: 
                    366: : bit-equivalent ( w1 w2 -- w3 )
                    367:  xor invert ;
                    368: 
                    369: : complement ( set1 -- set2 )
                    370:  empty ['] bit-equivalent binary-set-operation ;
                    371: 
1.121     anton     372: \ forward declaration for inst-stream (breaks cycle in definitions)
                    373: defer inst-stream-f ( -- stack )
                    374: 
1.80      anton     375: \ stack access stuff
1.79      anton     376: 
1.126     anton     377: : normal-stack-access0 { n stack -- }
1.144     anton     378:     \ n has the ss-offset already applied (see ...-access1)
1.126     anton     379:     n stack stack-access-transform @ execute ." [" 0 .r ." ]" ;
1.144     anton     380: 
                    381: : state-ss { stack state -- ss }
                    382:     state state-sss stack stack-number @ th @ ;
                    383: 
                    384: : stack-reg { n stack state -- reg }
                    385:     \ n is the index (TOS=0); reg is 0 if the access is to memory
                    386:     stack state state-ss ss-registers 2@ n u> if ( addr ) \ in ss-registers?
                    387:        n th @
1.49      anton     388:     else
1.144     anton     389:        drop 0
1.49      anton     390:     endif ;
1.1       anton     391: 
1.144     anton     392: : .reg ( reg -- )
                    393:     register-name 2@ type ;
                    394: 
                    395: : stack-offset ( stack state -- n )
                    396:     \ offset for stack in state
                    397:     state-ss ss-offset @ ;
                    398: 
                    399: : normal-stack-access1 { n stack state -- }
                    400:     n stack state stack-reg ?dup-if
                    401:        .reg exit
                    402:     endif
                    403:     stack stack-pointer 2@ type
                    404:     n stack state stack-offset - stack normal-stack-access0 ;
                    405: 
                    406: : normal-stack-access ( n stack state -- )
                    407:     over inst-stream-f = if
1.121     anton     408:        ." IMM_ARG(" normal-stack-access1 ." ," immarg ? ." )"
                    409:        1 immarg +!
                    410:     else
                    411:        normal-stack-access1
                    412:     endif ;
1.80      anton     413: 
1.118     anton     414: : stack-depth { stack -- n }
                    415:     current-depth stack stack-number @ th @ ;
                    416: 
1.79      anton     417: : part-stack-access { n stack -- }
1.80      anton     418:     \ print _<stack><x>, x=inst-stream? n : maxdepth-currentdepth-n-1
1.79      anton     419:     ." _" stack stack-pointer 2@ type
                    420:     stack stack-number @ { stack# }
1.118     anton     421:     stack stack-depth n + { access-depth }
1.80      anton     422:     stack inst-stream-f = if
                    423:        access-depth
                    424:     else
                    425:        combined prim-stacks-in stack# th @
                    426:        assert( dup max-depth stack# th @ = )
                    427:        access-depth - 1-
                    428:     endif
1.79      anton     429:     0 .r ;
                    430: 
1.118     anton     431: : part-stack-read { n stack -- }
                    432:     stack stack-depth n + ( ndepth )
                    433:     stack stack-number @ part-num @ s-c-max-depth @
                    434: \    max-depth stack stack-number @ th @ ( ndepth nmaxdepth )
                    435:     over <= if ( ndepth ) \ load from memory
1.144     anton     436:        stack state-in normal-stack-access
1.118     anton     437:     else
                    438:        drop n stack part-stack-access
                    439:     endif ;
                    440: 
1.119     anton     441: : stack-diff ( stack -- n )
                    442:     \ in-out
                    443:     dup stack-in @ swap stack-out @ - ;
                    444: 
                    445: : part-stack-write { n stack -- }
                    446:     stack stack-depth n +
                    447:     stack stack-number @ part-num @ s-c-max-back-depth @
                    448:     over <= if ( ndepth )
                    449:        stack combined ['] stack-diff prim-context -
1.144     anton     450:        stack state-out normal-stack-access
1.119     anton     451:     else
                    452:        drop n stack part-stack-access
                    453:     endif ;
1.118     anton     454: 
                    455: : stack-read ( n stack -- )
                    456:     \ print a stack access at index n of stack
                    457:     in-part @ if
                    458:        part-stack-read
                    459:     else
1.144     anton     460:        state-in normal-stack-access
1.118     anton     461:     endif ;
                    462: 
                    463: : stack-write ( n stack -- )
1.79      anton     464:     \ print a stack access at index n of stack
                    465:     in-part @ if
1.118     anton     466:        part-stack-write
1.79      anton     467:     else
1.144     anton     468:        state-out normal-stack-access
1.79      anton     469:     endif ;
                    470: 
1.53      anton     471: : item-in-index { item -- n }
1.49      anton     472:     \ n is the index of item (in the in-effect)
1.53      anton     473:     item item-stack @ dup >r stack-in @ ( in-size r:stack )
                    474:     item r> stack-in-index-xt @ execute ;
1.1       anton     475: 
1.78      anton     476: : item-stack-type-name ( item -- addr u )
                    477:     item-stack @ stack-type @ type-c-name 2@ ;
                    478: 
1.1       anton     479: : fetch-single ( item -- )
1.106     anton     480:     \ fetch a single stack item from its stack
                    481:     >r
                    482:     ." vm_" r@ item-stack-type-name type
                    483:     ." 2" r@ item-type @ print-type-prefix ." ("
1.118     anton     484:     r@ item-in-index r@ item-stack @ stack-read ." ,"
1.106     anton     485:     r@ item-name 2@ type
                    486:     ." );" cr
                    487:     rdrop ; 
1.1       anton     488: 
                    489: : fetch-double ( item -- )
1.106     anton     490:     \ fetch a double stack item from its stack
                    491:     >r
                    492:     ." vm_two"
                    493:     r@ item-stack-type-name type ." 2"
                    494:     r@ item-type @ print-type-prefix ." ("
1.118     anton     495:     r@ item-in-index r@ item-stack @ 2dup ." (Cell)" stack-read
                    496:     ." , "                      -1 under+ ." (Cell)" stack-read
1.106     anton     497:     ." , " r@ item-name 2@ type
                    498:     ." )" cr
                    499:     rdrop ;
1.1       anton     500: 
1.49      anton     501: : same-as-in? ( item -- f )
                    502:  \ f is true iff the offset and stack of item is the same as on input
1.1       anton     503:  >r
1.74      anton     504:  r@ item-first @ if
                    505:      rdrop false exit
                    506:  endif
1.75      anton     507:  r@ item-name 2@ prim prim-items-wordlist @ search-wordlist 0= abort" bug"
1.1       anton     508:  execute @
                    509:  dup r@ =
                    510:  if \ item first appeared in output
                    511:    drop false
                    512:  else
1.49      anton     513:    dup  item-stack  @ r@ item-stack  @ = 
                    514:    swap item-offset @ r@ item-offset @ = and
1.1       anton     515:  endif
                    516:  rdrop ;
                    517: 
1.49      anton     518: : item-out-index ( item -- n )
1.144     anton     519:     \ n is the index of item (in the out-effect)
1.49      anton     520:     >r r@ item-stack @ stack-out @ r> item-offset @ - 1- ;
1.31      pazsan    521: 
1.1       anton     522: : really-store-single ( item -- )
1.106     anton     523:     >r
                    524:     ." vm_"
                    525:     r@ item-type @ print-type-prefix ." 2"
                    526:     r@ item-stack-type-name type ." ("
                    527:     r@ item-name 2@ type ." ,"
1.118     anton     528:     r@ item-out-index r@ item-stack @ stack-write ." );"
1.106     anton     529:     rdrop ;
1.1       anton     530: 
1.144     anton     531: : store-single { item -- }
                    532:     item item-stack @ { stack }
                    533:     store-optimization @ in-part @ 0= and item same-as-in? and
                    534:     item item-in-index  stack state-in  stack-reg 0= and \  in in memory?
                    535:     item item-out-index stack state-out stack-reg 0= and \ out in memory?
                    536:     0= if
                    537:        item really-store-single cr
                    538:     endif ;
1.1       anton     539: 
                    540: : store-double ( item -- )
                    541: \ !! store optimization is not performed, because it is not yet needed
                    542:  >r
1.78      anton     543:  ." vm_"
                    544:  r@ item-type @ print-type-prefix ." 2two"
                    545:  r@ item-stack-type-name type ." ("
                    546:  r@ item-name 2@ type ." , "
1.118     anton     547:  r@ item-out-index r@ item-stack @ 2dup stack-write
                    548:  ." , "                       -1 under+ stack-write
1.106     anton     549:  ." )" cr
1.1       anton     550:  rdrop ;
                    551: 
1.54      anton     552: : single ( -- xt1 xt2 n )
                    553:     ['] fetch-single ['] store-single 1 ;
1.1       anton     554: 
1.54      anton     555: : double ( -- xt1 xt2 n )
                    556:     ['] fetch-double ['] store-double 2 ;
1.1       anton     557: 
                    558: : s, ( addr u -- )
                    559: \ allocate a string
                    560:  here swap dup allot move ;
                    561: 
1.50      anton     562: wordlist constant prefixes
                    563: 
                    564: : declare ( addr "name" -- )
                    565: \ remember that there is a stack item at addr called name
                    566:  create , ;
                    567: 
                    568: : !default ( w addr -- )
                    569:     dup @ if
                    570:        2drop \ leave nonzero alone
                    571:     else
                    572:        !
                    573:     endif ;
                    574: 
                    575: : create-type { addr u xt1 xt2 n stack -- } ( "prefix" -- )
1.49      anton     576:     \ describes a type
                    577:     \ addr u specifies the C type name
                    578:     \ stack effect entries of the type start with prefix
                    579:     create type% %allot >r
                    580:     addr u save-mem r@ type-c-name 2!
                    581:     xt1   r@ type-fetch !
                    582:     xt2   r@ type-store !
                    583:     n     r@ type-size !
                    584:     stack r@ type-stack !
                    585:     rdrop ;
1.1       anton     586: 
1.105     anton     587: : type-prefix ( addr u xt1 xt2 n stack "prefix" -- )
1.94      anton     588:     get-current >r prefixes set-current
                    589:     create-type r> set-current
1.50      anton     590: does> ( item -- )
                    591:     \ initialize item
                    592:     { item typ }
                    593:     typ item item-type !
                    594:     typ type-stack @ item item-stack !default
1.75      anton     595:     item item-name 2@ prim prim-items-wordlist @ search-wordlist 0= if
1.66      anton     596:        item item-name 2@ nextname item declare
                    597:        item item-first on
                    598:        \ typ type-c-name 2@ type space type  ." ;" cr
1.50      anton     599:     else
                    600:        drop
1.66      anton     601:        item item-first off
1.50      anton     602:     endif ;
                    603: 
                    604: : execute-prefix ( item addr1 u1 -- )
                    605:     \ execute the word ( item -- ) associated with the longest prefix
                    606:     \ of addr1 u1
                    607:     0 swap ?do
                    608:        dup i prefixes search-wordlist
                    609:        if \ ok, we have the type ( item addr1 xt )
                    610:            nip execute
                    611:            UNLOOP EXIT
                    612:        endif
                    613:        -1 s+loop
                    614:     \ we did not find a type, abort
1.81      anton     615:     false s" unknown prefix" ?print-error ;
1.1       anton     616: 
                    617: : declaration ( item -- )
1.50      anton     618:     dup item-name 2@ execute-prefix ;
1.1       anton     619: 
1.64      anton     620: : declaration-list ( addr1 addr2 -- )
                    621:     ['] declaration map-items ;
                    622: 
                    623: : declarations ( -- )
1.75      anton     624:  wordlist dup prim prim-items-wordlist ! set-current
1.69      anton     625:  prim prim-effect-in prim prim-effect-in-end @ declaration-list
                    626:  prim prim-effect-out prim prim-effect-out-end @ declaration-list ;
1.64      anton     627: 
1.66      anton     628: : print-declaration { item -- }
                    629:     item item-first @ if
                    630:        item item-type @ type-c-name 2@ type space
                    631:        item item-name 2@ type ." ;" cr
                    632:     endif ;
                    633: 
                    634: : print-declarations ( -- )
1.69      anton     635:     prim prim-effect-in  prim prim-effect-in-end  @ ['] print-declaration map-items
                    636:     prim prim-effect-out prim prim-effect-out-end @ ['] print-declaration map-items ;
1.66      anton     637:     
1.51      anton     638: : stack-prefix ( stack "prefix" -- )
1.94      anton     639:     get-current >r prefixes set-current
1.51      anton     640:     name tuck nextname create ( stack length ) 2,
1.94      anton     641:     r> set-current
1.51      anton     642: does> ( item -- )
                    643:     2@ { item stack prefix-length }
                    644:     item item-name 2@ prefix-length /string item item-name 2!
                    645:     stack item item-stack !
                    646:     item declaration ;
1.73      anton     647: 
1.74      anton     648: \ types pointed to by stacks for use in combined prims
1.83      anton     649: \ !! output-c-combined shouldn't use these names!
1.92      anton     650: : stack-type-name ( addr u "name" -- )
                    651:     single 0 create-type ;
                    652: 
1.93      anton     653: wordlist constant type-names \ this is here just to meet the requirement
                    654:                     \ that a type be a word; it is never used for lookup
1.83      anton     655: 
1.144     anton     656: : define-type ( addr u -- xt )
                    657:     \ define single type with name addr u, without stack
                    658:     get-current type-names set-current >r
                    659:     2dup nextname stack-type-name
                    660:     r> set-current
                    661:     latestxt ;
                    662: 
1.93      anton     663: : stack ( "name" "stack-pointer" "type" -- )
                    664:     \ define stack
                    665:     name { d: stack-name }
                    666:     name { d: stack-pointer }
                    667:     name { d: stack-type }
1.144     anton     668:     stack-type define-type
                    669:     stack-pointer rot >body stack-name nextname make-stack ;
1.93      anton     670: 
                    671: stack inst-stream IP Cell
1.73      anton     672: ' inst-in-index inst-stream stack-in-index-xt !
1.80      anton     673: ' inst-stream <is> inst-stream-f
1.73      anton     674: \ !! initialize stack-in and stack-out
1.1       anton     675: 
1.144     anton     676: \ registers
                    677: 
                    678: : make-register ( type addr u -- )
                    679:     \ define register with type TYPE and name ADDR U.
                    680:     nregisters @ max-registers < s" too many registers" ?print-error
                    681:     2dup nextname create register% %allot >r
                    682:     r@ register-name 2!
                    683:     r@ register-type !
                    684:     nregisters @ r@ register-number !
                    685:     1 nregisters +!
                    686:     rdrop ;
                    687: 
                    688: : register ( "name" "type" -- )
                    689:     \ define register
                    690:     name { d: reg-name }
                    691:     name { d: reg-type }
                    692:     reg-type define-type >body
                    693:     reg-name make-register ;
                    694: 
                    695: \ stack-states
                    696: 
                    697: : stack-state ( a-addr u uoffset "name" -- )
                    698:     create ss% %allot >r
                    699:     r@ ss-offset !
                    700:     r@ ss-registers 2!
                    701:     rdrop ;
                    702: 
                    703: 0 0 0 stack-state default-ss
                    704: 
                    705: \ state
                    706: 
                    707: : state ( "name" -- )
                    708:     \ create a state initialized with default-sss
                    709:     create state% %allot { s }
                    710:     next-state-number @ s state-number ! 1 next-state-number +!
                    711:     max-stacks 0 ?do
                    712:        default-ss s state-sss i th !
                    713:     loop ;
                    714: 
                    715: : set-ss ( ss stack state -- )
                    716:     state-sss swap stack-number @ th ! ;
                    717: 
1.1       anton     718: \ offset computation
                    719: \ the leftmost (i.e. deepest) item has offset 0
                    720: \ the rightmost item has the highest offset
                    721: 
1.49      anton     722: : compute-offset { item xt -- }
                    723:     \ xt specifies in/out; update stack-in/out and set item-offset
                    724:     item item-type @ type-size @
                    725:     item item-stack @ xt execute dup @ >r +!
                    726:     r> item item-offset ! ;
                    727: 
1.64      anton     728: : compute-offset-in ( addr1 addr2 -- )
                    729:     ['] stack-in compute-offset ;
                    730: 
                    731: : compute-offset-out ( addr1 addr2 -- )
                    732:     ['] stack-out compute-offset ;
1.49      anton     733: 
1.1       anton     734: : compute-offsets ( -- )
1.132     anton     735:     prim prim-stacks-in  max-stacks cells erase
                    736:     prim prim-stacks-out max-stacks cells erase
1.69      anton     737:     prim prim-effect-in  prim prim-effect-in-end  @ ['] compute-offset-in  map-items
                    738:     prim prim-effect-out prim prim-effect-out-end @ ['] compute-offset-out map-items
1.81      anton     739:     inst-stream stack-out @ 0= s" # can only be on the input side" ?print-error ;
                    740: 
                    741: : process-simple ( -- )
                    742:     prim prim { W^ key } key cell
1.82      anton     743:     combinations ['] constant insert-wordlist
1.81      anton     744:     declarations compute-offsets
1.82      anton     745:     output @ execute ;
1.49      anton     746: 
1.144     anton     747: : stack-state-items ( stack state -- n )
                    748:     state-ss ss-registers 2@ nip ;
                    749: 
                    750: : unused-stack-items { stack -- n-in n-out }
                    751:     \ n-in  are the stack items in state-in  not used    by prim
                    752:     \ n-out are the stack items in state-out not written by prim
                    753:     stack state-in  stack-state-items stack stack-in  @ - 0 max
                    754:     stack state-out stack-state-items stack stack-out @ - 0 max ;
                    755: 
                    756: : spill-stack { stack -- }
                    757:     \ spill regs of state-in that are not used by prim and are not in state-out
                    758:     stack state-in stack-offset { offset }
                    759:     stack state-in stack-state-items ( items )
                    760:     dup stack unused-stack-items - - +do
                    761:        \ loop through the bottom items
                    762:        stack stack-pointer 2@ type
                    763:        i offset - stack normal-stack-access0 ."  = "
                    764:        i stack state-in normal-stack-access1 ." ;" cr
                    765:     loop ;
1.1       anton     766: 
1.144     anton     767: : spill-state ( -- )
                    768:     ['] spill-stack map-stacks1 ;
1.49      anton     769: 
1.144     anton     770: : fill-stack { stack -- }
                    771:     stack state-out stack-offset { offset }
                    772:     stack state-out stack-state-items ( items )
                    773:     dup stack unused-stack-items - + +do
                    774:        \ loop through the bottom items
                    775:        i stack state-out normal-stack-access1 ."  = "
                    776:        stack stack-pointer 2@ type
                    777:        i offset - stack normal-stack-access0 ." ;" cr
                    778:     loop ;
1.1       anton     779: 
1.144     anton     780: : fill-state ( -- )
1.53      anton     781:     \ !! inst-stream for prefetching?
1.144     anton     782:     ['] fill-stack map-stacks1 ;
1.49      anton     783: 
                    784: : fetch ( addr -- )
1.72      anton     785:     dup item-type @ type-fetch @ execute ;
1.1       anton     786: 
                    787: : fetches ( -- )
1.69      anton     788:     prim prim-effect-in prim prim-effect-in-end @ ['] fetch map-items ;
1.49      anton     789: 
1.144     anton     790: : reg-reg-move ( reg-from reg-to -- )
                    791:     2dup = if
                    792:        2drop
                    793:     else
                    794:        .reg ."  = " .reg ." ;" cr
                    795:     endif ;
                    796: 
                    797: : stack-bottom-reg { n stack state -- reg }
                    798:     stack state stack-state-items n - 1- stack state stack-reg ;
                    799: 
                    800: : stack-moves { stack -- }
                    801:     \ generate moves between registers in state-in/state-out that are
                    802:     \ not spilled or consumed/produced by prim.
                    803:     \ !! this works only for a simple stack cache, not e.g., for
                    804:     \ rotating stack caches, or registers shared between stacks (the
                    805:     \ latter would also require a change in interface)
                    806:     \ !! maybe place this after NEXT_P1?
                    807:     stack unused-stack-items 2dup < if ( n-in n-out )
                    808:        \ move registers from 0..n_in-1 to n_out-n_in..n_out-1
                    809:        over - { diff } ( n-in )
                    810:        -1 swap 1- -do
                    811:            i stack state-in stack-bottom-reg ( reg-from )
                    812:            i diff + stack state-out stack-bottom-reg reg-reg-move
                    813:        1 -loop
                    814:     else
                    815:        \ move registers from n_in-n_out..n_in-1 to 0..n_out-1
                    816:        swap over - { diff } ( n-out )
                    817:        0 +do
                    818:            i diff + stack state-in stack-bottom-reg ( reg-from )
                    819:            i stack state-out stack-bottom-reg reg-reg-move
                    820:        loop
                    821:     endif ;
                    822: 
1.126     anton     823: : stack-update-transform ( n1 stack -- n2 )
                    824:     \ n2 is the number by which the stack pointer should be
                    825:     \ incremented to pop n1 items
                    826:     stack-access-transform @ dup >r execute
                    827:     0 r> execute - ;
                    828: 
1.49      anton     829: : stack-pointer-update { stack -- }
1.144     anton     830:     \ and moves
1.123     anton     831:     \ stacks grow downwards
1.144     anton     832:     stack stack-diff ( in-out )
                    833:     stack state-in  stack-offset -
                    834:     stack state-out stack-offset + ( [in-in_offset]-[out-out_offset] )
1.49      anton     835:     ?dup-if \ this check is not necessary, gcc would do this for us
1.118     anton     836:        stack inst-stream = if
1.120     anton     837:            ." INC_IP(" 0 .r ." );" cr
1.118     anton     838:        else
1.126     anton     839:            stack stack-pointer 2@ type ."  += "
                    840:            stack stack-update-transform 0 .r ." ;" cr
1.118     anton     841:        endif
1.144     anton     842:     endif
                    843:     stack stack-moves ;
1.55      anton     844: 
1.1       anton     845: : stack-pointer-updates ( -- )
1.92      anton     846:     ['] stack-pointer-update map-stacks ;
1.1       anton     847: 
                    848: : store ( item -- )
                    849: \ f is true if the item should be stored
                    850: \ f is false if the store is probably not necessary
1.49      anton     851:  dup item-type @ type-store @ execute ;
1.1       anton     852: 
                    853: : stores ( -- )
1.69      anton     854:     prim prim-effect-out prim prim-effect-out-end @ ['] store map-items ;
1.8       pazsan    855: 
1.91      anton     856: : print-debug-arg { item -- }
                    857:     ." fputs(" quote space item item-name 2@ type ." =" quote ." , vm_out); "
                    858:     ." printarg_" item item-type @ print-type-prefix
                    859:     ." (" item item-name 2@ type ." );" cr ;
                    860:     
                    861: : print-debug-args ( -- )
                    862:     ." #ifdef VM_DEBUG" cr
                    863:     ." if (vm_debug) {" cr
                    864:     prim prim-effect-in prim prim-effect-in-end @ ['] print-debug-arg map-items
                    865: \    ." fputc('\n', vm_out);" cr
                    866:     ." }" cr
                    867:     ." #endif" cr ;
                    868: 
                    869: : print-debug-result { item -- }
                    870:     item item-first @ if
                    871:        item print-debug-arg
                    872:     endif ;
                    873: 
                    874: : print-debug-results ( -- )
                    875:     cr
                    876:     ." #ifdef VM_DEBUG" cr
                    877:     ." if (vm_debug) {" cr
                    878:     ." fputs(" quote ."  -- " quote ." , vm_out); "
                    879:     prim prim-effect-out prim prim-effect-out-end @ ['] print-debug-result map-items
                    880:     ." fputc('\n', vm_out);" cr
                    881:     ." }" cr
                    882:     ." #endif" cr ;
                    883: 
1.86      anton     884: : output-super-end ( -- )
                    885:     prim prim-c-code 2@ s" SET_IP" search if
                    886:        ." SUPER_END;" cr
                    887:     endif
                    888:     2drop ;
                    889: 
1.145   ! anton     890: 
        !           891: defer output-nextp0
        !           892: :noname ( -- )
        !           893:     ." NEXT_P0;" cr ;
        !           894: is output-nextp0
        !           895: 
        !           896: defer output-nextp1
        !           897: :noname ( -- )
        !           898:     ." NEXT_P1;" cr ;
        !           899: is output-nextp1
        !           900: 
1.124     anton     901: : output-nextp2 ( -- )
                    902:     ." NEXT_P2;" cr ;
                    903: 
                    904: variable tail-nextp2 \ xt to execute for printing NEXT_P2 in INST_TAIL
                    905: ' output-nextp2 tail-nextp2 !
                    906: 
1.120     anton     907: : output-label2 ( -- )
1.121     anton     908:     ." LABEL2(" prim prim-c-name 2@ type ." )" cr
                    909:     ." NEXT_P2;" cr ;
1.120     anton     910: 
                    911: : output-c-tail1 { xt -- }
                    912:     \ the final part of the generated C code, with xt printing LABEL2 or not.
1.86      anton     913:     output-super-end
1.91      anton     914:     print-debug-results
1.145   ! anton     915:     output-nextp1
1.52      anton     916:     stores
1.144     anton     917:     fill-state 
1.121     anton     918:     xt execute ;
1.108     anton     919: 
1.120     anton     920: : output-c-tail1-no-stores { xt -- }
                    921:     \ the final part of the generated C code for combinations
                    922:     output-super-end
1.145   ! anton     923:     output-nextp1
1.144     anton     924:     fill-state 
1.121     anton     925:     xt execute ;
1.120     anton     926: 
                    927: : output-c-tail ( -- )
1.124     anton     928:     tail-nextp2 @ output-c-tail1 ;
1.52      anton     929: 
1.108     anton     930: : output-c-tail2 ( -- )
1.120     anton     931:     ['] output-label2 output-c-tail1 ;
                    932: 
                    933: : output-c-tail-no-stores ( -- )
1.124     anton     934:     tail-nextp2 @ output-c-tail1-no-stores ;
1.119     anton     935: 
                    936: : output-c-tail2-no-stores ( -- )
1.120     anton     937:     ['] output-label2 output-c-tail1-no-stores ;
1.108     anton     938: 
1.85      anton     939: : type-c-code ( c-addr u xt -- )
1.109     anton     940:     \ like TYPE, but replaces "INST_TAIL;" with tail code produced by xt
1.85      anton     941:     { xt }
1.111     anton     942:     ." {" cr
                    943:     ." #line " c-line @ . quote c-filename 2@ type quote cr
1.52      anton     944:     begin ( c-addr1 u1 )
1.109     anton     945:        2dup s" INST_TAIL;" search
1.52      anton     946:     while ( c-addr1 u1 c-addr3 u3 )
                    947:        2dup 2>r drop nip over - type
1.85      anton     948:        xt execute
1.109     anton     949:        2r> 10 /string
1.52      anton     950:        \ !! resync #line missing
                    951:     repeat
1.111     anton     952:     2drop type
                    953:     ." #line " out-nls @ 2 + . quote out-filename 2@ type quote cr
                    954:     ." }" cr ;
1.63      anton     955: 
1.72      anton     956: : print-entry ( -- )
1.109     anton     957:     ." LABEL(" prim prim-c-name 2@ type ." )" ;
1.63      anton     958:     
1.43      jwilke    959: : output-c ( -- ) 
1.111     anton     960:     print-entry ."  /* " prim prim-name 2@ type ."  ( " prim prim-stack-string 2@ type ." ) */" cr
                    961:     ." /* " prim prim-doc 2@ type ."  */" cr
                    962:     ." NAME(" quote prim prim-name 2@ type quote ." )" cr \ debugging
                    963:     ." {" cr
                    964:     ." DEF_CA" cr
                    965:     print-declarations
1.145   ! anton     966:     output-nextp0
1.144     anton     967:     spill-state
1.111     anton     968:     fetches
                    969:     print-debug-args
                    970:     stack-pointer-updates
                    971:     prim prim-c-code 2@ ['] output-c-tail type-c-code
                    972:     output-c-tail2
                    973:     ." }" cr
                    974:     cr
1.1       anton     975: ;
                    976: 
1.56      anton     977: : disasm-arg { item -- }
                    978:     item item-stack @ inst-stream = if
1.107     anton     979:        ." {" cr
                    980:        item print-declaration
                    981:        item fetch
                    982:        item print-debug-arg
                    983:        ." }" cr
1.56      anton     984:     endif ;
                    985: 
                    986: : disasm-args ( -- )
1.69      anton     987:     prim prim-effect-in prim prim-effect-in-end @ ['] disasm-arg map-items ;
1.56      anton     988: 
                    989: : output-disasm ( -- )
                    990:     \ generate code for disassembling VM instructions
1.106     anton     991:     ." if (VM_IS_INST(*ip, " function-number @ 0 .r ." )) {" cr
1.69      anton     992:     ."   fputs(" quote prim prim-name 2@ type quote ." , vm_out);" cr
1.56      anton     993:     disasm-args
                    994:     ."   ip += " inst-stream stack-in @ 1+ 0 .r ." ;" cr
1.91      anton     995:     ."   goto _endif_;" cr
                    996:     ." }" cr ;
1.56      anton     997: 
1.86      anton     998: : output-profile ( -- )
                    999:     \ generate code for postprocessing the VM block profile stuff
1.87      anton    1000:     ." if (VM_IS_INST(*ip, " function-number @ 0 .r ." )) {" cr
1.104     anton    1001:     ."   add_inst(b, " quote prim prim-name 2@ type quote ." );" cr
1.86      anton    1002:     ."   ip += " inst-stream stack-in @ 1+ 0 .r ." ;" cr
                   1003:     prim prim-c-code 2@  s" SET_IP"    search nip nip
                   1004:     prim prim-c-code 2@  s" SUPER_END" search nip nip or if
                   1005:        ."   return;" cr
1.91      anton    1006:     else
                   1007:        ."   goto _endif_;" cr
1.86      anton    1008:     endif
1.91      anton    1009:     ." }" cr ;
1.86      anton    1010: 
1.114     anton    1011: : output-profile-part ( p )
                   1012:     ."   add_inst(b, " quote
                   1013:     prim-name 2@ type
                   1014:     quote ." );" cr ;
                   1015:     
1.104     anton    1016: : output-profile-combined ( -- )
                   1017:     \ generate code for postprocessing the VM block profile stuff
                   1018:     ." if (VM_IS_INST(*ip, " function-number @ 0 .r ." )) {" cr
1.114     anton    1019:     ['] output-profile-part map-combined
1.104     anton    1020:     ."   ip += " inst-stream stack-in @ 1+ 0 .r ." ;" cr
                   1021:     combined-prims num-combined @ 1- th @ prim-c-code 2@  s" SET_IP"    search nip nip
                   1022:     combined-prims num-combined @ 1- th @ prim-c-code 2@  s" SUPER_END" search nip nip or if
                   1023:        ."   return;" cr
                   1024:     else
                   1025:        ."   goto _endif_;" cr
                   1026:     endif
                   1027:     ." }" cr ;
                   1028: 
1.143     anton    1029: : prim-branch? { prim -- f }
                   1030:     \ true if prim is a branch or super-end
                   1031:     prim prim-c-code 2@  s" SET_IP" search nip nip 0<> ;
                   1032: 
1.103     anton    1033: : output-superend ( -- )
                   1034:     \ output flag specifying whether the current word ends a dynamic superinst
1.143     anton    1035:     prim prim-branch?
                   1036:     prim prim-c-code 2@  s" SUPER_END" search nip nip 0<> or
1.103     anton    1037:     prim prim-c-code 2@  s" SUPER_CONTINUE" search nip nip 0= and
                   1038:     negate 0 .r ." , /* " prim prim-name 2@ type ."  */" cr ;
                   1039: 
1.60      anton    1040: : gen-arg-parm { item -- }
                   1041:     item item-stack @ inst-stream = if
                   1042:        ." , " item item-type @ type-c-name 2@ type space
                   1043:        item item-name 2@ type
                   1044:     endif ;
                   1045: 
                   1046: : gen-args-parm ( -- )
1.69      anton    1047:     prim prim-effect-in prim prim-effect-in-end @ ['] gen-arg-parm map-items ;
1.60      anton    1048: 
                   1049: : gen-arg-gen { item -- }
                   1050:     item item-stack @ inst-stream = if
                   1051:        ."   genarg_" item item-type @ print-type-prefix
                   1052:         ." (ctp, " item item-name 2@ type ." );" cr
                   1053:     endif ;
                   1054: 
                   1055: : gen-args-gen ( -- )
1.69      anton    1056:     prim prim-effect-in prim prim-effect-in-end @ ['] gen-arg-gen map-items ;
1.60      anton    1057: 
                   1058: : output-gen ( -- )
                   1059:     \ generate C code for generating VM instructions
1.69      anton    1060:     ." void gen_" prim prim-c-name 2@ type ." (Inst **ctp" gen-args-parm ." )" cr
1.60      anton    1061:     ." {" cr
                   1062:     ."   gen_inst(ctp, vm_prim[" function-number @ 0 .r ." ]);" cr
                   1063:     gen-args-gen
1.68      anton    1064:     ." }" cr ;
1.60      anton    1065: 
1.49      anton    1066: : stack-used? { stack -- f }
                   1067:     stack stack-in @ stack stack-out @ or 0<> ;
1.44      jwilke   1068: 
1.30      pazsan   1069: : output-funclabel ( -- )
1.69      anton    1070:   ." &I_" prim prim-c-name 2@ type ." ," cr ;
1.30      pazsan   1071: 
                   1072: : output-forthname ( -- )
1.69      anton    1073:   '" emit prim prim-name 2@ type '" emit ." ," cr ;
1.30      pazsan   1074: 
1.92      anton    1075: \  : output-c-func ( -- )
                   1076: \  \ used for word libraries
                   1077: \      ." Cell * I_" prim prim-c-name 2@ type ." (Cell *SP, Cell **FP)      /* " prim prim-name 2@ type
                   1078: \      ."  ( " prim prim-stack-string 2@ type ."  ) */" cr
                   1079: \      ." /* " prim prim-doc 2@ type ."  */" cr
                   1080: \      ." NAME(" quote prim prim-name 2@ type quote ." )" cr
                   1081: \      \ debugging
                   1082: \      ." {" cr
                   1083: \      print-declarations
                   1084: \      \ !! don't know what to do about that
                   1085: \      inst-stream  stack-used? IF ." Cell *ip=IP;" cr THEN
                   1086: \      data-stack   stack-used? IF ." Cell *sp=SP;" cr THEN
                   1087: \      fp-stack     stack-used? IF ." Cell *fp=*FP;" cr THEN
                   1088: \      return-stack stack-used? IF ." Cell *rp=*RP;" cr THEN
1.144     anton    1089: \      spill-state
1.92      anton    1090: \      fetches
                   1091: \      stack-pointer-updates
                   1092: \      fp-stack   stack-used? IF ." *FP=fp;" cr THEN
                   1093: \      ." {" cr
                   1094: \      ." #line " c-line @ . quote c-filename 2@ type quote cr
                   1095: \      prim prim-c-code 2@ type
                   1096: \      ." }" cr
                   1097: \      stores
1.144     anton    1098: \      fill-state
1.92      anton    1099: \      ." return (sp);" cr
                   1100: \      ." }" cr
                   1101: \      cr ;
1.30      pazsan   1102: 
1.43      jwilke   1103: : output-label ( -- )  
1.127     anton    1104:     ." INST_ADDR(" prim prim-c-name 2@ type ." )," cr ;
1.1       anton    1105: 
1.43      jwilke   1106: : output-alias ( -- ) 
1.69      anton    1107:     ( primitive-number @ . ." alias " ) ." Primitive " prim prim-name 2@ type cr ;
1.1       anton    1108: 
1.122     anton    1109: : output-c-prim-num ( -- )
1.141     anton    1110:     ." N_" prim prim-c-name 2@ type ." ," cr ;
1.114     anton    1111: 
1.43      jwilke   1112: : output-forth ( -- )  
1.69      anton    1113:     prim prim-forth-code @ 0=
1.30      pazsan   1114:     IF         \ output-alias
1.28      jwilke   1115:        \ this is bad for ec: an alias is compiled if tho word does not exist!
                   1116:        \ JAW
1.69      anton    1117:     ELSE  ." : " prim prim-name 2@ type ."   ( "
                   1118:        prim prim-stack-string 2@ type ." )" cr
                   1119:        prim prim-forth-code 2@ type cr
1.30      pazsan   1120:     THEN ;
1.10      anton    1121: 
1.17      anton    1122: : output-tag-file ( -- )
1.117     pazsan   1123:     name-filename 2@ last-name-filename 2@ compare if
1.17      anton    1124:        name-filename 2@ last-name-filename 2!
                   1125:        #ff emit cr
                   1126:        name-filename 2@ type
                   1127:        ." ,0" cr
                   1128:     endif ;
                   1129: 
                   1130: : output-tag ( -- )
                   1131:     output-tag-file
1.69      anton    1132:     prim prim-name 2@ 1+ type
1.17      anton    1133:     127 emit
1.69      anton    1134:     space prim prim-name 2@ type space
1.17      anton    1135:     1 emit
                   1136:     name-line @ 0 .r
                   1137:     ." ,0" cr ;
                   1138: 
1.100     pazsan   1139: : output-vi-tag ( -- )
                   1140:     name-filename 2@ type #tab emit
                   1141:     prim prim-name 2@ type #tab emit
                   1142:     ." /^" prim prim-name 2@ type ."  *(/" cr ;
                   1143: 
1.10      anton    1144: [IFDEF] documentation
                   1145: : register-doc ( -- )
1.82      anton    1146:     prim prim-name 2@ documentation ['] create insert-wordlist
1.69      anton    1147:     prim prim-name 2@ 2,
                   1148:     prim prim-stack-string 2@ condition-stack-effect 2,
                   1149:     prim prim-wordset 2@ 2,
                   1150:     prim prim-c-name 2@ condition-pronounciation 2,
1.82      anton    1151:     prim prim-doc 2@ 2, ;
1.10      anton    1152: [THEN]
1.67      anton    1153: 
                   1154: 
1.69      anton    1155: \ combining instructions
                   1156: 
                   1157: \ The input should look like this:
                   1158: 
                   1159: \ lit_+ = lit +
                   1160: 
                   1161: \ The output should look like this:
                   1162: 
                   1163: \  I_lit_+:
                   1164: \  {
                   1165: \  DEF_CA
                   1166: \  Cell _x_ip0;
                   1167: \  Cell _x_sp0;
                   1168: \  Cell _x_sp1;
                   1169: \  NEXT_P0;
                   1170: \  _x_ip0 = (Cell) IPTOS;
                   1171: \  _x_sp0 = (Cell) spTOS;
                   1172: \  INC_IP(1);
                   1173: \  /* sp += 0; */
                   1174: \  /* lit ( #w -- w ) */
                   1175: \  /*  */
                   1176: \  NAME("lit")
                   1177: \  {
                   1178: \  Cell w;
                   1179: \  w = (Cell) _x_ip0;
                   1180: \  #ifdef VM_DEBUG
                   1181: \  if (vm_debug) {
                   1182: \  fputs(" w=", vm_out); printarg_w (w);
                   1183: \  fputc('\n', vm_out);
                   1184: \  }
                   1185: \  #endif
                   1186: \  {
                   1187: \  #line 136 "./prim"
                   1188: \  }
                   1189: \  _x_sp1 = (Cell)w;
                   1190: \  }
                   1191: \  I_plus:     /* + ( n1 n2 -- n ) */
                   1192: \  /*  */
                   1193: \  NAME("+")
                   1194: \  {
                   1195: \  DEF_CA
                   1196: \  Cell n1;
                   1197: \  Cell n2;
                   1198: \  Cell n;
                   1199: \  NEXT_P0;
                   1200: \  n1 = (Cell) _x_sp0;
                   1201: \  n2 = (Cell) _x_sp1;
                   1202: \  #ifdef VM_DEBUG
                   1203: \  if (vm_debug) {
                   1204: \  fputs(" n1=", vm_out); printarg_n (n1);
                   1205: \  fputs(" n2=", vm_out); printarg_n (n2);
                   1206: \  fputc('\n', vm_out);
                   1207: \  }
                   1208: \  #endif
                   1209: \  {
                   1210: \  #line 516 "./prim"
                   1211: \  n = n1+n2;
                   1212: \  }
                   1213: \  _x_sp0 = (Cell)n;
                   1214: \  }
                   1215: \  NEXT_P1;
                   1216: \  spTOS = (Cell)_x_sp0;
                   1217: \  NEXT_P2;
                   1218: 
1.71      anton    1219: : init-combined ( -- )
1.79      anton    1220:     prim to combined
1.71      anton    1221:     0 num-combined !
                   1222:     current-depth max-stacks cells erase
1.116     anton    1223:     include-skipped-insts @ current-depth 0 th !
1.72      anton    1224:     max-depth     max-stacks cells erase
                   1225:     min-depth     max-stacks cells erase
                   1226:     prim prim-effect-in  prim prim-effect-in-end  !
                   1227:     prim prim-effect-out prim prim-effect-out-end ! ;
1.71      anton    1228: 
                   1229: : max! ( n addr -- )
                   1230:     tuck @ max swap ! ;
                   1231: 
1.72      anton    1232: : min! ( n addr -- )
                   1233:     tuck @ min swap ! ;
                   1234: 
1.119     anton    1235: : inst-stream-adjustment ( nstack -- n )
                   1236:     \ number of stack items to add for each part
                   1237:     0= include-skipped-insts @ and negate ;
1.116     anton    1238: 
1.71      anton    1239: : add-depths { p -- }
                   1240:     \ combine stack effect of p with *-depths
                   1241:     max-stacks 0 ?do
1.72      anton    1242:        current-depth i th @
1.119     anton    1243:        p prim-stacks-in  i th @ + i inst-stream-adjustment +
1.72      anton    1244:        dup max-depth i th max!
                   1245:        p prim-stacks-out i th @ -
                   1246:        dup min-depth i th min!
                   1247:        current-depth i th !
1.71      anton    1248:     loop ;
                   1249: 
1.118     anton    1250: : copy-maxdepths ( n -- )
                   1251:     max-depth max-depths rot max-stacks * th max-stacks cells move ;
                   1252: 
1.71      anton    1253: : add-prim ( addr u -- )
                   1254:     \ add primitive given by "addr u" to combined-prims
                   1255:     primitives search-wordlist s" unknown primitive" ?print-error
                   1256:     execute { p }
1.72      anton    1257:     p combined-prims num-combined @ th !
1.118     anton    1258:     num-combined @ copy-maxdepths
1.71      anton    1259:     1 num-combined +!
1.118     anton    1260:     p add-depths
                   1261:     num-combined @ copy-maxdepths ;
1.71      anton    1262: 
                   1263: : compute-effects { q -- }
                   1264:     \ compute the stack effects of q from the depths
                   1265:     max-stacks 0 ?do
1.72      anton    1266:        max-depth i th @ dup
                   1267:        q prim-stacks-in i th !
                   1268:        current-depth i th @ -
                   1269:        q prim-stacks-out i th !
                   1270:     loop ;
                   1271: 
                   1272: : make-effect-items { stack# items effect-endp -- }
                   1273:     \ effect-endp points to a pointer to the end of the current item-array
                   1274:     \ and has to be updated
                   1275:     stacks stack# th @ { stack }
                   1276:     items 0 +do
                   1277:        effect-endp @ { item }
                   1278:        i 0 <# #s stack stack-pointer 2@ holds [char] _ hold #> save-mem
                   1279:        item item-name 2!
                   1280:        stack item item-stack !
1.74      anton    1281:        stack stack-type @ item item-type !
1.72      anton    1282:        i item item-offset !
                   1283:        item item-first on
                   1284:        item% %size effect-endp +!
                   1285:     loop ;
                   1286: 
                   1287: : init-effects { q -- }
                   1288:     \ initialize effects field for FETCHES and STORES
                   1289:     max-stacks 0 ?do
                   1290:        i q prim-stacks-in  i th @ q prim-effect-in-end  make-effect-items
                   1291:        i q prim-stacks-out i th @ q prim-effect-out-end make-effect-items
1.71      anton    1292:     loop ;
                   1293: 
1.119     anton    1294: : compute-stack-max-back-depths ( stack -- )
                   1295:     stack-number @ { stack# }
                   1296:     current-depth stack# th @ dup
                   1297:     dup stack# num-combined @ s-c-max-back-depth !
                   1298:     -1 num-combined @ 1- -do ( max-depth current-depth )
                   1299:        combined-prims i th @ { p }
                   1300:        p prim-stacks-out stack# th @ +
                   1301:        dup >r max r>
                   1302:        over stack# i s-c-max-back-depth !
                   1303:        p prim-stacks-in stack# th @ -
                   1304:        stack# inst-stream-adjustment -
                   1305:     1 -loop
                   1306:     assert( dup stack# inst-stream-adjustment negate = )
                   1307:     assert( over max-depth stack# th @ = )
                   1308:     2drop ;
                   1309: 
                   1310: : compute-max-back-depths ( -- )
                   1311:     \ compute max-back-depths.
                   1312:     \ assumes that current-depths is correct for the end of the combination
                   1313:     ['] compute-stack-max-back-depths map-stacks ;
                   1314: 
1.71      anton    1315: : process-combined ( -- )
1.81      anton    1316:     combined combined-prims num-combined @ cells
1.82      anton    1317:     combinations ['] constant insert-wordlist
1.86      anton    1318:     combined-prims num-combined @ 1- th ( last-part )
                   1319:     @ prim-c-code 2@ prim prim-c-code 2! \ used by output-super-end
1.72      anton    1320:     prim compute-effects
                   1321:     prim init-effects
1.119     anton    1322:     compute-max-back-depths
1.72      anton    1323:     output-combined perform ;
                   1324: 
1.144     anton    1325: \ reprocessing (typically to generate versions for another cache states)
                   1326: \ !! use prim-context
                   1327: 
                   1328: variable reprocessed-num 0 reprocessed-num !
                   1329: 
                   1330: : new-name ( -- c-addr u )
                   1331:     reprocessed-num @ 0
                   1332:     1 reprocessed-num +!
                   1333:     <# #s 'p hold '_ hold #> save-mem ;
                   1334: 
                   1335: : reprocess-simple ( prim -- )
                   1336:     to prim
                   1337:     new-name prim prim-c-name 2!
                   1338:     output @ execute ;
                   1339: 
                   1340: : lookup-prim ( c-addr u -- prim )
                   1341:     primitives search-wordlist 0= -13 and throw execute ;
                   1342: 
                   1343: : state-prim1 { in-state out-state prim -- }
                   1344:     in-state out-state state-default dup d= ?EXIT
                   1345:     in-state  to state-in
                   1346:     out-state to state-out
                   1347:     prim reprocess-simple ;
                   1348: 
                   1349: : state-prim ( in-state out-state "name" -- )
                   1350:     parse-word lookup-prim state-prim1 ;
                   1351: 
                   1352: \ reprocessing with default states
                   1353: 
                   1354: \ This is a simple scheme and should be generalized
                   1355: \ assumes we only cache one stack and use simple states for that
                   1356: 
                   1357: 0 value cache-stack  \ stack that we cache
                   1358: 2variable cache-states \ states of the cache, starting with the empty state
                   1359: 
                   1360: : compute-default-state-out ( n-in -- n-out )
                   1361:     \ for the current prim
                   1362:     cache-stack stack-in @ - 0 max
                   1363:     cache-stack stack-out @ + cache-states 2@ nip 1- min ;
                   1364: 
                   1365: : gen-prim-states ( prim -- )
                   1366:     to prim
                   1367:     cache-states 2@ swap { states } ( nstates )
                   1368:     cache-stack stack-in @ +do
                   1369:        states i th @
                   1370:        states i compute-default-state-out th @
                   1371:        prim state-prim1
                   1372:     loop ;
                   1373: 
                   1374: : prim-states ( "name" -- )
                   1375:     parse-word lookup-prim gen-prim-states ;
                   1376: 
                   1377: : gen-branch-states ( prim -- )
                   1378:     \ generate versions that produce state-default; useful for branches
                   1379:     to prim
                   1380:     cache-states 2@ swap { states } ( nstates )
                   1381:     cache-stack stack-in @ +do
                   1382:        states i th @ state-default prim state-prim1
                   1383:     loop ;
                   1384: 
                   1385: : branch-states ( out-state "name" -- )
                   1386:     parse-word lookup-prim gen-branch-states ;
                   1387: 
                   1388: \ producing state transitions
                   1389: 
                   1390: : gen-transitions ( "name" -- )
                   1391:     parse-word lookup-prim { prim }
                   1392:     cache-states 2@ { states nstates }
                   1393:     nstates 0 +do
                   1394:        nstates 0 +do
                   1395:            i j <> if
                   1396:                states i th @ states j th @ prim state-prim1
                   1397:            endif
                   1398:        loop
                   1399:     loop ;
                   1400: 
1.72      anton    1401: \ C output
                   1402: 
                   1403: : print-item { n stack -- }
                   1404:     \ print nth stack item name
1.79      anton    1405:     stack stack-type @ type-c-name 2@ type space
1.142     anton    1406:     ." MAYBE_UNUSED _" stack stack-pointer 2@ type n 0 .r ;
1.72      anton    1407: 
                   1408: : print-declarations-combined ( -- )
                   1409:     max-stacks 0 ?do
                   1410:        max-depth i th @ min-depth i th @ - 0 +do
                   1411:            i stacks j th @ print-item ." ;" cr
                   1412:        loop
                   1413:     loop ;
1.79      anton    1414: 
                   1415: : part-fetches ( -- )
                   1416:     fetches ;
                   1417: 
                   1418: : part-output-c-tail ( -- )
1.91      anton    1419:     print-debug-results
1.85      anton    1420:     stores ;
                   1421: 
                   1422: : output-combined-tail ( -- )
                   1423:     part-output-c-tail
                   1424:     in-part @ >r in-part off
1.119     anton    1425:     combined ['] output-c-tail-no-stores prim-context
1.118     anton    1426:     r> in-part ! ;
                   1427: 
                   1428: : part-stack-pointer-updates ( -- )
1.123     anton    1429:     next-stack-number @ 0 +do
1.118     anton    1430:        i part-num @ 1+ s-c-max-depth @ dup
                   1431:        i num-combined @ s-c-max-depth @ =    \ final depth
                   1432:        swap i part-num @ s-c-max-depth @ <> \ just reached now
                   1433:        part-num @ 0= \ first part
                   1434:        or and if
                   1435:            stacks i th @ stack-pointer-update
                   1436:        endif
                   1437:     loop ;
1.79      anton    1438: 
                   1439: : output-part ( p -- )
                   1440:     to prim
                   1441:     ." /* " prim prim-name 2@ type ."  ( " prim prim-stack-string 2@ type ." ) */" cr
                   1442:     ." NAME(" quote prim prim-name 2@ type quote ." )" cr \ debugging
                   1443:     ." {" cr
                   1444:     print-declarations
                   1445:     part-fetches
                   1446:     print-debug-args
1.118     anton    1447:     combined ['] part-stack-pointer-updates prim-context
                   1448:     1 part-num +!
1.79      anton    1449:     prim add-depths \ !! right place?
1.85      anton    1450:     prim prim-c-code 2@ ['] output-combined-tail type-c-code
1.79      anton    1451:     part-output-c-tail
                   1452:     ." }" cr ;
                   1453: 
1.74      anton    1454: : output-parts ( -- )
1.79      anton    1455:     prim >r in-part on
                   1456:     current-depth max-stacks cells erase
1.118     anton    1457:     0 part-num !
1.114     anton    1458:     ['] output-part map-combined
1.79      anton    1459:     in-part off
1.74      anton    1460:     r> to prim ;
                   1461: 
1.72      anton    1462: : output-c-combined ( -- )
                   1463:     print-entry cr
1.74      anton    1464:     \ debugging messages just in parts
1.72      anton    1465:     ." {" cr
                   1466:     ." DEF_CA" cr
                   1467:     print-declarations-combined
1.145   ! anton    1468:     output-nextp0
1.144     anton    1469:     spill-state
1.118     anton    1470:     \ fetches \ now in parts
1.74      anton    1471:     \ print-debug-args
1.118     anton    1472:     \ stack-pointer-updates now in parts
1.74      anton    1473:     output-parts
1.119     anton    1474:     output-c-tail2-no-stores
1.74      anton    1475:     ." }" cr
                   1476:     cr ;
1.72      anton    1477: 
                   1478: : output-forth-combined ( -- )
1.81      anton    1479: ;
                   1480: 
                   1481: 
1.83      anton    1482: \ peephole optimization rules
1.81      anton    1483: 
1.114     anton    1484: \ data for a simple peephole optimizer that always tries to combine
                   1485: \ the currently compiled instruction with the last one.
                   1486: 
1.81      anton    1487: \ in order for this to work as intended, shorter combinations for each
                   1488: \ length must be present, and the longer combinations must follow
                   1489: \ shorter ones (this restriction may go away in the future).
                   1490:   
1.83      anton    1491: : output-peephole ( -- )
1.81      anton    1492:     combined-prims num-combined @ 1- cells combinations search-wordlist
1.114     anton    1493:     s" the prefix for this superinstruction must be defined earlier" ?print-error
1.82      anton    1494:     ." {"
                   1495:     execute prim-num @ 5 .r ." ,"
                   1496:     combined-prims num-combined @ 1- th @ prim-num @ 5 .r ." ,"
                   1497:     combined prim-num @ 5 .r ." }, /* "
                   1498:     combined prim-c-name 2@ type ."  */"
                   1499:     cr ;
                   1500: 
1.114     anton    1501: 
1.115     anton    1502: \ cost and superinstruction data for a sophisticated combiner (e.g.,
                   1503: \ shortest path)
1.114     anton    1504: 
                   1505: \ This is intended as initializer for a structure like this
                   1506: 
1.116     anton    1507: \  struct cost {
1.114     anton    1508: \    int loads;       /* number of stack loads */
                   1509: \    int stores;      /* number of stack stores */
                   1510: \    int updates;     /* number of stack pointer updates */
1.138     anton    1511: \    int offset;      /* offset into super2 table */
1.114     anton    1512: \    int length;      /* number of components */
                   1513: \  };
                   1514: 
1.115     anton    1515: \ How do you know which primitive or combined instruction this
                   1516: \ structure refers to?  By the order of cost structures, as in most
                   1517: \ other cases.
                   1518: 
1.139     anton    1519: : super2-length ( -- n )
                   1520:     combined if
                   1521:        num-combined @
                   1522:     else
                   1523:        1
                   1524:     endif ;
                   1525: 
1.115     anton    1526: : compute-costs { p -- nloads nstores nupdates }
                   1527:     \ compute the number of loads, stores, and stack pointer updates
                   1528:     \ of a primitive or combined instruction; does not take TOS
1.139     anton    1529:     \ caching into account
1.115     anton    1530:     0 max-stacks 0 +do
                   1531:        p prim-stacks-in i th @ +
                   1532:     loop
1.139     anton    1533:     super2-length 1- - \ don't count instruction fetches of subsumed insts
1.115     anton    1534:     0 max-stacks 0 +do
                   1535:        p prim-stacks-out i th @ +
                   1536:     loop
1.139     anton    1537:     0 max-stacks 1 +do \ don't count ip updates, therefore "1 +do"
1.115     anton    1538:        p prim-stacks-in i th @ p prim-stacks-out i th @ <> -
                   1539:     loop ;
1.114     anton    1540: 
                   1541: : output-num-part ( p -- )
1.141     anton    1542:     ." N_" prim-c-name 2@ type ." ," ;
                   1543:     \ prim-num @ 4 .r ." ," ;
1.138     anton    1544: 
                   1545: : output-name-comment ( -- )
                   1546:     ."  /* " prim prim-name 2@ type ."  */" ;
                   1547: 
                   1548: variable offset-super2  0 offset-super2 ! \ offset into the super2 table
1.141     anton    1549: 
1.143     anton    1550: : output-costs-prefix ( -- )
                   1551:     ." {" prim compute-costs
                   1552:     rot 2 .r ." ," swap 2 .r ." ," 2 .r ." , "
1.144     anton    1553:     prim prim-branch? negate . ." ,"
                   1554:     state-in  state-number @ 2 .r ." ,"
                   1555:     state-out state-number @ 2 .r ." ," ;
1.143     anton    1556: 
1.141     anton    1557: : output-costs-gforth-simple ( -- )
1.143     anton    1558:     output-costs-prefix
1.141     anton    1559:     prim output-num-part
                   1560:     1 2 .r ." },"
                   1561:     output-name-comment
                   1562:     cr ;
                   1563: 
                   1564: : output-costs-gforth-combined ( -- )
1.143     anton    1565:     output-costs-prefix
1.141     anton    1566:     ." N_START_SUPER+" offset-super2 @ 5 .r ." ,"
                   1567:     super2-length dup 2 .r ." }," offset-super2 +!
                   1568:     output-name-comment
                   1569:     cr ;
1.138     anton    1570: 
1.115     anton    1571: : output-costs ( -- )
1.138     anton    1572:     \ description of superinstructions and simple instructions
1.115     anton    1573:     ." {" prim compute-costs
                   1574:     rot 2 .r ." ," swap 2 .r ." ," 2 .r ." ,"
1.138     anton    1575:     offset-super2 @ 5 .r ." ,"
                   1576:     super2-length dup 2 .r ." }," offset-super2 +!
                   1577:     output-name-comment
                   1578:     cr ;
                   1579: 
                   1580: : output-super2 ( -- )
                   1581:     \ table of superinstructions without requirement for existing prefixes
1.115     anton    1582:     combined if
1.138     anton    1583:        ['] output-num-part map-combined 
1.115     anton    1584:     else
1.138     anton    1585:        prim output-num-part
1.115     anton    1586:     endif
1.138     anton    1587:     output-name-comment
                   1588:     cr ;   
1.69      anton    1589: 
1.67      anton    1590: \ the parser
                   1591: 
                   1592: eof-char max-member \ the whole character set + EOF
                   1593: 
                   1594: : getinput ( -- n )
                   1595:  rawinput @ endrawinput @ =
                   1596:  if
                   1597:    eof-char
                   1598:  else
                   1599:    cookedinput @ c@
                   1600:  endif ;
                   1601: 
                   1602: :noname ( n -- )
                   1603:  dup bl > if
                   1604:   emit space
                   1605:  else
                   1606:   .
                   1607:  endif ;
                   1608: print-token !
                   1609: 
                   1610: : testchar? ( set -- f )
                   1611:  getinput member? ;
                   1612: ' testchar? test-vector !
                   1613: 
1.130     anton    1614: : checksynclines ( -- )
1.67      anton    1615:     \ when input points to a newline, check if the next line is a
                   1616:     \ sync line.  If it is, perform the appropriate actions.
1.131     anton    1617:     rawinput @ begin >r
1.130     anton    1618:        s" #line " r@ over compare if
                   1619:            rdrop 1 line +! EXIT
                   1620:        endif
                   1621:        0. r> 6 chars + 20 >number drop >r drop line ! r> ( c-addr )
                   1622:        dup c@ bl = if
                   1623:            char+ dup c@ [char] " <> 0= s" sync line syntax" ?print-error
                   1624:            char+ dup 100 [char] " scan drop swap 2dup - save-mem filename 2!
                   1625:            char+
                   1626:        endif
                   1627:        dup c@ nl-char <> 0= s" sync line syntax" ?print-error
                   1628:        skipsynclines @ if
1.131     anton    1629:            char+ dup rawinput !
1.130     anton    1630:            rawinput @ c@ cookedinput @ c!
                   1631:        endif
                   1632:     again ;
1.67      anton    1633: 
                   1634: : ?nextchar ( f -- )
1.71      anton    1635:     s" syntax error, wrong char" ?print-error
1.67      anton    1636:     rawinput @ endrawinput @ <> if
                   1637:        rawinput @ c@
                   1638:        1 chars rawinput +!
                   1639:        1 chars cookedinput +!
                   1640:        nl-char = if
1.130     anton    1641:            checksynclines
1.67      anton    1642:            rawinput @ line-start !
                   1643:        endif
1.130     anton    1644:        rawinput @ c@
                   1645:        cookedinput @ c!
1.67      anton    1646:     endif ;
                   1647: 
                   1648: : charclass ( set "name" -- )
                   1649:  ['] ?nextchar terminal ;
                   1650: 
                   1651: : .. ( c1 c2 -- set )
                   1652:  ( creates a set that includes the characters c, c1<=c<=c2 )
                   1653:  empty copy-set
                   1654:  swap 1+ rot do
                   1655:   i over add-member
                   1656:  loop ;
                   1657: 
                   1658: : ` ( -- terminal ) ( use: ` c )
                   1659:  ( creates anonymous terminal for the character c )
                   1660:  char singleton ['] ?nextchar make-terminal ;
                   1661: 
                   1662: char a char z ..  char A char Z ..  union char _ singleton union  charclass letter
                   1663: char 0 char 9 ..                                       charclass digit
                   1664: bl singleton tab-char over add-member                  charclass white
                   1665: nl-char singleton eof-char over add-member complement  charclass nonl
                   1666: nl-char singleton eof-char over add-member
                   1667:     char : over add-member complement                   charclass nocolonnl
1.110     anton    1668: nl-char singleton eof-char over add-member
                   1669:     char } over add-member complement                   charclass nobracenl
1.67      anton    1670: bl 1+ maxchar .. char \ singleton complement intersection
                   1671:                                                         charclass nowhitebq
                   1672: bl 1+ maxchar ..                                        charclass nowhite
                   1673: char " singleton eof-char over add-member complement   charclass noquote
                   1674: nl-char singleton                                      charclass nl
                   1675: eof-char singleton                                     charclass eof
1.79      anton    1676: nl-char singleton eof-char over add-member             charclass nleof
1.67      anton    1677: 
                   1678: (( letter (( letter || digit )) **
                   1679: )) <- c-ident ( -- )
                   1680: 
1.110     anton    1681: (( ` # ?? (( letter || digit || ` : )) ++
1.67      anton    1682: )) <- stack-ident ( -- )
                   1683: 
                   1684: (( nowhitebq nowhite ** ))
                   1685: <- forth-ident ( -- )
                   1686: 
                   1687: Variable forth-flag
                   1688: Variable c-flag
                   1689: 
                   1690: (( (( ` e || ` E )) {{ start }} nonl ** 
                   1691:    {{ end evaluate }}
                   1692: )) <- eval-comment ( ... -- ... )
                   1693: 
                   1694: (( (( ` f || ` F )) {{ start }} nonl ** 
                   1695:    {{ end forth-flag @ IF type cr ELSE 2drop THEN }}
                   1696: )) <- forth-comment ( -- )
                   1697: 
                   1698: (( (( ` c || ` C )) {{ start }} nonl ** 
                   1699:    {{ end c-flag @ IF type cr ELSE 2drop THEN }}
                   1700: )) <- c-comment ( -- )
                   1701: 
                   1702: (( ` - nonl ** {{ 
1.140     pazsan   1703:        forth-flag @ IF forth-fdiff ." [ELSE]" cr THEN
                   1704:        c-flag @ IF
                   1705:            function-diff
                   1706:            ." #else /* " function-number @ 0 .r ."  */" cr THEN }}
1.67      anton    1707: )) <- else-comment
                   1708: 
                   1709: (( ` + {{ start }} nonl ** {{ end
                   1710:        dup
                   1711:        IF      c-flag @
1.140     pazsan   1712:            IF
                   1713:                function-diff
                   1714:                ." #ifdef HAS_" bounds ?DO  I c@ toupper emit  LOOP cr
1.67      anton    1715:                THEN
                   1716:                forth-flag @
1.140     pazsan   1717:                IF  forth-fdiff  ." has? " type ."  [IF]"  cr THEN
1.67      anton    1718:        ELSE    2drop
1.140     pazsan   1719:            c-flag @      IF
                   1720:                function-diff  ." #endif" cr THEN
                   1721:            forth-flag @  IF  forth-fdiff  ." [THEN]"  cr THEN
1.67      anton    1722:        THEN }}
                   1723: )) <- if-comment
                   1724: 
1.98      pazsan   1725: (( (( ` g || ` G )) {{ start }} nonl **
                   1726:    {{ end
1.140     pazsan   1727:       forth-flag @ IF  forth-fdiff  ." group " type cr  THEN
                   1728:       c-flag @     IF  function-diff
                   1729:          ." GROUP(" type ." , " function-number @ 0 .r ." )" cr  THEN }}
1.98      pazsan   1730: )) <- group-comment
                   1731: 
                   1732: (( (( eval-comment || forth-comment || c-comment || else-comment || if-comment || group-comment )) ?? nonl ** )) <- comment-body
1.67      anton    1733: 
1.79      anton    1734: (( ` \ comment-body nleof )) <- comment ( -- )
1.67      anton    1735: 
                   1736: (( {{ start }} stack-ident {{ end 2 pick init-item item% %size + }} white ** )) **
                   1737: <- stack-items
                   1738: 
1.69      anton    1739: (( {{ prim prim-effect-in }}  stack-items {{ prim prim-effect-in-end ! }}
1.67      anton    1740:    ` - ` - white **
1.69      anton    1741:    {{ prim prim-effect-out }} stack-items {{ prim prim-effect-out-end ! }}
1.67      anton    1742: )) <- stack-effect ( -- )
                   1743: 
1.71      anton    1744: (( {{ prim create-prim }}
1.69      anton    1745:    ` ( white ** {{ start }} stack-effect {{ end prim prim-stack-string 2! }} ` ) white **
                   1746:    (( {{ start }} forth-ident {{ end prim prim-wordset 2! }} white **
                   1747:       (( {{ start }}  c-ident {{ end prim prim-c-name 2! }} )) ??
1.79      anton    1748:    )) ??  nleof
                   1749:    (( ` " ` "  {{ start }} (( noquote ++ ` " )) ++ {{ end 1- prim prim-doc 2! }} ` " white ** nleof )) ??
1.110     anton    1750:    {{ skipsynclines off line @ c-line ! filename 2@ c-filename 2! start }}
                   1751:    (( (( ` { nonl ** nleof (( (( nobracenl {{ line @ drop }} nonl ** )) ?? nleof )) ** ` } white ** nleof white ** ))
                   1752:    || (( nocolonnl nonl **  nleof white ** )) ** ))
                   1753:    {{ end prim prim-c-code 2! skipsynclines on }}
1.79      anton    1754:    (( ` :  white ** nleof
                   1755:       {{ start }} (( nonl ++  nleof white ** )) ++ {{ end prim prim-forth-code 2! }}
1.81      anton    1756:    )) ?? {{ process-simple }}
1.79      anton    1757:    nleof
1.69      anton    1758: )) <- simple-primitive ( -- )
                   1759: 
1.71      anton    1760: (( {{ init-combined }}
1.89      anton    1761:    ` = white ** (( {{ start }} forth-ident {{ end add-prim }} white ** )) ++
1.79      anton    1762:    nleof {{ process-combined }}
1.69      anton    1763: )) <- combined-primitive
                   1764: 
1.79      anton    1765: (( {{ make-prim to prim 0 to combined
1.69      anton    1766:       line @ name-line ! filename 2@ name-filename 2!
1.82      anton    1767:       function-number @ prim prim-num !
1.110     anton    1768:       start }} [ifdef] vmgen c-ident [else] forth-ident [then] {{ end
                   1769:       2dup prim prim-name 2! prim prim-c-name 2! }}  white **
1.104     anton    1770:    (( ` / white ** {{ start }} c-ident {{ end prim prim-c-name 2! }} white ** )) ??
1.138     anton    1771:    (( simple-primitive || combined-primitive ))
                   1772:    {{ 1 function-number +! }}
1.67      anton    1773: )) <- primitive ( -- )
                   1774: 
                   1775: (( (( comment || primitive || nl white ** )) ** eof ))
                   1776: parser primitives2something
                   1777: warnings @ [IF]
                   1778: .( parser generated ok ) cr
                   1779: [THEN]
                   1780: 
1.95      jwilke   1781: 
1.97      jwilke   1782: \ run with gforth-0.5.0 (slurp-file is missing)
1.95      jwilke   1783: [IFUNDEF] slurp-file
                   1784: : slurp-file ( c-addr1 u1 -- c-addr2 u2 )
                   1785:     \ c-addr1 u1 is the filename, c-addr2 u2 is the file's contents
                   1786:     r/o bin open-file throw >r
                   1787:     r@ file-size throw abort" file too large"
                   1788:     dup allocate throw swap
                   1789:     2dup r@ read-file throw over <> abort" could not read whole file"
                   1790:     r> close-file throw ;
                   1791: [THEN]
                   1792: 
1.69      anton    1793: : primfilter ( addr u -- )
                   1794:     \ process the string at addr u
                   1795:     over dup rawinput ! dup line-start ! cookedinput !
                   1796:     + endrawinput !
1.130     anton    1797:     checksynclines
1.69      anton    1798:     primitives2something ;    
1.8       pazsan   1799: 
1.130     anton    1800: : unixify ( c-addr u1 -- c-addr u2 )
                   1801:     \ delete crs from the string
                   1802:     bounds tuck tuck ?do ( c-addr1 )
                   1803:        i c@ dup #cr <> if
                   1804:            over c! char+
                   1805:        else
                   1806:            drop
                   1807:        endif
                   1808:     loop
                   1809:     over - ;
                   1810: 
1.72      anton    1811: : process-file ( addr u xt-simple x-combined -- )
                   1812:     output-combined ! output !
1.61      anton    1813:     save-mem 2dup filename 2!
1.130     anton    1814:     slurp-file unixify
1.17      anton    1815:     warnings @ if
                   1816:        ." ------------ CUT HERE -------------" cr  endif
1.69      anton    1817:     primfilter ;
1.30      pazsan   1818: 
1.72      anton    1819: \  : process      ( xt -- )
                   1820: \      bl word count rot
                   1821: \      process-file ;

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