1: \ Gforth primitives
2:
3: \ Copyright (C) 1995,1996,1997,1998,2000,2003,2004,2005,2006,2007 Free Software Foundation, Inc.
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 3
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, see http://www.gnu.org/licenses/.
19:
20:
21: \ WARNING: This file is processed by m4. Make sure your identifiers
22: \ don't collide with m4's (e.g. by undefining them).
23: \
24: \
25: \
26: \ This file contains primitive specifications in the following format:
27: \
28: \ forth name ( stack effect ) category [pronunciation]
29: \ [""glossary entry""]
30: \ C code
31: \ [:
32: \ Forth code]
33: \
34: \ Note: Fields in brackets are optional. Word specifications have to
35: \ be separated by at least one empty line
36: \
37: \ Both pronounciation and stack items (in the stack effect) must
38: \ conform to the C identifier syntax or the C compiler will complain.
39: \ If you don't have a pronounciation field, the Forth name is used,
40: \ and has to conform to the C identifier syntax.
41: \
42: \ These specifications are automatically translated into C-code for the
43: \ interpreter and into some other files. I hope that your C compiler has
44: \ decent optimization, otherwise the automatically generated code will
45: \ be somewhat slow. The Forth version of the code is included for manual
46: \ compilers, so they will need to compile only the important words.
47: \
48: \ Note that stack pointer adjustment is performed according to stack
49: \ effect by automatically generated code and NEXT is automatically
50: \ appended to the C code. Also, you can use the names in the stack
51: \ effect in the C code. Stack access is automatic. One exception: if
52: \ your code does not fall through, the results are not stored into the
53: \ stack. Use different names on both sides of the '--', if you change a
54: \ value (some stores to the stack are optimized away).
55: \
56: \ For superinstructions the syntax is:
57: \
58: \ forth-name [/ c-name] = forth-name forth-name ...
59: \
60: \
61: \ The stack variables have the following types:
62: \
63: \ name matches type
64: \ f.* Bool
65: \ c.* Char
66: \ [nw].* Cell
67: \ u.* UCell
68: \ d.* DCell
69: \ ud.* UDCell
70: \ r.* Float
71: \ a_.* Cell *
72: \ c_.* Char *
73: \ f_.* Float *
74: \ df_.* DFloat *
75: \ sf_.* SFloat *
76: \ xt.* XT
77: \ f83name.* F83Name *
78:
79: \E stack data-stack sp Cell
80: \E stack fp-stack fp Float
81: \E stack return-stack rp Cell
82: \E
83: \E get-current prefixes set-current
84: \E
85: \E s" Bool" single data-stack type-prefix f
86: \E s" Char" single data-stack type-prefix c
87: \E s" Cell" single data-stack type-prefix n
88: \E s" Cell" single data-stack type-prefix w
89: \E s" UCell" single data-stack type-prefix u
90: \E s" DCell" double data-stack type-prefix d
91: \E s" UDCell" double data-stack type-prefix ud
92: \E s" Float" single fp-stack type-prefix r
93: \E s" Cell *" single data-stack type-prefix a_
94: \E s" Char *" single data-stack type-prefix c_
95: \E s" Float *" single data-stack type-prefix f_
96: \E s" DFloat *" single data-stack type-prefix df_
97: \E s" SFloat *" single data-stack type-prefix sf_
98: \E s" Xt" single data-stack type-prefix xt
99: \E s" struct F83Name *" single data-stack type-prefix f83name
100: \E s" struct Longname *" single data-stack type-prefix longname
101: \E
102: \E data-stack stack-prefix S:
103: \E fp-stack stack-prefix F:
104: \E return-stack stack-prefix R:
105: \E inst-stream stack-prefix #
106: \E
107: \E set-current
108: \E store-optimization on
109: \E ' noop tail-nextp2 ! \ now INST_TAIL just stores, but does not jump
110: \E
111: \E `include-skipped-insts' on \ static superinsts include cells for components
112: \E \ useful for dynamic programming and
113: \E \ superinsts across entry points
114:
115: \
116: \
117: \
118: \ In addition the following names can be used:
119: \ ip the instruction pointer
120: \ sp the data stack pointer
121: \ rp the parameter stack pointer
122: \ lp the locals stack pointer
123: \ NEXT executes NEXT
124: \ cfa
125: \ NEXT1 executes NEXT1
126: \ FLAG(x) makes a Forth flag from a C flag
127: \
128: \
129: \
130: \ Percentages in comments are from Koopmans book: average/maximum use
131: \ (taken from four, not very representative benchmarks)
132: \
133: \
134: \
135: \ To do:
136: \
137: \ throw execute, cfa and NEXT1 out?
138: \ macroize *ip, ip++, *ip++ (pipelining)?
139:
140: \ Stack caching setup
141:
142: ifdef(`STACK_CACHE_FILE', `include(STACK_CACHE_FILE)', `include(cache0.vmg)')
143:
144: \ these m4 macros would collide with identifiers
145: undefine(`index')
146: undefine(`shift')
147: undefine(`symbols')
148:
149: \F 0 [if]
150:
151: \ run-time routines for non-primitives. They are defined as
152: \ primitives, because that simplifies things.
153:
154: (docol) ( -- R:a_retaddr ) gforth-internal paren_docol
155: ""run-time routine for colon definitions""
156: #ifdef NO_IP
157: a_retaddr = next_code;
158: INST_TAIL;
159: goto **(Label *)PFA(CFA);
160: #else /* !defined(NO_IP) */
161: a_retaddr = (Cell *)IP;
162: SET_IP((Xt *)PFA(CFA));
163: #endif /* !defined(NO_IP) */
164:
165: (docon) ( -- w ) gforth-internal paren_docon
166: ""run-time routine for constants""
167: w = *(Cell *)PFA(CFA);
168: #ifdef NO_IP
169: INST_TAIL;
170: goto *next_code;
171: #endif /* defined(NO_IP) */
172:
173: (dovar) ( -- a_body ) gforth-internal paren_dovar
174: ""run-time routine for variables and CREATEd words""
175: a_body = PFA(CFA);
176: #ifdef NO_IP
177: INST_TAIL;
178: goto *next_code;
179: #endif /* defined(NO_IP) */
180:
181: (douser) ( -- a_user ) gforth-internal paren_douser
182: ""run-time routine for constants""
183: a_user = (Cell *)(up+*(Cell *)PFA(CFA));
184: #ifdef NO_IP
185: INST_TAIL;
186: goto *next_code;
187: #endif /* defined(NO_IP) */
188:
189: (dodefer) ( -- ) gforth-internal paren_dodefer
190: ""run-time routine for deferred words""
191: #ifndef NO_IP
192: ip=IP; /* undo any ip updating that may have been performed by NEXT_P0 */
193: #endif /* !defined(NO_IP) */
194: SUPER_END; /* !! probably unnecessary and may lead to measurement errors */
195: VM_JUMP(EXEC1(*(Xt *)PFA(CFA)));
196:
197: (dofield) ( n1 -- n2 ) gforth-internal paren_field
198: ""run-time routine for fields""
199: n2 = n1 + *(Cell *)PFA(CFA);
200: #ifdef NO_IP
201: INST_TAIL;
202: goto *next_code;
203: #endif /* defined(NO_IP) */
204:
205: (dovalue) ( -- w ) gforth-internal paren_doval
206: ""run-time routine for constants""
207: w = *(Cell *)PFA(CFA);
208: #ifdef NO_IP
209: INST_TAIL;
210: goto *next_code;
211: #endif /* defined(NO_IP) */
212:
213: (dodoes) ( -- a_body R:a_retaddr ) gforth-internal paren_dodoes
214: ""run-time routine for @code{does>}-defined words""
215: #ifdef NO_IP
216: a_retaddr = next_code;
217: a_body = PFA(CFA);
218: INST_TAIL;
219: #ifdef DEBUG
220: fprintf(stderr, "dodoes to %x, push %x\n", a_retaddr, a_body);
221: #endif
222: goto **(Label *)DOES_CODE1(CFA);
223: #else /* !defined(NO_IP) */
224: a_retaddr = (Cell *)IP;
225: a_body = PFA(CFA);
226: #ifdef DEBUG
227: fprintf(stderr, "dodoes to %x, push %x\n", a_retaddr, a_body);
228: #endif
229: SET_IP(DOES_CODE1(CFA));
230: #endif /* !defined(NO_IP) */
231:
232: (does-handler) ( -- ) gforth-internal paren_does_handler
233: ""just a slot to have an encoding for the DOESJUMP,
234: which is no longer used anyway (!! eliminate this)""
235:
236: \F [endif]
237:
238: \g control
239:
240: noop ( -- ) gforth
241: :
242: ;
243:
244: call ( #a_callee -- R:a_retaddr ) new
245: ""Call callee (a variant of docol with inline argument).""
246: #ifdef NO_IP
247: assert(0);
248: INST_TAIL;
249: JUMP(a_callee);
250: #else
251: #ifdef DEBUG
252: {
253: CFA_TO_NAME((((Cell *)a_callee)-2));
254: fprintf(stderr,"%08lx: call %08lx %.*s\n",(Cell)ip,(Cell)a_callee,
255: len,name);
256: }
257: #endif
258: a_retaddr = (Cell *)IP;
259: SET_IP((Xt *)a_callee);
260: #endif
261:
262: execute ( xt -- ) core
263: ""Perform the semantics represented by the execution token, @i{xt}.""
264: #ifdef DEBUG
265: fprintf(stderr, "execute %08x\n", xt);
266: #endif
267: #ifndef NO_IP
268: ip=IP;
269: #endif
270: SUPER_END;
271: VM_JUMP(EXEC1(xt));
272:
273: perform ( a_addr -- ) gforth
274: ""@code{@@ execute}.""
275: /* and pfe */
276: #ifndef NO_IP
277: ip=IP;
278: #endif
279: SUPER_END;
280: VM_JUMP(EXEC1(*(Xt *)a_addr));
281: :
282: @ execute ;
283:
284: ;s ( R:w -- ) gforth semis
285: ""The primitive compiled by @code{EXIT}.""
286: #ifdef NO_IP
287: INST_TAIL;
288: goto *(void *)w;
289: #else
290: SET_IP((Xt *)w);
291: #endif
292:
293: unloop ( R:w1 R:w2 -- ) core
294: /* !! alias for 2rdrop */
295: :
296: r> rdrop rdrop >r ;
297:
298: lit-perform ( #a_addr -- ) new lit_perform
299: #ifndef NO_IP
300: ip=IP;
301: #endif
302: SUPER_END;
303: VM_JUMP(EXEC1(*(Xt *)a_addr));
304:
305: does-exec ( #a_cfa -- R:nest a_pfa ) new does_exec
306: #ifdef NO_IP
307: /* compiled to LIT CALL by compile_prim */
308: assert(0);
309: #else
310: a_pfa = PFA(a_cfa);
311: nest = (Cell)IP;
312: #ifdef DEBUG
313: {
314: CFA_TO_NAME(a_cfa);
315: fprintf(stderr,"%08lx: does %08lx %.*s\n",
316: (Cell)ip,(Cell)a_cfa,len,name);
317: }
318: #endif
319: SET_IP(DOES_CODE1(a_cfa));
320: #endif
321:
322: \+glocals
323:
324: branch-lp+!# ( #a_target #nlocals -- ) gforth branch_lp_plus_store_number
325: /* this will probably not be used */
326: lp += nlocals;
327: #ifdef NO_IP
328: INST_TAIL;
329: JUMP(a_target);
330: #else
331: SET_IP((Xt *)a_target);
332: #endif
333:
334: \+
335:
336: branch ( #a_target -- ) gforth
337: #ifdef NO_IP
338: INST_TAIL;
339: JUMP(a_target);
340: #else
341: SET_IP((Xt *)a_target);
342: #endif
343: :
344: r> @ >r ;
345:
346: \ condbranch(forthname,stackeffect,restline,code1,code2,forthcode)
347: \ this is non-syntactical: code must open a brace that is closed by the macro
348: define(condbranch,
349: $1 ( `#'a_target $2 ) $3
350: $4 #ifdef NO_IP
351: INST_TAIL;
352: #endif
353: $5 #ifdef NO_IP
354: JUMP(a_target);
355: #else
356: SET_IP((Xt *)a_target);
357: ifelse(condbranch_opt,`1',`INST_TAIL; NEXT_P2;',`/* condbranch_opt=0 */')
358: #endif
359: }
360: ifelse(condbranch_opt,`1',`SUPER_CONTINUE;',`/* condbranch_opt=0 */')
361: $6
362:
363: \+glocals
364:
365: $1-lp+!`#' ( `#'a_target `#'nlocals $2 ) $3_lp_plus_store_number
366: $4 #ifdef NO_IP
367: INST_TAIL;
368: #endif
369: $5 lp += nlocals;
370: #ifdef NO_IP
371: JUMP(a_target);
372: #else
373: SET_IP((Xt *)a_target);
374: ifelse(condbranch_opt,`1',`INST_TAIL; NEXT_P2;',`/* condbranch_opt=0 */')
375: #endif
376: }
377: ifelse(condbranch_opt,`1',`SUPER_CONTINUE;',`/* condbranch_opt=0 */')
378:
379: \+
380: )
381:
382: condbranch(?branch,f --,f83 question_branch,
383: ,if (f==0) {
384: ,:
385: 0= dup 0= \ !f f
386: r> tuck cell+ \ !f branchoffset f IP+
387: and -rot @ and or \ f&IP+|!f&branch
388: >r ;)
389:
390: \ we don't need an lp_plus_store version of the ?dup-stuff, because it
391: \ is only used in if's (yet)
392:
393: \+xconds
394:
395: ?dup-?branch ( #a_target f -- S:... ) new question_dupe_question_branch
396: ""The run-time procedure compiled by @code{?DUP-IF}.""
397: if (f==0) {
398: #ifdef NO_IP
399: INST_TAIL;
400: JUMP(a_target);
401: #else
402: SET_IP((Xt *)a_target);
403: #endif
404: } else {
405: sp--;
406: sp[0]=f;
407: }
408:
409: ?dup-0=-?branch ( #a_target f -- S:... ) new question_dupe_zero_equals_question_branch
410: ""The run-time procedure compiled by @code{?DUP-0=-IF}.""
411: if (f!=0) {
412: sp--;
413: sp[0]=f;
414: #ifdef NO_IP
415: JUMP(a_target);
416: #else
417: SET_IP((Xt *)a_target);
418: #endif
419: }
420:
421: \+
422: \fhas? skiploopprims 0= [IF]
423:
424: condbranch((next),R:n1 -- R:n2,cmFORTH paren_next,
425: n2=n1-1;
426: ,if (n1) {
427: ,:
428: r> r> dup 1- >r
429: IF @ >r ELSE cell+ >r THEN ;)
430:
431: condbranch((loop),R:nlimit R:n1 -- R:nlimit R:n2,gforth paren_loop,
432: n2=n1+1;
433: ,if (n2 != nlimit) {
434: ,:
435: r> r> 1+ r> 2dup =
436: IF >r 1- >r cell+ >r
437: ELSE >r >r @ >r THEN ;)
438:
439: condbranch((+loop),n R:nlimit R:n1 -- R:nlimit R:n2,gforth paren_plus_loop,
440: /* !! check this thoroughly */
441: /* sign bit manipulation and test: (x^y)<0 is equivalent to (x<0) != (y<0) */
442: /* dependent upon two's complement arithmetic */
443: Cell olddiff = n1-nlimit;
444: n2=n1+n;
445: ,if (((olddiff^(olddiff+n)) /* the limit is not crossed */
446: &(olddiff^n)) /* OR it is a wrap-around effect */
447: >=0) { /* & is used to avoid having two branches for gforth-native */
448: ,:
449: r> swap
450: r> r> 2dup - >r
451: 2 pick r@ + r@ xor 0< 0=
452: 3 pick r> xor 0< 0= or
453: IF >r + >r @ >r
454: ELSE >r >r drop cell+ >r THEN ;)
455:
456: \+xconds
457:
458: condbranch((-loop),u R:nlimit R:n1 -- R:nlimit R:n2,gforth paren_minus_loop,
459: UCell olddiff = n1-nlimit;
460: n2=n1-u;
461: ,if (olddiff>u) {
462: ,)
463:
464: condbranch((s+loop),n R:nlimit R:n1 -- R:nlimit R:n2,gforth paren_symmetric_plus_loop,
465: ""The run-time procedure compiled by S+LOOP. It loops until the index
466: crosses the boundary between limit and limit-sign(n). I.e. a symmetric
467: version of (+LOOP).""
468: /* !! check this thoroughly */
469: Cell diff = n1-nlimit;
470: Cell newdiff = diff+n;
471: if (n<0) {
472: diff = -diff;
473: newdiff = -newdiff;
474: }
475: n2=n1+n;
476: ,if (((~diff)|newdiff)<0) { /* use | to avoid two branches for gforth-native */
477: ,)
478:
479: \+
480:
481: (for) ( ncount -- R:nlimit R:ncount ) cmFORTH paren_for
482: /* or (for) = >r -- collides with unloop! */
483: nlimit=0;
484: :
485: r> swap 0 >r >r >r ;
486:
487: (do) ( nlimit nstart -- R:nlimit R:nstart ) gforth paren_do
488: :
489: r> swap rot >r >r >r ;
490:
491: (?do) ( #a_target nlimit nstart -- R:nlimit R:nstart ) gforth paren_question_do
492: #ifdef NO_IP
493: INST_TAIL;
494: #endif
495: if (nstart == nlimit) {
496: #ifdef NO_IP
497: JUMP(a_target);
498: #else
499: SET_IP((Xt *)a_target);
500: #endif
501: }
502: :
503: 2dup =
504: IF r> swap rot >r >r
505: @ >r
506: ELSE r> swap rot >r >r
507: cell+ >r
508: THEN ; \ --> CORE-EXT
509:
510: \+xconds
511:
512: (+do) ( #a_target nlimit nstart -- R:nlimit R:nstart ) gforth paren_plus_do
513: #ifdef NO_IP
514: INST_TAIL;
515: #endif
516: if (nstart >= nlimit) {
517: #ifdef NO_IP
518: JUMP(a_target);
519: #else
520: SET_IP((Xt *)a_target);
521: #endif
522: }
523: :
524: swap 2dup
525: r> swap >r swap >r
526: >=
527: IF
528: @
529: ELSE
530: cell+
531: THEN >r ;
532:
533: (u+do) ( #a_target ulimit ustart -- R:ulimit R:ustart ) gforth paren_u_plus_do
534: #ifdef NO_IP
535: INST_TAIL;
536: #endif
537: if (ustart >= ulimit) {
538: #ifdef NO_IP
539: JUMP(a_target);
540: #else
541: SET_IP((Xt *)a_target);
542: #endif
543: }
544: :
545: swap 2dup
546: r> swap >r swap >r
547: u>=
548: IF
549: @
550: ELSE
551: cell+
552: THEN >r ;
553:
554: (-do) ( #a_target nlimit nstart -- R:nlimit R:nstart ) gforth paren_minus_do
555: #ifdef NO_IP
556: INST_TAIL;
557: #endif
558: if (nstart <= nlimit) {
559: #ifdef NO_IP
560: JUMP(a_target);
561: #else
562: SET_IP((Xt *)a_target);
563: #endif
564: }
565: :
566: swap 2dup
567: r> swap >r swap >r
568: <=
569: IF
570: @
571: ELSE
572: cell+
573: THEN >r ;
574:
575: (u-do) ( #a_target ulimit ustart -- R:ulimit R:ustart ) gforth paren_u_minus_do
576: #ifdef NO_IP
577: INST_TAIL;
578: #endif
579: if (ustart <= ulimit) {
580: #ifdef NO_IP
581: JUMP(a_target);
582: #else
583: SET_IP((Xt *)a_target);
584: #endif
585: }
586: :
587: swap 2dup
588: r> swap >r swap >r
589: u<=
590: IF
591: @
592: ELSE
593: cell+
594: THEN >r ;
595:
596: \+
597:
598: \ don't make any assumptions where the return stack is!!
599: \ implement this in machine code if it should run quickly!
600:
601: i ( R:n -- R:n n ) core
602: :
603: \ rp@ cell+ @ ;
604: r> r> tuck >r >r ;
605:
606: i' ( R:w R:w2 -- R:w R:w2 w ) gforth i_tick
607: :
608: \ rp@ cell+ cell+ @ ;
609: r> r> r> dup itmp ! >r >r >r itmp @ ;
610: variable itmp
611:
612: j ( R:w R:w1 R:w2 -- w R:w R:w1 R:w2 ) core
613: :
614: \ rp@ cell+ cell+ cell+ @ ;
615: r> r> r> r> dup itmp ! >r >r >r >r itmp @ ;
616: [IFUNDEF] itmp variable itmp [THEN]
617:
618: k ( R:w R:w1 R:w2 R:w3 R:w4 -- w R:w R:w1 R:w2 R:w3 R:w4 ) gforth
619: :
620: \ rp@ [ 5 cells ] Literal + @ ;
621: r> r> r> r> r> r> dup itmp ! >r >r >r >r >r >r itmp @ ;
622: [IFUNDEF] itmp variable itmp [THEN]
623:
624: \f[THEN]
625:
626: \ digit is high-level: 0/0%
627:
628: \g strings
629:
630: move ( c_from c_to ucount -- ) core
631: ""Copy the contents of @i{ucount} aus at @i{c-from} to
632: @i{c-to}. @code{move} works correctly even if the two areas overlap.""
633: /* !! note that the standard specifies addr, not c-addr */
634: memmove(c_to,c_from,ucount);
635: /* make an Ifdef for bsd and others? */
636: :
637: >r 2dup u< IF r> cmove> ELSE r> cmove THEN ;
638:
639: cmove ( c_from c_to u -- ) string c_move
640: ""Copy the contents of @i{ucount} characters from data space at
641: @i{c-from} to @i{c-to}. The copy proceeds @code{char}-by-@code{char}
642: from low address to high address; i.e., for overlapping areas it is
643: safe if @i{c-to}=<@i{c-from}.""
644: cmove(c_from,c_to,u);
645: :
646: bounds ?DO dup c@ I c! 1+ LOOP drop ;
647:
648: cmove> ( c_from c_to u -- ) string c_move_up
649: ""Copy the contents of @i{ucount} characters from data space at
650: @i{c-from} to @i{c-to}. The copy proceeds @code{char}-by-@code{char}
651: from high address to low address; i.e., for overlapping areas it is
652: safe if @i{c-to}>=@i{c-from}.""
653: cmove_up(c_from,c_to,u);
654: :
655: dup 0= IF drop 2drop exit THEN
656: rot over + -rot bounds swap 1-
657: DO 1- dup c@ I c! -1 +LOOP drop ;
658:
659: fill ( c_addr u c -- ) core
660: ""Store @i{c} in @i{u} chars starting at @i{c-addr}.""
661: memset(c_addr,c,u);
662: :
663: -rot bounds
664: ?DO dup I c! LOOP drop ;
665:
666: compare ( c_addr1 u1 c_addr2 u2 -- n ) string
667: ""Compare two strings lexicographically. If they are equal, @i{n} is 0; if
668: the first string is smaller, @i{n} is -1; if the first string is larger, @i{n}
669: is 1. Currently this is based on the machine's character
670: comparison. In the future, this may change to consider the current
671: locale and its collation order.""
672: /* close ' to keep fontify happy */
673: n = compare(c_addr1, u1, c_addr2, u2);
674: :
675: rot 2dup swap - >r min swap -text dup
676: IF rdrop ELSE drop r> sgn THEN ;
677: : -text ( c_addr1 u c_addr2 -- n )
678: swap bounds
679: ?DO dup c@ I c@ = WHILE 1+ LOOP drop 0
680: ELSE c@ I c@ - unloop THEN sgn ;
681: : sgn ( n -- -1/0/1 )
682: dup 0= IF EXIT THEN 0< 2* 1+ ;
683:
684: \ -text is only used by replaced primitives now; move it elsewhere
685: \ -text ( c_addr1 u c_addr2 -- n ) new dash_text
686: \ n = memcmp(c_addr1, c_addr2, u);
687: \ if (n<0)
688: \ n = -1;
689: \ else if (n>0)
690: \ n = 1;
691: \ :
692: \ swap bounds
693: \ ?DO dup c@ I c@ = WHILE 1+ LOOP drop 0
694: \ ELSE c@ I c@ - unloop THEN sgn ;
695: \ : sgn ( n -- -1/0/1 )
696: \ dup 0= IF EXIT THEN 0< 2* 1+ ;
697:
698: toupper ( c1 -- c2 ) gforth
699: ""If @i{c1} is a lower-case character (in the current locale), @i{c2}
700: is the equivalent upper-case character. All other characters are unchanged.""
701: c2 = toupper(c1);
702: :
703: dup [char] a - [ char z char a - 1 + ] Literal u< bl and - ;
704:
705: capscompare ( c_addr1 u1 c_addr2 u2 -- n ) gforth
706: ""Compare two strings lexicographically. If they are equal, @i{n} is 0; if
707: the first string is smaller, @i{n} is -1; if the first string is larger, @i{n}
708: is 1. Currently this is based on the machine's character
709: comparison. In the future, this may change to consider the current
710: locale and its collation order.""
711: /* close ' to keep fontify happy */
712: n = capscompare(c_addr1, u1, c_addr2, u2);
713:
714: /string ( c_addr1 u1 n -- c_addr2 u2 ) string slash_string
715: ""Adjust the string specified by @i{c-addr1, u1} to remove @i{n}
716: characters from the start of the string.""
717: c_addr2 = c_addr1+n;
718: u2 = u1-n;
719: :
720: tuck - >r + r> dup 0< IF - 0 THEN ;
721:
722: \g arith
723:
724: lit ( #w -- w ) gforth
725: :
726: r> dup @ swap cell+ >r ;
727:
728: + ( n1 n2 -- n ) core plus
729: n = n1+n2;
730:
731: \ lit+ / lit_plus = lit +
732:
733: lit+ ( n1 #n2 -- n ) new lit_plus
734: #ifdef DEBUG
735: fprintf(stderr, "lit+ %08x\n", n2);
736: #endif
737: n=n1+n2;
738:
739: \ PFE-0.9.14 has it differently, but the next release will have it as follows
740: under+ ( n1 n2 n3 -- n n2 ) gforth under_plus
741: ""add @i{n3} to @i{n1} (giving @i{n})""
742: n = n1+n3;
743: :
744: rot + swap ;
745:
746: - ( n1 n2 -- n ) core minus
747: n = n1-n2;
748: :
749: negate + ;
750:
751: negate ( n1 -- n2 ) core
752: /* use minus as alias */
753: n2 = -n1;
754: :
755: invert 1+ ;
756:
757: 1+ ( n1 -- n2 ) core one_plus
758: n2 = n1+1;
759: :
760: 1 + ;
761:
762: 1- ( n1 -- n2 ) core one_minus
763: n2 = n1-1;
764: :
765: 1 - ;
766:
767: max ( n1 n2 -- n ) core
768: if (n1<n2)
769: n = n2;
770: else
771: n = n1;
772: :
773: 2dup < IF swap THEN drop ;
774:
775: min ( n1 n2 -- n ) core
776: if (n1<n2)
777: n = n1;
778: else
779: n = n2;
780: :
781: 2dup > IF swap THEN drop ;
782:
783: abs ( n -- u ) core
784: if (n<0)
785: u = -n;
786: else
787: u = n;
788: :
789: dup 0< IF negate THEN ;
790:
791: * ( n1 n2 -- n ) core star
792: n = n1*n2;
793: :
794: um* drop ;
795:
796: / ( n1 n2 -- n ) core slash
797: n = n1/n2;
798: if (CHECK_DIVISION_SW && n2 == 0)
799: throw(BALL_DIVZERO);
800: if (CHECK_DIVISION_SW && n2 == -1 && n1 == CELL_MIN)
801: throw(BALL_RESULTRANGE);
802: if (FLOORED_DIV && ((n1^n2) < 0) && (n1%n2 != 0))
803: n--;
804: :
805: /mod nip ;
806:
807: mod ( n1 n2 -- n ) core
808: n = n1%n2;
809: if (CHECK_DIVISION_SW && n2 == 0)
810: throw(BALL_DIVZERO);
811: if (CHECK_DIVISION_SW && n2 == -1 && n1 == CELL_MIN)
812: throw(BALL_RESULTRANGE);
813: if(FLOORED_DIV && ((n1^n2) < 0) && n!=0) n += n2;
814: :
815: /mod drop ;
816:
817: /mod ( n1 n2 -- n3 n4 ) core slash_mod
818: n4 = n1/n2;
819: n3 = n1%n2; /* !! is this correct? look into C standard! */
820: if (CHECK_DIVISION_SW && n2 == 0)
821: throw(BALL_DIVZERO);
822: if (CHECK_DIVISION_SW && n2 == -1 && n1 == CELL_MIN)
823: throw(BALL_RESULTRANGE);
824: if (FLOORED_DIV && ((n1^n2) < 0) && n3!=0) {
825: n4--;
826: n3+=n2;
827: }
828: :
829: >r s>d r> fm/mod ;
830:
831: */mod ( n1 n2 n3 -- n4 n5 ) core star_slash_mod
832: ""n1*n2=n3*n5+n4, with the intermediate result (n1*n2) being double.""
833: #ifdef BUGGY_LL_MUL
834: DCell d = mmul(n1,n2);
835: #else
836: DCell d = (DCell)n1 * (DCell)n2;
837: #endif
838: #ifdef ASM_SM_SLASH_REM
839: ASM_SM_SLASH_REM(DLO(d), DHI(d), n3, n4, n5);
840: if (FLOORED_DIV && ((DHI(d)^n3)<0) && n4!=0) {
841: if (CHECK_DIVISION && n5 == CELL_MIN)
842: throw(BALL_RESULTRANGE);
843: n5--;
844: n4+=n3;
845: }
846: #else
847: DCell r = FLOORED_DIV ? fmdiv(d,n3) : smdiv(d,n3);
848: n4=DHI(r);
849: n5=DLO(r);
850: #endif
851: :
852: >r m* r> fm/mod ;
853:
854: */ ( n1 n2 n3 -- n4 ) core star_slash
855: ""n4=(n1*n2)/n3, with the intermediate result being double.""
856: #ifdef BUGGY_LL_MUL
857: DCell d = mmul(n1,n2);
858: #else
859: DCell d = (DCell)n1 * (DCell)n2;
860: #endif
861: #ifdef ASM_SM_SLASH_REM
862: Cell remainder;
863: ASM_SM_SLASH_REM(DLO(d), DHI(d), n3, remainder, n4);
864: if (FLOORED_DIV && ((DHI(d)^n3)<0) && remainder!=0) {
865: if (CHECK_DIVISION && n4 == CELL_MIN)
866: throw(BALL_RESULTRANGE);
867: n4--;
868: }
869: #else
870: DCell r = FLOORED_DIV ? fmdiv(d,n3) : smdiv(d,n3);
871: n4=DLO(r);
872: #endif
873: :
874: */mod nip ;
875:
876: 2* ( n1 -- n2 ) core two_star
877: ""Shift left by 1; also works on unsigned numbers""
878: n2 = 2*n1;
879: :
880: dup + ;
881:
882: 2/ ( n1 -- n2 ) core two_slash
883: ""Arithmetic shift right by 1. For signed numbers this is a floored
884: division by 2 (note that @code{/} not necessarily floors).""
885: n2 = n1>>1;
886: :
887: dup MINI and IF 1 ELSE 0 THEN
888: [ bits/char cell * 1- ] literal
889: 0 DO 2* swap dup 2* >r MINI and
890: IF 1 ELSE 0 THEN or r> swap
891: LOOP nip ;
892:
893: fm/mod ( d1 n1 -- n2 n3 ) core f_m_slash_mod
894: ""Floored division: @i{d1} = @i{n3}*@i{n1}+@i{n2}, @i{n1}>@i{n2}>=0 or 0>=@i{n2}>@i{n1}.""
895: #ifdef ASM_SM_SLASH_REM
896: ASM_SM_SLASH_REM(DLO(d1), DHI(d1), n1, n2, n3);
897: if (((DHI(d1)^n1)<0) && n2!=0) {
898: if (CHECK_DIVISION && n3 == CELL_MIN)
899: throw(BALL_RESULTRANGE);
900: n3--;
901: n2+=n1;
902: }
903: #else /* !defined(ASM_SM_SLASH_REM) */
904: DCell r = fmdiv(d1,n1);
905: n2=DHI(r);
906: n3=DLO(r);
907: #endif /* !defined(ASM_SM_SLASH_REM) */
908: :
909: dup >r dup 0< IF negate >r dnegate r> THEN
910: over 0< IF tuck + swap THEN
911: um/mod
912: r> 0< IF swap negate swap THEN ;
913:
914: sm/rem ( d1 n1 -- n2 n3 ) core s_m_slash_rem
915: ""Symmetric division: @i{d1} = @i{n3}*@i{n1}+@i{n2}, sign(@i{n2})=sign(@i{d1}) or 0.""
916: #ifdef ASM_SM_SLASH_REM
917: ASM_SM_SLASH_REM(DLO(d1), DHI(d1), n1, n2, n3);
918: #else /* !defined(ASM_SM_SLASH_REM) */
919: DCell r = smdiv(d1,n1);
920: n2=DHI(r);
921: n3=DLO(r);
922: #endif /* !defined(ASM_SM_SLASH_REM) */
923: :
924: over >r dup >r abs -rot
925: dabs rot um/mod
926: r> r@ xor 0< IF negate THEN
927: r> 0< IF swap negate swap THEN ;
928:
929: m* ( n1 n2 -- d ) core m_star
930: #ifdef BUGGY_LL_MUL
931: d = mmul(n1,n2);
932: #else
933: d = (DCell)n1 * (DCell)n2;
934: #endif
935: :
936: 2dup 0< and >r
937: 2dup swap 0< and >r
938: um* r> - r> - ;
939:
940: um* ( u1 u2 -- ud ) core u_m_star
941: /* use u* as alias */
942: #ifdef BUGGY_LL_MUL
943: ud = ummul(u1,u2);
944: #else
945: ud = (UDCell)u1 * (UDCell)u2;
946: #endif
947: :
948: 0 -rot dup [ 8 cells ] literal -
949: DO
950: dup 0< I' and d2*+ drop
951: LOOP ;
952: : d2*+ ( ud n -- ud+n c )
953: over MINI
954: and >r >r 2dup d+ swap r> + swap r> ;
955:
956: um/mod ( ud u1 -- u2 u3 ) core u_m_slash_mod
957: ""ud=u3*u1+u2, u1>u2>=0""
958: #ifdef ASM_UM_SLASH_MOD
959: ASM_UM_SLASH_MOD(DLO(ud), DHI(ud), u1, u2, u3);
960: #else /* !defined(ASM_UM_SLASH_MOD) */
961: UDCell r = umdiv(ud,u1);
962: u2=DHI(r);
963: u3=DLO(r);
964: #endif /* !defined(ASM_UM_SLASH_MOD) */
965: :
966: 0 swap [ 8 cells 1 + ] literal 0
967: ?DO /modstep
968: LOOP drop swap 1 rshift or swap ;
969: : /modstep ( ud c R: u -- ud-?u c R: u )
970: >r over r@ u< 0= or IF r@ - 1 ELSE 0 THEN d2*+ r> ;
971: : d2*+ ( ud n -- ud+n c )
972: over MINI
973: and >r >r 2dup d+ swap r> + swap r> ;
974:
975: m+ ( d1 n -- d2 ) double m_plus
976: #ifdef BUGGY_LL_ADD
977: DLO_IS(d2, DLO(d1)+n);
978: DHI_IS(d2, DHI(d1) - (n<0) + (DLO(d2)<DLO(d1)));
979: #else
980: d2 = d1+n;
981: #endif
982: :
983: s>d d+ ;
984:
985: d+ ( d1 d2 -- d ) double d_plus
986: #ifdef BUGGY_LL_ADD
987: DLO_IS(d, DLO(d1) + DLO(d2));
988: DHI_IS(d, DHI(d1) + DHI(d2) + (d.lo<DLO(d1)));
989: #else
990: d = d1+d2;
991: #endif
992: :
993: rot + >r tuck + swap over u> r> swap - ;
994:
995: d- ( d1 d2 -- d ) double d_minus
996: #ifdef BUGGY_LL_ADD
997: DLO_IS(d, DLO(d1) - DLO(d2));
998: DHI_IS(d, DHI(d1)-DHI(d2)-(DLO(d1)<DLO(d2)));
999: #else
1000: d = d1-d2;
1001: #endif
1002: :
1003: dnegate d+ ;
1004:
1005: dnegate ( d1 -- d2 ) double d_negate
1006: /* use dminus as alias */
1007: #ifdef BUGGY_LL_ADD
1008: d2 = dnegate(d1);
1009: #else
1010: d2 = -d1;
1011: #endif
1012: :
1013: invert swap negate tuck 0= - ;
1014:
1015: d2* ( d1 -- d2 ) double d_two_star
1016: ""Shift left by 1; also works on unsigned numbers""
1017: d2 = DLSHIFT(d1,1);
1018: :
1019: 2dup d+ ;
1020:
1021: d2/ ( d1 -- d2 ) double d_two_slash
1022: ""Arithmetic shift right by 1. For signed numbers this is a floored
1023: division by 2.""
1024: #ifdef BUGGY_LL_SHIFT
1025: DHI_IS(d2, DHI(d1)>>1);
1026: DLO_IS(d2, (DLO(d1)>>1) | (DHI(d1)<<(CELL_BITS-1)));
1027: #else
1028: d2 = d1>>1;
1029: #endif
1030: :
1031: dup 1 and >r 2/ swap 2/ [ 1 8 cells 1- lshift 1- ] Literal and
1032: r> IF [ 1 8 cells 1- lshift ] Literal + THEN swap ;
1033:
1034: and ( w1 w2 -- w ) core
1035: w = w1&w2;
1036:
1037: or ( w1 w2 -- w ) core
1038: w = w1|w2;
1039: :
1040: invert swap invert and invert ;
1041:
1042: xor ( w1 w2 -- w ) core x_or
1043: w = w1^w2;
1044:
1045: invert ( w1 -- w2 ) core
1046: w2 = ~w1;
1047: :
1048: MAXU xor ;
1049:
1050: rshift ( u1 n -- u2 ) core r_shift
1051: ""Logical shift right by @i{n} bits.""
1052: #ifdef BROKEN_SHIFT
1053: u2 = rshift(u1, n);
1054: #else
1055: u2 = u1 >> n;
1056: #endif
1057: :
1058: 0 ?DO 2/ MAXI and LOOP ;
1059:
1060: lshift ( u1 n -- u2 ) core l_shift
1061: #ifdef BROKEN_SHIFT
1062: u2 = lshift(u1, n);
1063: #else
1064: u2 = u1 << n;
1065: #endif
1066: :
1067: 0 ?DO 2* LOOP ;
1068:
1069: \g compare
1070:
1071: \ comparisons(prefix, args, prefix, arg1, arg2, wordsets...)
1072: define(comparisons,
1073: $1= ( $2 -- f ) $6 $3equals
1074: f = FLAG($4==$5);
1075: :
1076: [ char $1x char 0 = [IF]
1077: ] IF false ELSE true THEN [
1078: [ELSE]
1079: ] xor 0= [
1080: [THEN] ] ;
1081:
1082: $1<> ( $2 -- f ) $7 $3not_equals
1083: f = FLAG($4!=$5);
1084: :
1085: [ char $1x char 0 = [IF]
1086: ] IF true ELSE false THEN [
1087: [ELSE]
1088: ] xor 0<> [
1089: [THEN] ] ;
1090:
1091: $1< ( $2 -- f ) $8 $3less_than
1092: f = FLAG($4<$5);
1093: :
1094: [ char $1x char 0 = [IF]
1095: ] MINI and 0<> [
1096: [ELSE] char $1x char u = [IF]
1097: ] 2dup xor 0< IF nip ELSE - THEN 0< [
1098: [ELSE]
1099: ] MINI xor >r MINI xor r> u< [
1100: [THEN]
1101: [THEN] ] ;
1102:
1103: $1> ( $2 -- f ) $9 $3greater_than
1104: f = FLAG($4>$5);
1105: :
1106: [ char $1x char 0 = [IF] ] negate [ [ELSE] ] swap [ [THEN] ]
1107: $1< ;
1108:
1109: $1<= ( $2 -- f ) gforth $3less_or_equal
1110: f = FLAG($4<=$5);
1111: :
1112: $1> 0= ;
1113:
1114: $1>= ( $2 -- f ) gforth $3greater_or_equal
1115: f = FLAG($4>=$5);
1116: :
1117: [ char $1x char 0 = [IF] ] negate [ [ELSE] ] swap [ [THEN] ]
1118: $1<= ;
1119:
1120: )
1121:
1122: comparisons(0, n, zero_, n, 0, core, core-ext, core, core-ext)
1123: comparisons(, n1 n2, , n1, n2, core, core-ext, core, core)
1124: comparisons(u, u1 u2, u_, u1, u2, gforth, gforth, core, core-ext)
1125:
1126: \ dcomparisons(prefix, args, prefix, arg1, arg2, wordsets...)
1127: define(dcomparisons,
1128: $1= ( $2 -- f ) $6 $3equals
1129: #ifdef BUGGY_LL_CMP
1130: f = FLAG($4.lo==$5.lo && $4.hi==$5.hi);
1131: #else
1132: f = FLAG($4==$5);
1133: #endif
1134:
1135: $1<> ( $2 -- f ) $7 $3not_equals
1136: #ifdef BUGGY_LL_CMP
1137: f = FLAG($4.lo!=$5.lo || $4.hi!=$5.hi);
1138: #else
1139: f = FLAG($4!=$5);
1140: #endif
1141:
1142: $1< ( $2 -- f ) $8 $3less_than
1143: #ifdef BUGGY_LL_CMP
1144: f = FLAG($4.hi==$5.hi ? $4.lo<$5.lo : $4.hi<$5.hi);
1145: #else
1146: f = FLAG($4<$5);
1147: #endif
1148:
1149: $1> ( $2 -- f ) $9 $3greater_than
1150: #ifdef BUGGY_LL_CMP
1151: f = FLAG($4.hi==$5.hi ? $4.lo>$5.lo : $4.hi>$5.hi);
1152: #else
1153: f = FLAG($4>$5);
1154: #endif
1155:
1156: $1<= ( $2 -- f ) gforth $3less_or_equal
1157: #ifdef BUGGY_LL_CMP
1158: f = FLAG($4.hi==$5.hi ? $4.lo<=$5.lo : $4.hi<=$5.hi);
1159: #else
1160: f = FLAG($4<=$5);
1161: #endif
1162:
1163: $1>= ( $2 -- f ) gforth $3greater_or_equal
1164: #ifdef BUGGY_LL_CMP
1165: f = FLAG($4.hi==$5.hi ? $4.lo>=$5.lo : $4.hi>=$5.hi);
1166: #else
1167: f = FLAG($4>=$5);
1168: #endif
1169:
1170: )
1171:
1172: \+dcomps
1173:
1174: dcomparisons(d, d1 d2, d_, d1, d2, double, gforth, double, gforth)
1175: dcomparisons(d0, d, d_zero_, d, DZERO, double, gforth, double, gforth)
1176: dcomparisons(du, ud1 ud2, d_u_, ud1, ud2, gforth, gforth, double-ext, gforth)
1177:
1178: \+
1179:
1180: within ( u1 u2 u3 -- f ) core-ext
1181: ""u2=<u1<u3 or: u3=<u2 and u1 is not in [u3,u2). This works for
1182: unsigned and signed numbers (but not a mixture). Another way to think
1183: about this word is to consider the numbers as a circle (wrapping
1184: around from @code{max-u} to 0 for unsigned, and from @code{max-n} to
1185: min-n for signed numbers); now consider the range from u2 towards
1186: increasing numbers up to and excluding u3 (giving an empty range if
1187: u2=u3); if u1 is in this range, @code{within} returns true.""
1188: f = FLAG(u1-u2 < u3-u2);
1189: :
1190: over - >r - r> u< ;
1191:
1192: \g stack
1193:
1194: useraddr ( #u -- a_addr ) new
1195: a_addr = (Cell *)(up+u);
1196:
1197: up! ( a_addr -- ) gforth up_store
1198: gforth_UP=up=(Address)a_addr;
1199: :
1200: up ! ;
1201: Variable UP
1202:
1203: sp@ ( S:... -- a_addr ) gforth sp_fetch
1204: a_addr = sp;
1205:
1206: sp! ( a_addr -- S:... ) gforth sp_store
1207: sp = a_addr;
1208:
1209: rp@ ( -- a_addr ) gforth rp_fetch
1210: a_addr = rp;
1211:
1212: rp! ( a_addr -- ) gforth rp_store
1213: rp = a_addr;
1214:
1215: \+floating
1216:
1217: fp@ ( f:... -- f_addr ) gforth fp_fetch
1218: f_addr = fp;
1219:
1220: fp! ( f_addr -- f:... ) gforth fp_store
1221: fp = f_addr;
1222:
1223: \+
1224:
1225: >r ( w -- R:w ) core to_r
1226: :
1227: (>r) ;
1228: : (>r) rp@ cell+ @ rp@ ! rp@ cell+ ! ;
1229:
1230: r> ( R:w -- w ) core r_from
1231: :
1232: rp@ cell+ @ rp@ @ rp@ cell+ ! (rdrop) rp@ ! ;
1233: Create (rdrop) ' ;s A,
1234:
1235: rdrop ( R:w -- ) gforth
1236: :
1237: r> r> drop >r ;
1238:
1239: 2>r ( d -- R:d ) core-ext two_to_r
1240: :
1241: swap r> swap >r swap >r >r ;
1242:
1243: 2r> ( R:d -- d ) core-ext two_r_from
1244: :
1245: r> r> swap r> swap >r swap ;
1246:
1247: 2r@ ( R:d -- R:d d ) core-ext two_r_fetch
1248: :
1249: i' j ;
1250:
1251: 2rdrop ( R:d -- ) gforth two_r_drop
1252: :
1253: r> r> drop r> drop >r ;
1254:
1255: over ( w1 w2 -- w1 w2 w1 ) core
1256: :
1257: sp@ cell+ @ ;
1258:
1259: drop ( w -- ) core
1260: :
1261: IF THEN ;
1262:
1263: swap ( w1 w2 -- w2 w1 ) core
1264: :
1265: >r (swap) ! r> (swap) @ ;
1266: Variable (swap)
1267:
1268: dup ( w -- w w ) core dupe
1269: :
1270: sp@ @ ;
1271:
1272: rot ( w1 w2 w3 -- w2 w3 w1 ) core rote
1273: :
1274: [ defined? (swap) [IF] ]
1275: (swap) ! (rot) ! >r (rot) @ (swap) @ r> ;
1276: Variable (rot)
1277: [ELSE] ]
1278: >r swap r> swap ;
1279: [THEN]
1280:
1281: -rot ( w1 w2 w3 -- w3 w1 w2 ) gforth not_rote
1282: :
1283: rot rot ;
1284:
1285: nip ( w1 w2 -- w2 ) core-ext
1286: :
1287: swap drop ;
1288:
1289: tuck ( w1 w2 -- w2 w1 w2 ) core-ext
1290: :
1291: swap over ;
1292:
1293: ?dup ( w -- S:... w ) core question_dupe
1294: ""Actually the stack effect is: @code{( w -- 0 | w w )}. It performs a
1295: @code{dup} if w is nonzero.""
1296: if (w!=0) {
1297: *--sp = w;
1298: }
1299: :
1300: dup IF dup THEN ;
1301:
1302: pick ( S:... u -- S:... w ) core-ext
1303: ""Actually the stack effect is @code{ x0 ... xu u -- x0 ... xu x0 }.""
1304: w = sp[u];
1305: :
1306: 1+ cells sp@ + @ ;
1307:
1308: 2drop ( w1 w2 -- ) core two_drop
1309: :
1310: drop drop ;
1311:
1312: 2dup ( w1 w2 -- w1 w2 w1 w2 ) core two_dupe
1313: :
1314: over over ;
1315:
1316: 2over ( w1 w2 w3 w4 -- w1 w2 w3 w4 w1 w2 ) core two_over
1317: :
1318: 3 pick 3 pick ;
1319:
1320: 2swap ( w1 w2 w3 w4 -- w3 w4 w1 w2 ) core two_swap
1321: :
1322: rot >r rot r> ;
1323:
1324: 2rot ( w1 w2 w3 w4 w5 w6 -- w3 w4 w5 w6 w1 w2 ) double-ext two_rote
1325: :
1326: >r >r 2swap r> r> 2swap ;
1327:
1328: 2nip ( w1 w2 w3 w4 -- w3 w4 ) gforth two_nip
1329: :
1330: 2swap 2drop ;
1331:
1332: 2tuck ( w1 w2 w3 w4 -- w3 w4 w1 w2 w3 w4 ) gforth two_tuck
1333: :
1334: 2swap 2over ;
1335:
1336: \ toggle is high-level: 0.11/0.42%
1337:
1338: \g memory
1339:
1340: @ ( a_addr -- w ) core fetch
1341: ""@i{w} is the cell stored at @i{a_addr}.""
1342: w = *a_addr;
1343:
1344: \ lit@ / lit_fetch = lit @
1345:
1346: lit@ ( #a_addr -- w ) new lit_fetch
1347: w = *a_addr;
1348:
1349: ! ( w a_addr -- ) core store
1350: ""Store @i{w} into the cell at @i{a-addr}.""
1351: *a_addr = w;
1352:
1353: +! ( n a_addr -- ) core plus_store
1354: ""Add @i{n} to the cell at @i{a-addr}.""
1355: *a_addr += n;
1356: :
1357: tuck @ + swap ! ;
1358:
1359: c@ ( c_addr -- c ) core c_fetch
1360: ""@i{c} is the char stored at @i{c_addr}.""
1361: c = *c_addr;
1362: :
1363: [ bigendian [IF] ]
1364: [ cell>bit 4 = [IF] ]
1365: dup [ 0 cell - ] Literal and @ swap 1 and
1366: IF $FF and ELSE 8>> THEN ;
1367: [ [ELSE] ]
1368: dup [ cell 1- ] literal and
1369: tuck - @ swap [ cell 1- ] literal xor
1370: 0 ?DO 8>> LOOP $FF and
1371: [ [THEN] ]
1372: [ [ELSE] ]
1373: [ cell>bit 4 = [IF] ]
1374: dup [ 0 cell - ] Literal and @ swap 1 and
1375: IF 8>> ELSE $FF and THEN
1376: [ [ELSE] ]
1377: dup [ cell 1- ] literal and
1378: tuck - @ swap
1379: 0 ?DO 8>> LOOP 255 and
1380: [ [THEN] ]
1381: [ [THEN] ]
1382: ;
1383: : 8>> 2/ 2/ 2/ 2/ 2/ 2/ 2/ 2/ ;
1384:
1385: c! ( c c_addr -- ) core c_store
1386: ""Store @i{c} into the char at @i{c-addr}.""
1387: *c_addr = c;
1388: :
1389: [ bigendian [IF] ]
1390: [ cell>bit 4 = [IF] ]
1391: tuck 1 and IF $FF and ELSE 8<< THEN >r
1392: dup -2 and @ over 1 and cells masks + @ and
1393: r> or swap -2 and ! ;
1394: Create masks $00FF , $FF00 ,
1395: [ELSE] ]
1396: dup [ cell 1- ] literal and dup
1397: [ cell 1- ] literal xor >r
1398: - dup @ $FF r@ 0 ?DO 8<< LOOP invert and
1399: rot $FF and r> 0 ?DO 8<< LOOP or swap ! ;
1400: [THEN]
1401: [ELSE] ]
1402: [ cell>bit 4 = [IF] ]
1403: tuck 1 and IF 8<< ELSE $FF and THEN >r
1404: dup -2 and @ over 1 and cells masks + @ and
1405: r> or swap -2 and ! ;
1406: Create masks $FF00 , $00FF ,
1407: [ELSE] ]
1408: dup [ cell 1- ] literal and dup >r
1409: - dup @ $FF r@ 0 ?DO 8<< LOOP invert and
1410: rot $FF and r> 0 ?DO 8<< LOOP or swap ! ;
1411: [THEN]
1412: [THEN]
1413: : 8<< 2* 2* 2* 2* 2* 2* 2* 2* ;
1414:
1415: 2! ( w1 w2 a_addr -- ) core two_store
1416: ""Store @i{w2} into the cell at @i{c-addr} and @i{w1} into the next cell.""
1417: a_addr[0] = w2;
1418: a_addr[1] = w1;
1419: :
1420: tuck ! cell+ ! ;
1421:
1422: 2@ ( a_addr -- w1 w2 ) core two_fetch
1423: ""@i{w2} is the content of the cell stored at @i{a-addr}, @i{w1} is
1424: the content of the next cell.""
1425: w2 = a_addr[0];
1426: w1 = a_addr[1];
1427: :
1428: dup cell+ @ swap @ ;
1429:
1430: cell+ ( a_addr1 -- a_addr2 ) core cell_plus
1431: ""@code{1 cells +}""
1432: a_addr2 = a_addr1+1;
1433: :
1434: cell + ;
1435:
1436: cells ( n1 -- n2 ) core
1437: "" @i{n2} is the number of address units of @i{n1} cells.""
1438: n2 = n1 * sizeof(Cell);
1439: :
1440: [ cell
1441: 2/ dup [IF] ] 2* [ [THEN]
1442: 2/ dup [IF] ] 2* [ [THEN]
1443: 2/ dup [IF] ] 2* [ [THEN]
1444: 2/ dup [IF] ] 2* [ [THEN]
1445: drop ] ;
1446:
1447: char+ ( c_addr1 -- c_addr2 ) core char_plus
1448: ""@code{1 chars +}.""
1449: c_addr2 = c_addr1 + 1;
1450: :
1451: 1+ ;
1452:
1453: (chars) ( n1 -- n2 ) gforth paren_chars
1454: n2 = n1 * sizeof(Char);
1455: :
1456: ;
1457:
1458: count ( c_addr1 -- c_addr2 u ) core
1459: ""@i{c-addr2} is the first character and @i{u} the length of the
1460: counted string at @i{c-addr1}.""
1461: u = *c_addr1;
1462: c_addr2 = c_addr1+1;
1463: :
1464: dup 1+ swap c@ ;
1465:
1466: \g compiler
1467:
1468: \+f83headerstring
1469:
1470: (f83find) ( c_addr u f83name1 -- f83name2 ) new paren_f83find
1471: for (; f83name1 != NULL; f83name1 = (struct F83Name *)(f83name1->next))
1472: if ((UCell)F83NAME_COUNT(f83name1)==u &&
1473: memcasecmp(c_addr, f83name1->name, u)== 0 /* or inline? */)
1474: break;
1475: f83name2=f83name1;
1476: #ifdef DEBUG
1477: fprintf(stderr, "F83find ");
1478: fwrite(c_addr, u, 1, stderr);
1479: fprintf(stderr, " found %08x\n", f83name2);
1480: #endif
1481: :
1482: BEGIN dup WHILE (find-samelen) dup WHILE
1483: >r 2dup r@ cell+ char+ capscomp 0=
1484: IF 2drop r> EXIT THEN
1485: r> @
1486: REPEAT THEN nip nip ;
1487: : (find-samelen) ( u f83name1 -- u f83name2/0 )
1488: BEGIN 2dup cell+ c@ $1F and <> WHILE @ dup 0= UNTIL THEN ;
1489: : capscomp ( c_addr1 u c_addr2 -- n )
1490: swap bounds
1491: ?DO dup c@ I c@ <>
1492: IF dup c@ toupper I c@ toupper =
1493: ELSE true THEN WHILE 1+ LOOP drop 0
1494: ELSE c@ toupper I c@ toupper - unloop THEN sgn ;
1495: : sgn ( n -- -1/0/1 )
1496: dup 0= IF EXIT THEN 0< 2* 1+ ;
1497:
1498: \-
1499:
1500: (listlfind) ( c_addr u longname1 -- longname2 ) new paren_listlfind
1501: longname2=listlfind(c_addr, u, longname1);
1502: :
1503: BEGIN dup WHILE (findl-samelen) dup WHILE
1504: >r 2dup r@ cell+ cell+ capscomp 0=
1505: IF 2drop r> EXIT THEN
1506: r> @
1507: REPEAT THEN nip nip ;
1508: : (findl-samelen) ( u longname1 -- u longname2/0 )
1509: BEGIN 2dup cell+ @ lcount-mask and <> WHILE @ dup 0= UNTIL THEN ;
1510: : capscomp ( c_addr1 u c_addr2 -- n )
1511: swap bounds
1512: ?DO dup c@ I c@ <>
1513: IF dup c@ toupper I c@ toupper =
1514: ELSE true THEN WHILE 1+ LOOP drop 0
1515: ELSE c@ toupper I c@ toupper - unloop THEN sgn ;
1516: : sgn ( n -- -1/0/1 )
1517: dup 0= IF EXIT THEN 0< 2* 1+ ;
1518:
1519: \+hash
1520:
1521: (hashlfind) ( c_addr u a_addr -- longname2 ) new paren_hashlfind
1522: longname2 = hashlfind(c_addr, u, a_addr);
1523: :
1524: BEGIN dup WHILE
1525: 2@ >r >r dup r@ cell+ @ lcount-mask and =
1526: IF 2dup r@ cell+ cell+ capscomp 0=
1527: IF 2drop r> rdrop EXIT THEN THEN
1528: rdrop r>
1529: REPEAT nip nip ;
1530:
1531: (tablelfind) ( c_addr u a_addr -- longname2 ) new paren_tablelfind
1532: ""A case-sensitive variant of @code{(hashfind)}""
1533: longname2 = tablelfind(c_addr, u, a_addr);
1534: :
1535: BEGIN dup WHILE
1536: 2@ >r >r dup r@ cell+ @ lcount-mask and =
1537: IF 2dup r@ cell+ cell+ -text 0=
1538: IF 2drop r> rdrop EXIT THEN THEN
1539: rdrop r>
1540: REPEAT nip nip ;
1541: : -text ( c_addr1 u c_addr2 -- n )
1542: swap bounds
1543: ?DO dup c@ I c@ = WHILE 1+ LOOP drop 0
1544: ELSE c@ I c@ - unloop THEN sgn ;
1545: : sgn ( n -- -1/0/1 )
1546: dup 0= IF EXIT THEN 0< 2* 1+ ;
1547:
1548: (hashkey1) ( c_addr u ubits -- ukey ) gforth paren_hashkey1
1549: ""ukey is the hash key for the string c_addr u fitting in ubits bits""
1550: ukey = hashkey1(c_addr, u, ubits);
1551: :
1552: dup rot-values + c@ over 1 swap lshift 1- >r
1553: tuck - 2swap r> 0 2swap bounds
1554: ?DO dup 4 pick lshift swap 3 pick rshift or
1555: I c@ toupper xor
1556: over and LOOP
1557: nip nip nip ;
1558: Create rot-values
1559: 5 c, 0 c, 1 c, 2 c, 3 c, 4 c, 5 c, 5 c, 5 c, 5 c,
1560: 3 c, 5 c, 5 c, 5 c, 5 c, 7 c, 5 c, 5 c, 5 c, 5 c,
1561: 7 c, 5 c, 5 c, 5 c, 5 c, 6 c, 5 c, 5 c, 5 c, 5 c,
1562: 7 c, 5 c, 5 c,
1563:
1564: \+
1565:
1566: \+
1567:
1568: (parse-white) ( c_addr1 u1 -- c_addr2 u2 ) gforth paren_parse_white
1569: struct Cellpair r=parse_white(c_addr1, u1);
1570: c_addr2 = (Char *)(r.n1);
1571: u2 = r.n2;
1572: :
1573: BEGIN dup WHILE over c@ bl <= WHILE 1 /string
1574: REPEAT THEN 2dup
1575: BEGIN dup WHILE over c@ bl > WHILE 1 /string
1576: REPEAT THEN nip - ;
1577:
1578: aligned ( c_addr -- a_addr ) core
1579: "" @i{a-addr} is the first aligned address greater than or equal to @i{c-addr}.""
1580: a_addr = (Cell *)((((Cell)c_addr)+(sizeof(Cell)-1))&(-sizeof(Cell)));
1581: :
1582: [ cell 1- ] Literal + [ -1 cells ] Literal and ;
1583:
1584: faligned ( c_addr -- f_addr ) float f_aligned
1585: "" @i{f-addr} is the first float-aligned address greater than or equal to @i{c-addr}.""
1586: f_addr = (Float *)((((Cell)c_addr)+(sizeof(Float)-1))&(-sizeof(Float)));
1587: :
1588: [ 1 floats 1- ] Literal + [ -1 floats ] Literal and ;
1589:
1590: \ threading stuff is currently only interesting if we have a compiler
1591: \fhas? standardthreading has? compiler and [IF]
1592: threading-method ( -- n ) gforth threading_method
1593: ""0 if the engine is direct threaded. Note that this may change during
1594: the lifetime of an image.""
1595: #if defined(DOUBLY_INDIRECT)
1596: n=2;
1597: #else
1598: # if defined(DIRECT_THREADED)
1599: n=0;
1600: # else
1601: n=1;
1602: # endif
1603: #endif
1604: :
1605: 1 ;
1606:
1607: \f[THEN]
1608:
1609: \g hostos
1610:
1611: key-file ( wfileid -- c ) gforth paren_key_file
1612: ""Read one character @i{c} from @i{wfileid}. This word disables
1613: buffering for @i{wfileid}. If you want to read characters from a
1614: terminal in non-canonical (raw) mode, you have to put the terminal in
1615: non-canonical mode yourself (using the C interface); the exception is
1616: @code{stdin}: Gforth automatically puts it into non-canonical mode.""
1617: #ifdef HAS_FILE
1618: fflush(stdout);
1619: c = key((FILE*)wfileid);
1620: #else
1621: c = key(stdin);
1622: #endif
1623:
1624: key?-file ( wfileid -- f ) gforth key_q_file
1625: ""@i{f} is true if at least one character can be read from @i{wfileid}
1626: without blocking. If you also want to use @code{read-file} or
1627: @code{read-line} on the file, you have to call @code{key?-file} or
1628: @code{key-file} first (these two words disable buffering).""
1629: #ifdef HAS_FILE
1630: fflush(stdout);
1631: f = key_query((FILE*)wfileid);
1632: #else
1633: f = key_query(stdin);
1634: #endif
1635:
1636: stdin ( -- wfileid ) gforth
1637: ""The standard input file of the Gforth process.""
1638: wfileid = (Cell)stdin;
1639:
1640: stdout ( -- wfileid ) gforth
1641: ""The standard output file of the Gforth process.""
1642: wfileid = (Cell)stdout;
1643:
1644: stderr ( -- wfileid ) gforth
1645: ""The standard error output file of the Gforth process.""
1646: wfileid = (Cell)stderr;
1647:
1648: \+os
1649:
1650: form ( -- urows ucols ) gforth
1651: ""The number of lines and columns in the terminal. These numbers may
1652: change with the window size. Note that it depends on the OS whether
1653: this reflects the actual size and changes with the window size
1654: (currently only on Unix-like OSs). On other OSs you just get a
1655: default, and can tell Gforth the terminal size by setting the
1656: environment variables @code{COLUMNS} and @code{LINES} before starting
1657: Gforth.""
1658: /* we could block SIGWINCH here to get a consistent size, but I don't
1659: think this is necessary or always beneficial */
1660: urows=rows;
1661: ucols=cols;
1662:
1663: wcwidth ( u -- n ) gforth
1664: ""The number of fixed-width characters per unicode character u""
1665: n = wcwidth(u);
1666:
1667: flush-icache ( c_addr u -- ) gforth flush_icache
1668: ""Make sure that the instruction cache of the processor (if there is
1669: one) does not contain stale data at @i{c-addr} and @i{u} bytes
1670: afterwards. @code{END-CODE} performs a @code{flush-icache}
1671: automatically. Caveat: @code{flush-icache} might not work on your
1672: installation; this is usually the case if direct threading is not
1673: supported on your machine (take a look at your @file{machine.h}) and
1674: your machine has a separate instruction cache. In such cases,
1675: @code{flush-icache} does nothing instead of flushing the instruction
1676: cache.""
1677: FLUSH_ICACHE(c_addr,u);
1678:
1679: (bye) ( n -- ) gforth paren_bye
1680: SUPER_END;
1681: return (Label *)n;
1682:
1683: (system) ( c_addr u -- wretval wior ) gforth paren_system
1684: wretval = gforth_system(c_addr, u);
1685: wior = IOR(wretval==-1 || (wretval==127 && errno != 0));
1686:
1687: getenv ( c_addr1 u1 -- c_addr2 u2 ) gforth
1688: ""The string @i{c-addr1 u1} specifies an environment variable. The string @i{c-addr2 u2}
1689: is the host operating system's expansion of that environment variable. If the
1690: environment variable does not exist, @i{c-addr2 u2} specifies a string 0 characters
1691: in length.""
1692: /* close ' to keep fontify happy */
1693: c_addr2 = (Char *)getenv(cstr(c_addr1,u1,1));
1694: u2 = (c_addr2 == NULL ? 0 : strlen((char *)c_addr2));
1695:
1696: open-pipe ( c_addr u wfam -- wfileid wior ) gforth open_pipe
1697: wfileid=(Cell)popen(cstr(c_addr,u,1),pfileattr[wfam]); /* ~ expansion of 1st arg? */
1698: wior = IOR(wfileid==0); /* !! the man page says that errno is not set reliably */
1699:
1700: close-pipe ( wfileid -- wretval wior ) gforth close_pipe
1701: wretval = pclose((FILE *)wfileid);
1702: wior = IOR(wretval==-1);
1703:
1704: time&date ( -- nsec nmin nhour nday nmonth nyear ) facility-ext time_and_date
1705: ""Report the current time of day. Seconds, minutes and hours are numbered from 0.
1706: Months are numbered from 1.""
1707: #if 1
1708: time_t now;
1709: struct tm *ltime;
1710: time(&now);
1711: ltime=localtime(&now);
1712: #else
1713: struct timeval time1;
1714: struct timezone zone1;
1715: struct tm *ltime;
1716: gettimeofday(&time1,&zone1);
1717: /* !! Single Unix specification:
1718: If tzp is not a null pointer, the behaviour is unspecified. */
1719: ltime=localtime((time_t *)&time1.tv_sec);
1720: #endif
1721: nyear =ltime->tm_year+1900;
1722: nmonth=ltime->tm_mon+1;
1723: nday =ltime->tm_mday;
1724: nhour =ltime->tm_hour;
1725: nmin =ltime->tm_min;
1726: nsec =ltime->tm_sec;
1727:
1728: ms ( u -- ) facility-ext
1729: ""Wait at least @i{n} milli-second.""
1730: gforth_ms(u);
1731:
1732: allocate ( u -- a_addr wior ) memory
1733: ""Allocate @i{u} address units of contiguous data space. The initial
1734: contents of the data space is undefined. If the allocation is successful,
1735: @i{a-addr} is the start address of the allocated region and @i{wior}
1736: is 0. If the allocation fails, @i{a-addr} is undefined and @i{wior}
1737: is a non-zero I/O result code.""
1738: a_addr = (Cell *)malloc(u?u:1);
1739: wior = IOR(a_addr==NULL);
1740:
1741: free ( a_addr -- wior ) memory
1742: ""Return the region of data space starting at @i{a-addr} to the system.
1743: The region must originally have been obtained using @code{allocate} or
1744: @code{resize}. If the operational is successful, @i{wior} is 0.
1745: If the operation fails, @i{wior} is a non-zero I/O result code.""
1746: free(a_addr);
1747: wior = 0;
1748:
1749: resize ( a_addr1 u -- a_addr2 wior ) memory
1750: ""Change the size of the allocated area at @i{a-addr1} to @i{u}
1751: address units, possibly moving the contents to a different
1752: area. @i{a-addr2} is the address of the resulting area.
1753: If the operation is successful, @i{wior} is 0.
1754: If the operation fails, @i{wior} is a non-zero
1755: I/O result code. If @i{a-addr1} is 0, Gforth's (but not the Standard)
1756: @code{resize} @code{allocate}s @i{u} address units.""
1757: /* the following check is not necessary on most OSs, but it is needed
1758: on SunOS 4.1.2. */
1759: /* close ' to keep fontify happy */
1760: if (a_addr1==NULL)
1761: a_addr2 = (Cell *)malloc(u);
1762: else
1763: a_addr2 = (Cell *)realloc(a_addr1, u);
1764: wior = IOR(a_addr2==NULL); /* !! Define a return code */
1765:
1766: strerror ( n -- c_addr u ) gforth
1767: c_addr = (Char *)strerror(n);
1768: u = strlen((char *)c_addr);
1769:
1770: strsignal ( n -- c_addr u ) gforth
1771: c_addr = (Char *)strsignal(n);
1772: u = strlen((char *)c_addr);
1773:
1774: call-c ( ... w -- ... ) gforth call_c
1775: ""Call the C function pointed to by @i{w}. The C function has to
1776: access the stack itself. The stack pointers are exported in the global
1777: variables @code{gforth_SP} and @code{gforth_FP}.""
1778: /* This is a first attempt at support for calls to C. This may change in
1779: the future */
1780: gforth_FP=fp;
1781: gforth_SP=sp;
1782: ((void (*)())w)();
1783: sp=gforth_SP;
1784: fp=gforth_FP;
1785:
1786: \+
1787: \+file
1788:
1789: close-file ( wfileid -- wior ) file close_file
1790: wior = IOR(fclose((FILE *)wfileid)==EOF);
1791:
1792: open-file ( c_addr u wfam -- wfileid wior ) file open_file
1793: wfileid = opencreate_file(tilde_cstr(c_addr,u,1), wfam, 0, &wior);
1794:
1795: create-file ( c_addr u wfam -- wfileid wior ) file create_file
1796: wfileid = opencreate_file(tilde_cstr(c_addr,u,1), wfam, O_CREAT|O_TRUNC, &wior);
1797:
1798: delete-file ( c_addr u -- wior ) file delete_file
1799: wior = IOR(unlink(tilde_cstr(c_addr, u, 1))==-1);
1800:
1801: rename-file ( c_addr1 u1 c_addr2 u2 -- wior ) file-ext rename_file
1802: ""Rename file @i{c_addr1 u1} to new name @i{c_addr2 u2}""
1803: wior = rename_file(c_addr1, u1, c_addr2, u2);
1804:
1805: file-position ( wfileid -- ud wior ) file file_position
1806: /* !! use tell and lseek? */
1807: ud = OFF2UD(ftello((FILE *)wfileid));
1808: wior = IOR(UD2OFF(ud)==-1);
1809:
1810: reposition-file ( ud wfileid -- wior ) file reposition_file
1811: wior = IOR(fseeko((FILE *)wfileid, UD2OFF(ud), SEEK_SET)==-1);
1812:
1813: file-size ( wfileid -- ud wior ) file file_size
1814: struct stat buf;
1815: wior = IOR(fstat(fileno((FILE *)wfileid), &buf)==-1);
1816: ud = OFF2UD(buf.st_size);
1817:
1818: resize-file ( ud wfileid -- wior ) file resize_file
1819: wior = IOR(ftruncate(fileno((FILE *)wfileid), UD2OFF(ud))==-1);
1820:
1821: read-file ( c_addr u1 wfileid -- u2 wior ) file read_file
1822: /* !! fread does not guarantee enough */
1823: u2 = fread(c_addr, sizeof(Char), u1, (FILE *)wfileid);
1824: wior = FILEIO(u2<u1 && ferror((FILE *)wfileid));
1825: /* !! is the value of ferror errno-compatible? */
1826: if (wior)
1827: clearerr((FILE *)wfileid);
1828:
1829: (read-line) ( c_addr u1 wfileid -- u2 flag u3 wior ) file paren_read_line
1830: struct Cellquad r = read_line(c_addr, u1, wfileid);
1831: u2 = r.n1;
1832: flag = r.n2;
1833: u3 = r.n3;
1834: wior = r.n4;
1835:
1836: \+
1837:
1838: write-file ( c_addr u1 wfileid -- wior ) file write_file
1839: /* !! fwrite does not guarantee enough */
1840: #ifdef HAS_FILE
1841: {
1842: UCell u2 = fwrite(c_addr, sizeof(Char), u1, (FILE *)wfileid);
1843: wior = FILEIO(u2<u1 && ferror((FILE *)wfileid));
1844: if (wior)
1845: clearerr((FILE *)wfileid);
1846: }
1847: #else
1848: TYPE(c_addr, u1);
1849: #endif
1850:
1851: emit-file ( c wfileid -- wior ) gforth emit_file
1852: #ifdef HAS_FILE
1853: wior = FILEIO(putc(c, (FILE *)wfileid)==EOF);
1854: if (wior)
1855: clearerr((FILE *)wfileid);
1856: #else
1857: PUTC(c);
1858: #endif
1859:
1860: \+file
1861:
1862: flush-file ( wfileid -- wior ) file-ext flush_file
1863: wior = IOR(fflush((FILE *) wfileid)==EOF);
1864:
1865: file-status ( c_addr u -- wfam wior ) file-ext file_status
1866: struct Cellpair r = file_status(c_addr, u);
1867: wfam = r.n1;
1868: wior = r.n2;
1869:
1870: file-eof? ( wfileid -- flag ) gforth file_eof_query
1871: flag = FLAG(feof((FILE *) wfileid));
1872:
1873: open-dir ( c_addr u -- wdirid wior ) gforth open_dir
1874: ""Open the directory specified by @i{c-addr, u}
1875: and return @i{dir-id} for futher access to it.""
1876: wdirid = (Cell)opendir(tilde_cstr(c_addr, u, 1));
1877: wior = IOR(wdirid == 0);
1878:
1879: read-dir ( c_addr u1 wdirid -- u2 flag wior ) gforth read_dir
1880: ""Attempt to read the next entry from the directory specified
1881: by @i{dir-id} to the buffer of length @i{u1} at address @i{c-addr}.
1882: If the attempt fails because there is no more entries,
1883: @i{ior}=0, @i{flag}=0, @i{u2}=0, and the buffer is unmodified.
1884: If the attempt to read the next entry fails because of any other reason,
1885: return @i{ior}<>0.
1886: If the attempt succeeds, store file name to the buffer at @i{c-addr}
1887: and return @i{ior}=0, @i{flag}=true and @i{u2} equal to the size of the file name.
1888: If the length of the file name is greater than @i{u1},
1889: store first @i{u1} characters from file name into the buffer and
1890: indicate "name too long" with @i{ior}, @i{flag}=true, and @i{u2}=@i{u1}.""
1891: struct dirent * dent;
1892: dent = readdir((DIR *)wdirid);
1893: wior = 0;
1894: flag = -1;
1895: if(dent == NULL) {
1896: u2 = 0;
1897: flag = 0;
1898: } else {
1899: u2 = strlen((char *)dent->d_name);
1900: if(u2 > u1) {
1901: u2 = u1;
1902: wior = -512-ENAMETOOLONG;
1903: }
1904: memmove(c_addr, dent->d_name, u2);
1905: }
1906:
1907: close-dir ( wdirid -- wior ) gforth close_dir
1908: ""Close the directory specified by @i{dir-id}.""
1909: wior = IOR(closedir((DIR *)wdirid));
1910:
1911: filename-match ( c_addr1 u1 c_addr2 u2 -- flag ) gforth match_file
1912: char * string = cstr(c_addr1, u1, 1);
1913: char * pattern = cstr(c_addr2, u2, 0);
1914: flag = FLAG(!fnmatch(pattern, string, 0));
1915:
1916: set-dir ( c_addr u -- wior ) gforth set_dir
1917: ""Change the current directory to @i{c-addr, u}.
1918: Return an error if this is not possible""
1919: wior = IOR(chdir(tilde_cstr(c_addr, u, 1)));
1920:
1921: get-dir ( c_addr1 u1 -- c_addr2 u2 ) gforth get_dir
1922: ""Store the current directory in the buffer specified by @{c-addr1, u1}.
1923: If the buffer size is not sufficient, return 0 0""
1924: c_addr2 = (Char *)getcwd((char *)c_addr1, u1);
1925: if(c_addr2 != NULL) {
1926: u2 = strlen((char *)c_addr2);
1927: } else {
1928: u2 = 0;
1929: }
1930:
1931: \+
1932:
1933: newline ( -- c_addr u ) gforth
1934: ""String containing the newline sequence of the host OS""
1935: char newline[] = {
1936: #if DIRSEP=='/'
1937: /* Unix */
1938: '\n'
1939: #else
1940: /* DOS, Win, OS/2 */
1941: '\r','\n'
1942: #endif
1943: };
1944: c_addr=(Char *)newline;
1945: u=sizeof(newline);
1946: :
1947: "newline count ;
1948: Create "newline e? crlf [IF] 2 c, $0D c, [ELSE] 1 c, [THEN] $0A c,
1949:
1950: \+os
1951:
1952: utime ( -- dtime ) gforth
1953: ""Report the current time in microseconds since some epoch.""
1954: struct timeval time1;
1955: gettimeofday(&time1,NULL);
1956: dtime = timeval2us(&time1);
1957:
1958: cputime ( -- duser dsystem ) gforth
1959: ""duser and dsystem are the respective user- and system-level CPU
1960: times used since the start of the Forth system (excluding child
1961: processes), in microseconds (the granularity may be much larger,
1962: however). On platforms without the getrusage call, it reports elapsed
1963: time (since some epoch) for duser and 0 for dsystem.""
1964: #ifdef HAVE_GETRUSAGE
1965: struct rusage usage;
1966: getrusage(RUSAGE_SELF, &usage);
1967: duser = timeval2us(&usage.ru_utime);
1968: dsystem = timeval2us(&usage.ru_stime);
1969: #else
1970: struct timeval time1;
1971: gettimeofday(&time1,NULL);
1972: duser = timeval2us(&time1);
1973: dsystem = DZERO;
1974: #endif
1975:
1976: \+
1977:
1978: \+floating
1979:
1980: \g floating
1981:
1982: comparisons(f, r1 r2, f_, r1, r2, gforth, gforth, float, gforth)
1983: comparisons(f0, r, f_zero_, r, 0., float, gforth, float, gforth)
1984:
1985: s>f ( n -- r ) float s_to_f
1986: r = n;
1987:
1988: d>f ( d -- r ) float d_to_f
1989: #ifdef BUGGY_LL_D2F
1990: extern double ldexp(double x, int exp);
1991: if (DHI(d)<0) {
1992: #ifdef BUGGY_LL_ADD
1993: DCell d2=dnegate(d);
1994: #else
1995: DCell d2=-d;
1996: #endif
1997: r = -(ldexp((Float)DHI(d2),CELL_BITS) + (Float)DLO(d2));
1998: } else
1999: r = ldexp((Float)DHI(d),CELL_BITS) + (Float)DLO(d);
2000: #else
2001: r = d;
2002: #endif
2003:
2004: f>d ( r -- d ) float f_to_d
2005: extern DCell double2ll(Float r);
2006: d = double2ll(r);
2007:
2008: f>s ( r -- n ) float f_to_s
2009: n = (Cell)r;
2010:
2011: f! ( r f_addr -- ) float f_store
2012: ""Store @i{r} into the float at address @i{f-addr}.""
2013: *f_addr = r;
2014:
2015: f@ ( f_addr -- r ) float f_fetch
2016: ""@i{r} is the float at address @i{f-addr}.""
2017: r = *f_addr;
2018:
2019: df@ ( df_addr -- r ) float-ext d_f_fetch
2020: ""Fetch the double-precision IEEE floating-point value @i{r} from the address @i{df-addr}.""
2021: #ifdef IEEE_FP
2022: r = *df_addr;
2023: #else
2024: !! df@
2025: #endif
2026:
2027: df! ( r df_addr -- ) float-ext d_f_store
2028: ""Store @i{r} as double-precision IEEE floating-point value to the
2029: address @i{df-addr}.""
2030: #ifdef IEEE_FP
2031: *df_addr = r;
2032: #else
2033: !! df!
2034: #endif
2035:
2036: sf@ ( sf_addr -- r ) float-ext s_f_fetch
2037: ""Fetch the single-precision IEEE floating-point value @i{r} from the address @i{sf-addr}.""
2038: #ifdef IEEE_FP
2039: r = *sf_addr;
2040: #else
2041: !! sf@
2042: #endif
2043:
2044: sf! ( r sf_addr -- ) float-ext s_f_store
2045: ""Store @i{r} as single-precision IEEE floating-point value to the
2046: address @i{sf-addr}.""
2047: #ifdef IEEE_FP
2048: *sf_addr = r;
2049: #else
2050: !! sf!
2051: #endif
2052:
2053: f+ ( r1 r2 -- r3 ) float f_plus
2054: r3 = r1+r2;
2055:
2056: f- ( r1 r2 -- r3 ) float f_minus
2057: r3 = r1-r2;
2058:
2059: f* ( r1 r2 -- r3 ) float f_star
2060: r3 = r1*r2;
2061:
2062: f/ ( r1 r2 -- r3 ) float f_slash
2063: r3 = r1/r2;
2064:
2065: f** ( r1 r2 -- r3 ) float-ext f_star_star
2066: ""@i{r3} is @i{r1} raised to the @i{r2}th power.""
2067: r3 = pow(r1,r2);
2068:
2069: fm* ( r1 n -- r2 ) gforth fm_star
2070: r2 = r1*n;
2071:
2072: fm/ ( r1 n -- r2 ) gforth fm_slash
2073: r2 = r1/n;
2074:
2075: fm*/ ( r1 n1 n2 -- r2 ) gforth fm_star_slash
2076: r2 = (r1*n1)/n2;
2077:
2078: f**2 ( r1 -- r2 ) gforth fm_square
2079: r2 = r1*r1;
2080:
2081: fnegate ( r1 -- r2 ) float f_negate
2082: r2 = - r1;
2083:
2084: fdrop ( r -- ) float f_drop
2085:
2086: fdup ( r -- r r ) float f_dupe
2087:
2088: fswap ( r1 r2 -- r2 r1 ) float f_swap
2089:
2090: fover ( r1 r2 -- r1 r2 r1 ) float f_over
2091:
2092: frot ( r1 r2 r3 -- r2 r3 r1 ) float f_rote
2093:
2094: fnip ( r1 r2 -- r2 ) gforth f_nip
2095:
2096: ftuck ( r1 r2 -- r2 r1 r2 ) gforth f_tuck
2097:
2098: float+ ( f_addr1 -- f_addr2 ) float float_plus
2099: ""@code{1 floats +}.""
2100: f_addr2 = f_addr1+1;
2101:
2102: floats ( n1 -- n2 ) float
2103: ""@i{n2} is the number of address units of @i{n1} floats.""
2104: n2 = n1*sizeof(Float);
2105:
2106: floor ( r1 -- r2 ) float
2107: ""Round towards the next smaller integral value, i.e., round toward negative infinity.""
2108: /* !! unclear wording */
2109: r2 = floor(r1);
2110:
2111: fround ( r1 -- r2 ) float f_round
2112: ""Round to the nearest integral value.""
2113: r2 = rint(r1);
2114:
2115: fmax ( r1 r2 -- r3 ) float f_max
2116: if (r1<r2)
2117: r3 = r2;
2118: else
2119: r3 = r1;
2120:
2121: fmin ( r1 r2 -- r3 ) float f_min
2122: if (r1<r2)
2123: r3 = r1;
2124: else
2125: r3 = r2;
2126:
2127: represent ( r c_addr u -- n f1 f2 ) float
2128: char *sig;
2129: size_t siglen;
2130: int flag;
2131: int decpt;
2132: sig=ecvt(r, u, &decpt, &flag);
2133: n=(r==0. ? 1 : decpt);
2134: f1=FLAG(flag!=0);
2135: f2=FLAG(isdigit((unsigned)(sig[0]))!=0);
2136: siglen=strlen((char *)sig);
2137: if (siglen>u) /* happens in glibc-2.1.3 if 999.. is rounded up */
2138: siglen=u;
2139: if (!f2) /* workaround Cygwin trailing 0s for Inf and Nan */
2140: for (; sig[siglen-1]=='0'; siglen--);
2141: ;
2142: memcpy(c_addr,sig,siglen);
2143: memset(c_addr+siglen,f2?'0':' ',u-siglen);
2144:
2145: >float ( c_addr u -- f:... flag ) float to_float
2146: ""Actual stack effect: ( c_addr u -- r t | f ). Attempt to convert the
2147: character string @i{c-addr u} to internal floating-point
2148: representation. If the string represents a valid floating-point number
2149: @i{r} is placed on the floating-point stack and @i{flag} is
2150: true. Otherwise, @i{flag} is false. A string of blanks is a special
2151: case and represents the floating-point number 0.""
2152: Float r;
2153: flag = to_float(c_addr, u, &r);
2154: if (flag) {
2155: fp--;
2156: fp[0]=r;
2157: }
2158:
2159: fabs ( r1 -- r2 ) float-ext f_abs
2160: r2 = fabs(r1);
2161:
2162: facos ( r1 -- r2 ) float-ext f_a_cos
2163: r2 = acos(r1);
2164:
2165: fasin ( r1 -- r2 ) float-ext f_a_sine
2166: r2 = asin(r1);
2167:
2168: fatan ( r1 -- r2 ) float-ext f_a_tan
2169: r2 = atan(r1);
2170:
2171: fatan2 ( r1 r2 -- r3 ) float-ext f_a_tan_two
2172: ""@i{r1/r2}=tan(@i{r3}). ANS Forth does not require, but probably
2173: intends this to be the inverse of @code{fsincos}. In gforth it is.""
2174: r3 = atan2(r1,r2);
2175:
2176: fcos ( r1 -- r2 ) float-ext f_cos
2177: r2 = cos(r1);
2178:
2179: fexp ( r1 -- r2 ) float-ext f_e_x_p
2180: r2 = exp(r1);
2181:
2182: fexpm1 ( r1 -- r2 ) float-ext f_e_x_p_m_one
2183: ""@i{r2}=@i{e}**@i{r1}@minus{}1""
2184: #ifdef HAVE_EXPM1
2185: extern double
2186: #ifdef NeXT
2187: const
2188: #endif
2189: expm1(double);
2190: r2 = expm1(r1);
2191: #else
2192: r2 = exp(r1)-1.;
2193: #endif
2194:
2195: fln ( r1 -- r2 ) float-ext f_l_n
2196: r2 = log(r1);
2197:
2198: flnp1 ( r1 -- r2 ) float-ext f_l_n_p_one
2199: ""@i{r2}=ln(@i{r1}+1)""
2200: #ifdef HAVE_LOG1P
2201: extern double
2202: #ifdef NeXT
2203: const
2204: #endif
2205: log1p(double);
2206: r2 = log1p(r1);
2207: #else
2208: r2 = log(r1+1.);
2209: #endif
2210:
2211: flog ( r1 -- r2 ) float-ext f_log
2212: ""The decimal logarithm.""
2213: r2 = log10(r1);
2214:
2215: falog ( r1 -- r2 ) float-ext f_a_log
2216: ""@i{r2}=10**@i{r1}""
2217: extern double pow10(double);
2218: r2 = pow10(r1);
2219:
2220: fsin ( r1 -- r2 ) float-ext f_sine
2221: r2 = sin(r1);
2222:
2223: fsincos ( r1 -- r2 r3 ) float-ext f_sine_cos
2224: ""@i{r2}=sin(@i{r1}), @i{r3}=cos(@i{r1})""
2225: r2 = sin(r1);
2226: r3 = cos(r1);
2227:
2228: fsqrt ( r1 -- r2 ) float-ext f_square_root
2229: r2 = sqrt(r1);
2230:
2231: ftan ( r1 -- r2 ) float-ext f_tan
2232: r2 = tan(r1);
2233: :
2234: fsincos f/ ;
2235:
2236: fsinh ( r1 -- r2 ) float-ext f_cinch
2237: r2 = sinh(r1);
2238: :
2239: fexpm1 fdup fdup 1. d>f f+ f/ f+ f2/ ;
2240:
2241: fcosh ( r1 -- r2 ) float-ext f_cosh
2242: r2 = cosh(r1);
2243: :
2244: fexp fdup 1/f f+ f2/ ;
2245:
2246: ftanh ( r1 -- r2 ) float-ext f_tan_h
2247: r2 = tanh(r1);
2248: :
2249: f2* fexpm1 fdup 2. d>f f+ f/ ;
2250:
2251: fasinh ( r1 -- r2 ) float-ext f_a_cinch
2252: r2 = asinh(r1);
2253: :
2254: fdup fdup f* 1. d>f f+ fsqrt f/ fatanh ;
2255:
2256: facosh ( r1 -- r2 ) float-ext f_a_cosh
2257: r2 = acosh(r1);
2258: :
2259: fdup fdup f* 1. d>f f- fsqrt f+ fln ;
2260:
2261: fatanh ( r1 -- r2 ) float-ext f_a_tan_h
2262: r2 = atanh(r1);
2263: :
2264: fdup f0< >r fabs 1. d>f fover f- f/ f2* flnp1 f2/
2265: r> IF fnegate THEN ;
2266:
2267: sfloats ( n1 -- n2 ) float-ext s_floats
2268: ""@i{n2} is the number of address units of @i{n1}
2269: single-precision IEEE floating-point numbers.""
2270: n2 = n1*sizeof(SFloat);
2271:
2272: dfloats ( n1 -- n2 ) float-ext d_floats
2273: ""@i{n2} is the number of address units of @i{n1}
2274: double-precision IEEE floating-point numbers.""
2275: n2 = n1*sizeof(DFloat);
2276:
2277: sfaligned ( c_addr -- sf_addr ) float-ext s_f_aligned
2278: ""@i{sf-addr} is the first single-float-aligned address greater
2279: than or equal to @i{c-addr}.""
2280: sf_addr = (SFloat *)((((Cell)c_addr)+(sizeof(SFloat)-1))&(-sizeof(SFloat)));
2281: :
2282: [ 1 sfloats 1- ] Literal + [ -1 sfloats ] Literal and ;
2283:
2284: dfaligned ( c_addr -- df_addr ) float-ext d_f_aligned
2285: ""@i{df-addr} is the first double-float-aligned address greater
2286: than or equal to @i{c-addr}.""
2287: df_addr = (DFloat *)((((Cell)c_addr)+(sizeof(DFloat)-1))&(-sizeof(DFloat)));
2288: :
2289: [ 1 dfloats 1- ] Literal + [ -1 dfloats ] Literal and ;
2290:
2291: v* ( f_addr1 nstride1 f_addr2 nstride2 ucount -- r ) gforth v_star
2292: ""dot-product: r=v1*v2. The first element of v1 is at f_addr1, the
2293: next at f_addr1+nstride1 and so on (similar for v2). Both vectors have
2294: ucount elements.""
2295: r = v_star(f_addr1, nstride1, f_addr2, nstride2, ucount);
2296: :
2297: >r swap 2swap swap 0e r> 0 ?DO
2298: dup f@ over + 2swap dup f@ f* f+ over + 2swap
2299: LOOP 2drop 2drop ;
2300:
2301: faxpy ( ra f_x nstridex f_y nstridey ucount -- ) gforth
2302: ""vy=ra*vx+vy""
2303: faxpy(ra, f_x, nstridex, f_y, nstridey, ucount);
2304: :
2305: >r swap 2swap swap r> 0 ?DO
2306: fdup dup f@ f* over + 2swap dup f@ f+ dup f! over + 2swap
2307: LOOP 2drop 2drop fdrop ;
2308:
2309: \+
2310:
2311: \ The following words access machine/OS/installation-dependent
2312: \ Gforth internals
2313: \ !! how about environmental queries DIRECT-THREADED,
2314: \ INDIRECT-THREADED, TOS-CACHED, FTOS-CACHED, CODEFIELD-DOES */
2315:
2316: \ local variable implementation primitives
2317:
2318: \+glocals
2319:
2320: \g locals
2321:
2322: @local# ( #noffset -- w ) gforth fetch_local_number
2323: w = *(Cell *)(lp+noffset);
2324:
2325: @local0 ( -- w ) new fetch_local_zero
2326: w = ((Cell *)lp)[0];
2327:
2328: @local1 ( -- w ) new fetch_local_four
2329: w = ((Cell *)lp)[1];
2330:
2331: @local2 ( -- w ) new fetch_local_eight
2332: w = ((Cell *)lp)[2];
2333:
2334: @local3 ( -- w ) new fetch_local_twelve
2335: w = ((Cell *)lp)[3];
2336:
2337: \+floating
2338:
2339: f@local# ( #noffset -- r ) gforth f_fetch_local_number
2340: r = *(Float *)(lp+noffset);
2341:
2342: f@local0 ( -- r ) new f_fetch_local_zero
2343: r = ((Float *)lp)[0];
2344:
2345: f@local1 ( -- r ) new f_fetch_local_eight
2346: r = ((Float *)lp)[1];
2347:
2348: \+
2349:
2350: laddr# ( #noffset -- c_addr ) gforth laddr_number
2351: /* this can also be used to implement lp@ */
2352: c_addr = (Char *)(lp+noffset);
2353:
2354: lp+!# ( #noffset -- ) gforth lp_plus_store_number
2355: ""used with negative immediate values it allocates memory on the
2356: local stack, a positive immediate argument drops memory from the local
2357: stack""
2358: lp += noffset;
2359:
2360: lp- ( -- ) new minus_four_lp_plus_store
2361: lp += -sizeof(Cell);
2362:
2363: lp+ ( -- ) new eight_lp_plus_store
2364: lp += sizeof(Float);
2365:
2366: lp+2 ( -- ) new sixteen_lp_plus_store
2367: lp += 2*sizeof(Float);
2368:
2369: lp! ( c_addr -- ) gforth lp_store
2370: lp = (Address)c_addr;
2371:
2372: >l ( w -- ) gforth to_l
2373: lp -= sizeof(Cell);
2374: *(Cell *)lp = w;
2375:
2376: \+floating
2377:
2378: f>l ( r -- ) gforth f_to_l
2379: lp -= sizeof(Float);
2380: *(Float *)lp = r;
2381:
2382: fpick ( f:... u -- f:... r ) gforth
2383: ""Actually the stack effect is @code{ r0 ... ru u -- r0 ... ru r0 }.""
2384: r = fp[u];
2385: :
2386: floats fp@ + f@ ;
2387:
2388: \+
2389: \+
2390:
2391: \+OS
2392:
2393: \g syslib
2394:
2395: open-lib ( c_addr1 u1 -- u2 ) gforth open_lib
2396: #if 1
2397: u2 = (UCell)lt_dlopen(cstr(c_addr1, u1, 1));
2398: #elif defined(HAVE_LIBDL) || defined(HAVE_DLOPEN)
2399: #ifndef RTLD_GLOBAL
2400: #define RTLD_GLOBAL 0
2401: #endif
2402: u2=(UCell) dlopen(cstr(c_addr1, u1, 1), RTLD_GLOBAL | RTLD_LAZY);
2403: #else
2404: # ifdef _WIN32
2405: u2 = (Cell) GetModuleHandle(cstr(c_addr1, u1, 1));
2406: # else
2407: #warning Define open-lib!
2408: u2 = 0;
2409: # endif
2410: #endif
2411:
2412: lib-sym ( c_addr1 u1 u2 -- u3 ) gforth lib_sym
2413: #if 1
2414: u3 = (UCell) lt_dlsym((lt_dlhandle)u2, cstr(c_addr1, u1, 1));
2415: #elif defined(HAVE_LIBDL) || defined(HAVE_DLOPEN)
2416: u3 = (UCell) dlsym((void*)u2,cstr(c_addr1, u1, 1));
2417: #else
2418: # ifdef _WIN32
2419: u3 = (Cell) GetProcAddress((HMODULE)u2, cstr(c_addr1, u1, 1));
2420: # else
2421: #warning Define lib-sym!
2422: u3 = 0;
2423: # endif
2424: #endif
2425:
2426: wcall ( ... u -- ... ) gforth
2427: gforth_FP=fp;
2428: sp=(Cell*)(SYSCALL(Cell*(*)(Cell *, void *))u)(sp, &gforth_FP);
2429: fp=gforth_FP;
2430:
2431: uw@ ( c_addr -- u ) gforth u_w_fetch
2432: ""@i{u} is the zero-extended 16-bit value stored at @i{c_addr}.""
2433: u = *(UWyde*)(c_addr);
2434:
2435: sw@ ( c_addr -- n ) gforth s_w_fetch
2436: ""@i{n} is the sign-extended 16-bit value stored at @i{c_addr}.""
2437: n = *(Wyde*)(c_addr);
2438:
2439: w! ( w c_addr -- ) gforth w_store
2440: ""Store the bottom 16 bits of @i{w} at @i{c_addr}.""
2441: *(Wyde*)(c_addr) = w;
2442:
2443: ul@ ( c_addr -- u ) gforth u_l_fetch
2444: ""@i{u} is the zero-extended 32-bit value stored at @i{c_addr}.""
2445: u = *(UTetrabyte*)(c_addr);
2446:
2447: sl@ ( c_addr -- n ) gforth s_l_fetch
2448: ""@i{n} is the sign-extended 32-bit value stored at @i{c_addr}.""
2449: n = *(Tetrabyte*)(c_addr);
2450:
2451: l! ( w c_addr -- ) gforth l_store
2452: ""Store the bottom 32 bits of @i{w} at @i{c_addr}.""
2453: *(Tetrabyte*)(c_addr) = w;
2454:
2455: \+FFCALL
2456:
2457: av-start-void ( c_addr -- ) gforth av_start_void
2458: av_start_void(alist, c_addr);
2459:
2460: av-start-int ( c_addr -- ) gforth av_start_int
2461: av_start_int(alist, c_addr, &irv);
2462:
2463: av-start-float ( c_addr -- ) gforth av_start_float
2464: av_start_float(alist, c_addr, &frv);
2465:
2466: av-start-double ( c_addr -- ) gforth av_start_double
2467: av_start_double(alist, c_addr, &drv);
2468:
2469: av-start-longlong ( c_addr -- ) gforth av_start_longlong
2470: av_start_longlong(alist, c_addr, &llrv);
2471:
2472: av-start-ptr ( c_addr -- ) gforth av_start_ptr
2473: av_start_ptr(alist, c_addr, void*, &prv);
2474:
2475: av-int ( w -- ) gforth av_int
2476: av_int(alist, w);
2477:
2478: av-float ( r -- ) gforth av_float
2479: av_float(alist, r);
2480:
2481: av-double ( r -- ) gforth av_double
2482: av_double(alist, r);
2483:
2484: av-longlong ( d -- ) gforth av_longlong
2485: #ifdef BUGGY_LL_SIZE
2486: av_longlong(alist, DLO(d));
2487: #else
2488: av_longlong(alist, d);
2489: #endif
2490:
2491: av-ptr ( c_addr -- ) gforth av_ptr
2492: av_ptr(alist, void*, c_addr);
2493:
2494: av-int-r ( R:w -- ) gforth av_int_r
2495: av_int(alist, w);
2496:
2497: av-float-r ( -- ) gforth av_float_r
2498: float r = *(Float*)lp;
2499: lp += sizeof(Float);
2500: av_float(alist, r);
2501:
2502: av-double-r ( -- ) gforth av_double_r
2503: double r = *(Float*)lp;
2504: lp += sizeof(Float);
2505: av_double(alist, r);
2506:
2507: av-longlong-r ( R:d -- ) gforth av_longlong_r
2508: #ifdef BUGGY_LL_SIZE
2509: av_longlong(alist, DLO(d));
2510: #else
2511: av_longlong(alist, d);
2512: #endif
2513:
2514: av-ptr-r ( R:c_addr -- ) gforth av_ptr_r
2515: av_ptr(alist, void*, c_addr);
2516:
2517: av-call-void ( ... -- ... ) gforth av_call_void
2518: SAVE_REGS
2519: av_call(alist);
2520: REST_REGS
2521:
2522: av-call-int ( ... -- ... w ) gforth av_call_int
2523: SAVE_REGS
2524: av_call(alist);
2525: REST_REGS
2526: w = irv;
2527:
2528: av-call-float ( ... -- ... r ) gforth av_call_float
2529: SAVE_REGS
2530: av_call(alist);
2531: REST_REGS
2532: r = frv;
2533:
2534: av-call-double ( ... -- ... r ) gforth av_call_double
2535: SAVE_REGS
2536: av_call(alist);
2537: REST_REGS
2538: r = drv;
2539:
2540: av-call-longlong ( ... -- ... d ) gforth av_call_longlong
2541: SAVE_REGS
2542: av_call(alist);
2543: REST_REGS
2544: #ifdef BUGGY_LONG_LONG
2545: DLO_IS(d, llrv);
2546: DHI_IS(d, 0);
2547: #else
2548: d = llrv;
2549: #endif
2550:
2551: av-call-ptr ( ... -- ... c_addr ) gforth av_call_ptr
2552: SAVE_REGS
2553: av_call(alist);
2554: REST_REGS
2555: c_addr = prv;
2556:
2557: alloc-callback ( a_ip -- c_addr ) gforth alloc_callback
2558: c_addr = (char *)alloc_callback(gforth_callback, (Xt *)a_ip);
2559:
2560: va-start-void ( -- ) gforth va_start_void
2561: va_start_void(gforth_clist);
2562:
2563: va-start-int ( -- ) gforth va_start_int
2564: va_start_int(gforth_clist);
2565:
2566: va-start-longlong ( -- ) gforth va_start_longlong
2567: va_start_longlong(gforth_clist);
2568:
2569: va-start-ptr ( -- ) gforth va_start_ptr
2570: va_start_ptr(gforth_clist, (char *));
2571:
2572: va-start-float ( -- ) gforth va_start_float
2573: va_start_float(gforth_clist);
2574:
2575: va-start-double ( -- ) gforth va_start_double
2576: va_start_double(gforth_clist);
2577:
2578: va-arg-int ( -- w ) gforth va_arg_int
2579: w = va_arg_int(gforth_clist);
2580:
2581: va-arg-longlong ( -- d ) gforth va_arg_longlong
2582: #ifdef BUGGY_LONG_LONG
2583: DLO_IS(d, va_arg_longlong(gforth_clist));
2584: DHI_IS(d, 0);
2585: #else
2586: d = va_arg_longlong(gforth_clist);
2587: #endif
2588:
2589: va-arg-ptr ( -- c_addr ) gforth va_arg_ptr
2590: c_addr = (char *)va_arg_ptr(gforth_clist,char*);
2591:
2592: va-arg-float ( -- r ) gforth va_arg_float
2593: r = va_arg_float(gforth_clist);
2594:
2595: va-arg-double ( -- r ) gforth va_arg_double
2596: r = va_arg_double(gforth_clist);
2597:
2598: va-return-void ( -- ) gforth va_return_void
2599: va_return_void(gforth_clist);
2600: return 0;
2601:
2602: va-return-int ( w -- ) gforth va_return_int
2603: va_return_int(gforth_clist, w);
2604: return 0;
2605:
2606: va-return-ptr ( c_addr -- ) gforth va_return_ptr
2607: va_return_ptr(gforth_clist, void *, c_addr);
2608: return 0;
2609:
2610: va-return-longlong ( d -- ) gforth va_return_longlong
2611: #ifdef BUGGY_LONG_LONG
2612: va_return_longlong(gforth_clist, d.lo);
2613: #else
2614: va_return_longlong(gforth_clist, d);
2615: #endif
2616: return 0;
2617:
2618: va-return-float ( r -- ) gforth va_return_float
2619: va_return_float(gforth_clist, r);
2620: return 0;
2621:
2622: va-return-double ( r -- ) gforth va_return_double
2623: va_return_double(gforth_clist, r);
2624: return 0;
2625:
2626: \+
2627:
2628: \+LIBFFI
2629:
2630: ffi-type ( n -- a_type ) gforth ffi_type
2631: static void* ffi_types[] =
2632: { &ffi_type_void,
2633: &ffi_type_uint8, &ffi_type_sint8,
2634: &ffi_type_uint16, &ffi_type_sint16,
2635: &ffi_type_uint32, &ffi_type_sint32,
2636: &ffi_type_uint64, &ffi_type_sint64,
2637: &ffi_type_float, &ffi_type_double, &ffi_type_longdouble,
2638: &ffi_type_pointer };
2639: a_type = ffi_types[n];
2640:
2641: ffi-size ( n1 -- n2 ) gforth ffi_size
2642: static int ffi_sizes[] =
2643: { sizeof(ffi_cif), sizeof(ffi_closure) };
2644: n2 = ffi_sizes[n1];
2645:
2646: ffi-prep-cif ( a_atypes n a_rtype a_cif -- w ) gforth ffi_prep_cif
2647: w = ffi_prep_cif((ffi_cif *)a_cif, FFI_DEFAULT_ABI, n,
2648: (ffi_type *)a_rtype, (ffi_type **)a_atypes);
2649:
2650: ffi-call ( a_avalues a_rvalue a_ip a_cif -- ) gforth ffi_call
2651: SAVE_REGS
2652: ffi_call((ffi_cif *)a_cif, (void(*)())a_ip, (void *)a_rvalue, (void **)a_avalues);
2653: REST_REGS
2654:
2655: ffi-prep-closure ( a_ip a_cif a_closure -- w ) gforth ffi_prep_closure
2656: w = ffi_prep_closure((ffi_closure *)a_closure, (ffi_cif *)a_cif, gforth_callback, (void *)a_ip);
2657:
2658: ffi-2@ ( a_addr -- d ) gforth ffi_2fetch
2659: #ifdef BUGGY_LONG_LONG
2660: DLO_IS(d, *(Cell*)(*a_addr));
2661: DHI_IS(d, 0);
2662: #else
2663: d = *(DCell*)(a_addr);
2664: #endif
2665:
2666: ffi-2! ( d a_addr -- ) gforth ffi_2store
2667: #ifdef BUGGY_LONG_LONG
2668: *(Cell*)(a_addr) = DLO(d);
2669: #else
2670: *(DCell*)(a_addr) = d;
2671: #endif
2672:
2673: ffi-arg-int ( -- w ) gforth ffi_arg_int
2674: w = *(int *)(*gforth_clist++);
2675:
2676: ffi-arg-long ( -- w ) gforth ffi_arg_long
2677: w = *(long *)(*gforth_clist++);
2678:
2679: ffi-arg-longlong ( -- d ) gforth ffi_arg_longlong
2680: #ifdef BUGGY_LONG_LONG
2681: DLO_IS(d, *(Cell*)(*gforth_clist++));
2682: DHI_IS(d, -(*(Cell*)(*gforth_clist++)<0));
2683: #else
2684: d = *(DCell*)(*gforth_clist++);
2685: #endif
2686:
2687: ffi-arg-dlong ( -- d ) gforth ffi_arg_dlong
2688: #ifdef BUGGY_LONG_LONG
2689: DLO_IS(d, *(Cell*)(*gforth_clist++));
2690: DHI_IS(d, -(*(Cell*)(*gforth_clist++)<0));
2691: #else
2692: d = *(Cell*)(*gforth_clist++);
2693: #endif
2694:
2695: ffi-arg-ptr ( -- c_addr ) gforth ffi_arg_ptr
2696: c_addr = *(Char **)(*gforth_clist++);
2697:
2698: ffi-arg-float ( -- r ) gforth ffi_arg_float
2699: r = *(float*)(*gforth_clist++);
2700:
2701: ffi-arg-double ( -- r ) gforth ffi_arg_double
2702: r = *(double*)(*gforth_clist++);
2703:
2704: ffi-ret-void ( -- ) gforth ffi_ret_void
2705: return 0;
2706:
2707: ffi-ret-int ( w -- ) gforth ffi_ret_int
2708: *(int*)(gforth_ritem) = w;
2709: return 0;
2710:
2711: ffi-ret-longlong ( d -- ) gforth ffi_ret_longlong
2712: #ifdef BUGGY_LONG_LONG
2713: *(Cell*)(gforth_ritem) = DLO(d);
2714: #else
2715: *(DCell*)(gforth_ritem) = d;
2716: #endif
2717: return 0;
2718:
2719: ffi-ret-dlong ( d -- ) gforth ffi_ret_dlong
2720: #ifdef BUGGY_LONG_LONG
2721: *(Cell*)(gforth_ritem) = DLO(d);
2722: #else
2723: *(Cell*)(gforth_ritem) = d;
2724: #endif
2725: return 0;
2726:
2727: ffi-ret-long ( n -- ) gforth ffi_ret_long
2728: *(Cell*)(gforth_ritem) = n;
2729: return 0;
2730:
2731: ffi-ret-ptr ( c_addr -- ) gforth ffi_ret_ptr
2732: *(Char **)(gforth_ritem) = c_addr;
2733: return 0;
2734:
2735: ffi-ret-float ( r -- ) gforth ffi_ret_float
2736: *(float*)(gforth_ritem) = r;
2737: return 0;
2738:
2739: ffi-ret-double ( r -- ) gforth ffi_ret_double
2740: *(double*)(gforth_ritem) = r;
2741: return 0;
2742:
2743: \+
2744:
2745: \+OLDCALL
2746:
2747: define(`uploop',
2748: `pushdef(`$1', `$2')_uploop(`$1', `$2', `$3', `$4', `$5')`'popdef(`$1')')
2749: define(`_uploop',
2750: `ifelse($1, `$3', `$5',
2751: `$4`'define(`$1', incr($1))_uploop(`$1', `$2', `$3', `$4', `$5')')')
2752:
2753: \ argflist(argnum): Forth argument list
2754: define(argflist,
2755: `ifelse($1, 0, `',
2756: `uploop(`_i', 1, $1, ``u''`_i ', ``u''`_i')')')
2757: \ argdlist(argnum): declare C's arguments
2758: define(argdlist,
2759: `ifelse($1, 0, `',
2760: `uploop(`_i', 1, $1, `Cell, ', `Cell')')')
2761: \ argclist(argnum): pass C's arguments
2762: define(argclist,
2763: `ifelse($1, 0, `',
2764: `uploop(`_i', 1, $1, ``u''`_i, ', ``u''`_i')')')
2765: \ icall(argnum)
2766: define(icall,
2767: `icall$1 ( argflist($1) u -- uret ) gforth
2768: uret = (SYSCALL(Cell(*)(argdlist($1)))u)(argclist($1));
2769:
2770: ')
2771: define(fcall,
2772: `fcall$1 ( argflist($1) u -- rret ) gforth
2773: rret = (SYSCALL(Float(*)(argdlist($1)))u)(argclist($1));
2774:
2775: ')
2776:
2777: \ close ' to keep fontify happy
2778:
2779: uploop(i, 0, 7, `icall(i)')
2780: icall(20)
2781: uploop(i, 0, 7, `fcall(i)')
2782: fcall(20)
2783:
2784: \+
2785:
2786: lib-error ( -- c_addr u ) gforth lib_error
2787: c_addr = lt_dlerror();
2788: u = (c_addr == NULL) ? 0 : strlen(c_addr);
2789:
2790: \+
2791: \g peephole
2792:
2793: \+peephole
2794:
2795: compile-prim1 ( a_prim -- ) gforth compile_prim1
2796: ""compile prim (incl. immargs) at @var{a_prim}""
2797: compile_prim1(a_prim);
2798:
2799: finish-code ( ... -- ... ) gforth finish_code
2800: ""Perform delayed steps in code generation (branch resolution, I-cache
2801: flushing).""
2802: /* The ... above are a workaround for a bug in gcc-2.95, which fails
2803: to save spTOS (gforth-fast --enable-force-reg) */
2804: finish_code();
2805:
2806: forget-dyncode ( c_code -- f ) gforth-internal forget_dyncode
2807: f = forget_dyncode(c_code);
2808:
2809: decompile-prim ( a_code -- a_prim ) gforth-internal decompile_prim
2810: ""a_prim is the code address of the primitive that has been
2811: compile_prim1ed to a_code""
2812: a_prim = (Cell *)decompile_code((Label)a_code);
2813:
2814: \ set-next-code and call2 do not appear in images and can be
2815: \ renumbered arbitrarily
2816:
2817: set-next-code ( #w -- ) gforth set_next_code
2818: #ifdef NO_IP
2819: next_code = (Label)w;
2820: #endif
2821:
2822: call2 ( #a_callee #a_ret_addr -- R:a_ret_addr ) gforth
2823: /* call with explicit return address */
2824: #ifdef NO_IP
2825: INST_TAIL;
2826: JUMP(a_callee);
2827: #else
2828: assert(0);
2829: #endif
2830:
2831: tag-offsets ( -- a_addr ) gforth tag_offsets
2832: extern Cell groups[32];
2833: a_addr = groups;
2834:
2835: \+
2836:
2837: \g static_super
2838:
2839: ifdef(`STACK_CACHE_FILE',
2840: `include(peeprules.vmg)')
2841:
2842: \g end
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to prim CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [gforth] / gforth