]> git.ipfire.org Git - thirdparty/binutils-gdb.git/blob - gdb/f-lang.c
gdb/fortran: Print 'void' type in lower case
[thirdparty/binutils-gdb.git] / gdb / f-lang.c
1 /* Fortran language support routines for GDB, the GNU debugger.
2
3 Copyright (C) 1993-2019 Free Software Foundation, Inc.
4
5 Contributed by Motorola. Adapted from the C parser by Farooq Butt
6 (fmbutt@engage.sps.mot.com).
7
8 This file is part of GDB.
9
10 This program is free software; you can redistribute it and/or modify
11 it under the terms of the GNU General Public License as published by
12 the Free Software Foundation; either version 3 of the License, or
13 (at your option) any later version.
14
15 This program is distributed in the hope that it will be useful,
16 but WITHOUT ANY WARRANTY; without even the implied warranty of
17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 GNU General Public License for more details.
19
20 You should have received a copy of the GNU General Public License
21 along with this program. If not, see <http://www.gnu.org/licenses/>. */
22
23 #include "defs.h"
24 #include "symtab.h"
25 #include "gdbtypes.h"
26 #include "expression.h"
27 #include "parser-defs.h"
28 #include "language.h"
29 #include "varobj.h"
30 #include "gdbcore.h"
31 #include "f-lang.h"
32 #include "valprint.h"
33 #include "value.h"
34 #include "cp-support.h"
35 #include "charset.h"
36 #include "c-lang.h"
37 #include "target-float.h"
38
39 #include <math.h>
40
41 /* Local functions */
42
43 static void f_printchar (int c, struct type *type, struct ui_file * stream);
44 static void f_emit_char (int c, struct type *type,
45 struct ui_file * stream, int quoter);
46
47 /* Return the encoding that should be used for the character type
48 TYPE. */
49
50 static const char *
51 f_get_encoding (struct type *type)
52 {
53 const char *encoding;
54
55 switch (TYPE_LENGTH (type))
56 {
57 case 1:
58 encoding = target_charset (get_type_arch (type));
59 break;
60 case 4:
61 if (gdbarch_byte_order (get_type_arch (type)) == BFD_ENDIAN_BIG)
62 encoding = "UTF-32BE";
63 else
64 encoding = "UTF-32LE";
65 break;
66
67 default:
68 error (_("unrecognized character type"));
69 }
70
71 return encoding;
72 }
73
74 /* Print the character C on STREAM as part of the contents of a literal
75 string whose delimiter is QUOTER. Note that that format for printing
76 characters and strings is language specific.
77 FIXME: This is a copy of the same function from c-exp.y. It should
78 be replaced with a true F77 version. */
79
80 static void
81 f_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
82 {
83 const char *encoding = f_get_encoding (type);
84
85 generic_emit_char (c, type, stream, quoter, encoding);
86 }
87
88 /* Implementation of la_printchar. */
89
90 static void
91 f_printchar (int c, struct type *type, struct ui_file *stream)
92 {
93 fputs_filtered ("'", stream);
94 LA_EMIT_CHAR (c, type, stream, '\'');
95 fputs_filtered ("'", stream);
96 }
97
98 /* Print the character string STRING, printing at most LENGTH characters.
99 Printing stops early if the number hits print_max; repeat counts
100 are printed as appropriate. Print ellipses at the end if we
101 had to stop before printing LENGTH characters, or if FORCE_ELLIPSES.
102 FIXME: This is a copy of the same function from c-exp.y. It should
103 be replaced with a true F77 version. */
104
105 static void
106 f_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
107 unsigned int length, const char *encoding, int force_ellipses,
108 const struct value_print_options *options)
109 {
110 const char *type_encoding = f_get_encoding (type);
111
112 if (TYPE_LENGTH (type) == 4)
113 fputs_filtered ("4_", stream);
114
115 if (!encoding || !*encoding)
116 encoding = type_encoding;
117
118 generic_printstr (stream, type, string, length, encoding,
119 force_ellipses, '\'', 0, options);
120 }
121 \f
122
123 /* Table of operators and their precedences for printing expressions. */
124
125 static const struct op_print f_op_print_tab[] =
126 {
127 {"+", BINOP_ADD, PREC_ADD, 0},
128 {"+", UNOP_PLUS, PREC_PREFIX, 0},
129 {"-", BINOP_SUB, PREC_ADD, 0},
130 {"-", UNOP_NEG, PREC_PREFIX, 0},
131 {"*", BINOP_MUL, PREC_MUL, 0},
132 {"/", BINOP_DIV, PREC_MUL, 0},
133 {"DIV", BINOP_INTDIV, PREC_MUL, 0},
134 {"MOD", BINOP_REM, PREC_MUL, 0},
135 {"=", BINOP_ASSIGN, PREC_ASSIGN, 1},
136 {".OR.", BINOP_LOGICAL_OR, PREC_LOGICAL_OR, 0},
137 {".AND.", BINOP_LOGICAL_AND, PREC_LOGICAL_AND, 0},
138 {".NOT.", UNOP_LOGICAL_NOT, PREC_PREFIX, 0},
139 {".EQ.", BINOP_EQUAL, PREC_EQUAL, 0},
140 {".NE.", BINOP_NOTEQUAL, PREC_EQUAL, 0},
141 {".LE.", BINOP_LEQ, PREC_ORDER, 0},
142 {".GE.", BINOP_GEQ, PREC_ORDER, 0},
143 {".GT.", BINOP_GTR, PREC_ORDER, 0},
144 {".LT.", BINOP_LESS, PREC_ORDER, 0},
145 {"**", UNOP_IND, PREC_PREFIX, 0},
146 {"@", BINOP_REPEAT, PREC_REPEAT, 0},
147 {NULL, OP_NULL, PREC_REPEAT, 0}
148 };
149 \f
150 enum f_primitive_types {
151 f_primitive_type_character,
152 f_primitive_type_logical,
153 f_primitive_type_logical_s1,
154 f_primitive_type_logical_s2,
155 f_primitive_type_logical_s8,
156 f_primitive_type_integer,
157 f_primitive_type_integer_s2,
158 f_primitive_type_real,
159 f_primitive_type_real_s8,
160 f_primitive_type_real_s16,
161 f_primitive_type_complex_s8,
162 f_primitive_type_complex_s16,
163 f_primitive_type_void,
164 nr_f_primitive_types
165 };
166
167 static void
168 f_language_arch_info (struct gdbarch *gdbarch,
169 struct language_arch_info *lai)
170 {
171 const struct builtin_f_type *builtin = builtin_f_type (gdbarch);
172
173 lai->string_char_type = builtin->builtin_character;
174 lai->primitive_type_vector
175 = GDBARCH_OBSTACK_CALLOC (gdbarch, nr_f_primitive_types + 1,
176 struct type *);
177
178 lai->primitive_type_vector [f_primitive_type_character]
179 = builtin->builtin_character;
180 lai->primitive_type_vector [f_primitive_type_logical]
181 = builtin->builtin_logical;
182 lai->primitive_type_vector [f_primitive_type_logical_s1]
183 = builtin->builtin_logical_s1;
184 lai->primitive_type_vector [f_primitive_type_logical_s2]
185 = builtin->builtin_logical_s2;
186 lai->primitive_type_vector [f_primitive_type_logical_s8]
187 = builtin->builtin_logical_s8;
188 lai->primitive_type_vector [f_primitive_type_real]
189 = builtin->builtin_real;
190 lai->primitive_type_vector [f_primitive_type_real_s8]
191 = builtin->builtin_real_s8;
192 lai->primitive_type_vector [f_primitive_type_real_s16]
193 = builtin->builtin_real_s16;
194 lai->primitive_type_vector [f_primitive_type_complex_s8]
195 = builtin->builtin_complex_s8;
196 lai->primitive_type_vector [f_primitive_type_complex_s16]
197 = builtin->builtin_complex_s16;
198 lai->primitive_type_vector [f_primitive_type_void]
199 = builtin->builtin_void;
200
201 lai->bool_type_symbol = "logical";
202 lai->bool_type_default = builtin->builtin_logical_s2;
203 }
204
205 /* Remove the modules separator :: from the default break list. */
206
207 static const char *
208 f_word_break_characters (void)
209 {
210 static char *retval;
211
212 if (!retval)
213 {
214 char *s;
215
216 retval = xstrdup (default_word_break_characters ());
217 s = strchr (retval, ':');
218 if (s)
219 {
220 char *last_char = &s[strlen (s) - 1];
221
222 *s = *last_char;
223 *last_char = 0;
224 }
225 }
226 return retval;
227 }
228
229 /* Consider the modules separator :: as a valid symbol name character
230 class. */
231
232 static void
233 f_collect_symbol_completion_matches (completion_tracker &tracker,
234 complete_symbol_mode mode,
235 symbol_name_match_type compare_name,
236 const char *text, const char *word,
237 enum type_code code)
238 {
239 default_collect_symbol_completion_matches_break_on (tracker, mode,
240 compare_name,
241 text, word, ":", code);
242 }
243
244 /* Special expression evaluation cases for Fortran. */
245 struct value *
246 evaluate_subexp_f (struct type *expect_type, struct expression *exp,
247 int *pos, enum noside noside)
248 {
249 struct value *arg1 = NULL, *arg2 = NULL;
250 enum exp_opcode op;
251 int pc;
252 struct type *type;
253
254 pc = *pos;
255 *pos += 1;
256 op = exp->elts[pc].opcode;
257
258 switch (op)
259 {
260 default:
261 *pos -= 1;
262 return evaluate_subexp_standard (expect_type, exp, pos, noside);
263
264 case UNOP_ABS:
265 arg1 = evaluate_subexp (NULL_TYPE, exp, pos, noside);
266 if (noside == EVAL_SKIP)
267 return eval_skip_value (exp);
268 type = value_type (arg1);
269 switch (TYPE_CODE (type))
270 {
271 case TYPE_CODE_FLT:
272 {
273 double d
274 = fabs (target_float_to_host_double (value_contents (arg1),
275 value_type (arg1)));
276 return value_from_host_double (type, d);
277 }
278 case TYPE_CODE_INT:
279 {
280 LONGEST l = value_as_long (arg1);
281 l = llabs (l);
282 return value_from_longest (type, l);
283 }
284 }
285 error (_("ABS of type %s not supported"), TYPE_SAFE_NAME (type));
286
287 case BINOP_MOD:
288 arg1 = evaluate_subexp (NULL_TYPE, exp, pos, noside);
289 arg2 = evaluate_subexp (value_type (arg1), exp, pos, noside);
290 if (noside == EVAL_SKIP)
291 return eval_skip_value (exp);
292 type = value_type (arg1);
293 if (TYPE_CODE (type) != TYPE_CODE (value_type (arg2)))
294 error (_("non-matching types for parameters to MOD ()"));
295 switch (TYPE_CODE (type))
296 {
297 case TYPE_CODE_FLT:
298 {
299 double d1
300 = target_float_to_host_double (value_contents (arg1),
301 value_type (arg1));
302 double d2
303 = target_float_to_host_double (value_contents (arg2),
304 value_type (arg2));
305 double d3 = fmod (d1, d2);
306 return value_from_host_double (type, d3);
307 }
308 case TYPE_CODE_INT:
309 {
310 LONGEST v1 = value_as_long (arg1);
311 LONGEST v2 = value_as_long (arg2);
312 if (v2 == 0)
313 error (_("calling MOD (N, 0) is undefined"));
314 LONGEST v3 = v1 - (v1 / v2) * v2;
315 return value_from_longest (value_type (arg1), v3);
316 }
317 }
318 error (_("MOD of type %s not supported"), TYPE_SAFE_NAME (type));
319
320 case UNOP_FORTRAN_CEILING:
321 {
322 arg1 = evaluate_subexp (NULL_TYPE, exp, pos, noside);
323 if (noside == EVAL_SKIP)
324 return eval_skip_value (exp);
325 type = value_type (arg1);
326 if (TYPE_CODE (type) != TYPE_CODE_FLT)
327 error (_("argument to CEILING must be of type float"));
328 double val
329 = target_float_to_host_double (value_contents (arg1),
330 value_type (arg1));
331 val = ceil (val);
332 return value_from_host_double (type, val);
333 }
334
335 case UNOP_FORTRAN_FLOOR:
336 {
337 arg1 = evaluate_subexp (NULL_TYPE, exp, pos, noside);
338 if (noside == EVAL_SKIP)
339 return eval_skip_value (exp);
340 type = value_type (arg1);
341 if (TYPE_CODE (type) != TYPE_CODE_FLT)
342 error (_("argument to FLOOR must be of type float"));
343 double val
344 = target_float_to_host_double (value_contents (arg1),
345 value_type (arg1));
346 val = floor (val);
347 return value_from_host_double (type, val);
348 }
349
350 case BINOP_FORTRAN_MODULO:
351 {
352 arg1 = evaluate_subexp (NULL_TYPE, exp, pos, noside);
353 arg2 = evaluate_subexp (value_type (arg1), exp, pos, noside);
354 if (noside == EVAL_SKIP)
355 return eval_skip_value (exp);
356 type = value_type (arg1);
357 if (TYPE_CODE (type) != TYPE_CODE (value_type (arg2)))
358 error (_("non-matching types for parameters to MODULO ()"));
359 /* MODULO(A, P) = A - FLOOR (A / P) * P */
360 switch (TYPE_CODE (type))
361 {
362 case TYPE_CODE_INT:
363 {
364 LONGEST a = value_as_long (arg1);
365 LONGEST p = value_as_long (arg2);
366 LONGEST result = a - (a / p) * p;
367 if (result != 0 && (a < 0) != (p < 0))
368 result += p;
369 return value_from_longest (value_type (arg1), result);
370 }
371 case TYPE_CODE_FLT:
372 {
373 double a
374 = target_float_to_host_double (value_contents (arg1),
375 value_type (arg1));
376 double p
377 = target_float_to_host_double (value_contents (arg2),
378 value_type (arg2));
379 double result = fmod (a, p);
380 if (result != 0 && (a < 0.0) != (p < 0.0))
381 result += p;
382 return value_from_host_double (type, result);
383 }
384 }
385 error (_("MODULO of type %s not supported"), TYPE_SAFE_NAME (type));
386 }
387
388 case BINOP_FORTRAN_CMPLX:
389 arg1 = evaluate_subexp (NULL_TYPE, exp, pos, noside);
390 arg2 = evaluate_subexp (value_type (arg1), exp, pos, noside);
391 if (noside == EVAL_SKIP)
392 return eval_skip_value (exp);
393 type = builtin_f_type(exp->gdbarch)->builtin_complex_s16;
394 return value_literal_complex (arg1, arg2, type);
395
396 case UNOP_FORTRAN_KIND:
397 arg1 = evaluate_subexp (NULL, exp, pos, EVAL_AVOID_SIDE_EFFECTS);
398 type = value_type (arg1);
399
400 switch (TYPE_CODE (type))
401 {
402 case TYPE_CODE_STRUCT:
403 case TYPE_CODE_UNION:
404 case TYPE_CODE_MODULE:
405 case TYPE_CODE_FUNC:
406 error (_("argument to kind must be an intrinsic type"));
407 }
408
409 if (!TYPE_TARGET_TYPE (type))
410 return value_from_longest (builtin_type (exp->gdbarch)->builtin_int,
411 TYPE_LENGTH (type));
412 return value_from_longest (builtin_type (exp->gdbarch)->builtin_int,
413 TYPE_LENGTH (TYPE_TARGET_TYPE(type)));
414 }
415
416 /* Should be unreachable. */
417 return nullptr;
418 }
419
420 /* Return true if TYPE is a string. */
421
422 static bool
423 f_is_string_type_p (struct type *type)
424 {
425 type = check_typedef (type);
426 return (TYPE_CODE (type) == TYPE_CODE_STRING
427 || (TYPE_CODE (type) == TYPE_CODE_ARRAY
428 && TYPE_CODE (TYPE_TARGET_TYPE (type)) == TYPE_CODE_CHAR));
429 }
430
431 /* Special expression lengths for Fortran. */
432
433 static void
434 operator_length_f (const struct expression *exp, int pc, int *oplenp,
435 int *argsp)
436 {
437 int oplen = 1;
438 int args = 0;
439
440 switch (exp->elts[pc - 1].opcode)
441 {
442 default:
443 operator_length_standard (exp, pc, oplenp, argsp);
444 return;
445
446 case UNOP_FORTRAN_KIND:
447 case UNOP_FORTRAN_FLOOR:
448 case UNOP_FORTRAN_CEILING:
449 oplen = 1;
450 args = 1;
451 break;
452
453 case BINOP_FORTRAN_CMPLX:
454 case BINOP_FORTRAN_MODULO:
455 oplen = 1;
456 args = 2;
457 break;
458 }
459
460 *oplenp = oplen;
461 *argsp = args;
462 }
463
464 /* Helper for PRINT_SUBEXP_F. Arguments are as for PRINT_SUBEXP_F, except
465 the extra argument NAME which is the text that should be printed as the
466 name of this operation. */
467
468 static void
469 print_unop_subexp_f (struct expression *exp, int *pos,
470 struct ui_file *stream, enum precedence prec,
471 const char *name)
472 {
473 (*pos)++;
474 fprintf_filtered (stream, "%s(", name);
475 print_subexp (exp, pos, stream, PREC_SUFFIX);
476 fputs_filtered (")", stream);
477 }
478
479 /* Helper for PRINT_SUBEXP_F. Arguments are as for PRINT_SUBEXP_F, except
480 the extra argument NAME which is the text that should be printed as the
481 name of this operation. */
482
483 static void
484 print_binop_subexp_f (struct expression *exp, int *pos,
485 struct ui_file *stream, enum precedence prec,
486 const char *name)
487 {
488 (*pos)++;
489 fprintf_filtered (stream, "%s(", name);
490 print_subexp (exp, pos, stream, PREC_SUFFIX);
491 fputs_filtered (",", stream);
492 print_subexp (exp, pos, stream, PREC_SUFFIX);
493 fputs_filtered (")", stream);
494 }
495
496 /* Special expression printing for Fortran. */
497
498 static void
499 print_subexp_f (struct expression *exp, int *pos,
500 struct ui_file *stream, enum precedence prec)
501 {
502 int pc = *pos;
503 enum exp_opcode op = exp->elts[pc].opcode;
504
505 switch (op)
506 {
507 default:
508 print_subexp_standard (exp, pos, stream, prec);
509 return;
510
511 case UNOP_FORTRAN_KIND:
512 print_unop_subexp_f (exp, pos, stream, prec, "KIND");
513 return;
514
515 case UNOP_FORTRAN_FLOOR:
516 print_unop_subexp_f (exp, pos, stream, prec, "FLOOR");
517 return;
518
519 case UNOP_FORTRAN_CEILING:
520 print_unop_subexp_f (exp, pos, stream, prec, "CEILING");
521 return;
522
523 case BINOP_FORTRAN_CMPLX:
524 print_binop_subexp_f (exp, pos, stream, prec, "CMPLX");
525 return;
526
527 case BINOP_FORTRAN_MODULO:
528 print_binop_subexp_f (exp, pos, stream, prec, "MODULO");
529 return;
530 }
531 }
532
533 /* Special expression names for Fortran. */
534
535 static const char *
536 op_name_f (enum exp_opcode opcode)
537 {
538 switch (opcode)
539 {
540 default:
541 return op_name_standard (opcode);
542
543 #define OP(name) \
544 case name: \
545 return #name ;
546 #include "fortran-operator.def"
547 #undef OP
548 }
549 }
550
551 /* Special expression dumping for Fortran. */
552
553 static int
554 dump_subexp_body_f (struct expression *exp,
555 struct ui_file *stream, int elt)
556 {
557 int opcode = exp->elts[elt].opcode;
558 int oplen, nargs, i;
559
560 switch (opcode)
561 {
562 default:
563 return dump_subexp_body_standard (exp, stream, elt);
564
565 case UNOP_FORTRAN_KIND:
566 case UNOP_FORTRAN_FLOOR:
567 case UNOP_FORTRAN_CEILING:
568 case BINOP_FORTRAN_CMPLX:
569 case BINOP_FORTRAN_MODULO:
570 operator_length_f (exp, (elt + 1), &oplen, &nargs);
571 break;
572 }
573
574 elt += oplen;
575 for (i = 0; i < nargs; i += 1)
576 elt = dump_subexp (exp, stream, elt);
577
578 return elt;
579 }
580
581 /* Special expression checking for Fortran. */
582
583 static int
584 operator_check_f (struct expression *exp, int pos,
585 int (*objfile_func) (struct objfile *objfile,
586 void *data),
587 void *data)
588 {
589 const union exp_element *const elts = exp->elts;
590
591 switch (elts[pos].opcode)
592 {
593 case UNOP_FORTRAN_KIND:
594 case UNOP_FORTRAN_FLOOR:
595 case UNOP_FORTRAN_CEILING:
596 case BINOP_FORTRAN_CMPLX:
597 case BINOP_FORTRAN_MODULO:
598 /* Any references to objfiles are held in the arguments to this
599 expression, not within the expression itself, so no additional
600 checking is required here, the outer expression iteration code
601 will take care of checking each argument. */
602 break;
603
604 default:
605 return operator_check_standard (exp, pos, objfile_func, data);
606 }
607
608 return 0;
609 }
610
611 static const char *f_extensions[] =
612 {
613 ".f", ".F", ".for", ".FOR", ".ftn", ".FTN", ".fpp", ".FPP",
614 ".f90", ".F90", ".f95", ".F95", ".f03", ".F03", ".f08", ".F08",
615 NULL
616 };
617
618 /* Expression processing for Fortran. */
619 static const struct exp_descriptor exp_descriptor_f =
620 {
621 print_subexp_f,
622 operator_length_f,
623 operator_check_f,
624 op_name_f,
625 dump_subexp_body_f,
626 evaluate_subexp_f
627 };
628
629 extern const struct language_defn f_language_defn =
630 {
631 "fortran",
632 "Fortran",
633 language_fortran,
634 range_check_on,
635 case_sensitive_off,
636 array_column_major,
637 macro_expansion_no,
638 f_extensions,
639 &exp_descriptor_f,
640 f_parse, /* parser */
641 null_post_parser,
642 f_printchar, /* Print character constant */
643 f_printstr, /* function to print string constant */
644 f_emit_char, /* Function to print a single character */
645 f_print_type, /* Print a type using appropriate syntax */
646 default_print_typedef, /* Print a typedef using appropriate syntax */
647 f_val_print, /* Print a value using appropriate syntax */
648 c_value_print, /* FIXME */
649 default_read_var_value, /* la_read_var_value */
650 NULL, /* Language specific skip_trampoline */
651 NULL, /* name_of_this */
652 false, /* la_store_sym_names_in_linkage_form_p */
653 cp_lookup_symbol_nonlocal, /* lookup_symbol_nonlocal */
654 basic_lookup_transparent_type,/* lookup_transparent_type */
655
656 /* We could support demangling here to provide module namespaces
657 also for inferiors with only minimal symbol table (ELF symbols).
658 Just the mangling standard is not standardized across compilers
659 and there is no DW_AT_producer available for inferiors with only
660 the ELF symbols to check the mangling kind. */
661 NULL, /* Language specific symbol demangler */
662 NULL,
663 NULL, /* Language specific
664 class_name_from_physname */
665 f_op_print_tab, /* expression operators for printing */
666 0, /* arrays are first-class (not c-style) */
667 1, /* String lower bound */
668 f_word_break_characters,
669 f_collect_symbol_completion_matches,
670 f_language_arch_info,
671 default_print_array_index,
672 default_pass_by_reference,
673 default_get_string,
674 c_watch_location_expression,
675 NULL, /* la_get_symbol_name_matcher */
676 iterate_over_symbols,
677 default_search_name_hash,
678 &default_varobj_ops,
679 NULL,
680 NULL,
681 f_is_string_type_p,
682 "(...)" /* la_struct_too_deep_ellipsis */
683 };
684
685 static void *
686 build_fortran_types (struct gdbarch *gdbarch)
687 {
688 struct builtin_f_type *builtin_f_type
689 = GDBARCH_OBSTACK_ZALLOC (gdbarch, struct builtin_f_type);
690
691 builtin_f_type->builtin_void
692 = arch_type (gdbarch, TYPE_CODE_VOID, TARGET_CHAR_BIT, "void");
693
694 builtin_f_type->builtin_character
695 = arch_type (gdbarch, TYPE_CODE_CHAR, TARGET_CHAR_BIT, "character");
696
697 builtin_f_type->builtin_logical_s1
698 = arch_boolean_type (gdbarch, TARGET_CHAR_BIT, 1, "logical*1");
699
700 builtin_f_type->builtin_integer_s2
701 = arch_integer_type (gdbarch, gdbarch_short_bit (gdbarch), 0,
702 "integer*2");
703
704 builtin_f_type->builtin_integer_s8
705 = arch_integer_type (gdbarch, gdbarch_long_long_bit (gdbarch), 0,
706 "integer*8");
707
708 builtin_f_type->builtin_logical_s2
709 = arch_boolean_type (gdbarch, gdbarch_short_bit (gdbarch), 1,
710 "logical*2");
711
712 builtin_f_type->builtin_logical_s8
713 = arch_boolean_type (gdbarch, gdbarch_long_long_bit (gdbarch), 1,
714 "logical*8");
715
716 builtin_f_type->builtin_integer
717 = arch_integer_type (gdbarch, gdbarch_int_bit (gdbarch), 0,
718 "integer");
719
720 builtin_f_type->builtin_logical
721 = arch_boolean_type (gdbarch, gdbarch_int_bit (gdbarch), 1,
722 "logical*4");
723
724 builtin_f_type->builtin_real
725 = arch_float_type (gdbarch, gdbarch_float_bit (gdbarch),
726 "real", gdbarch_float_format (gdbarch));
727 builtin_f_type->builtin_real_s8
728 = arch_float_type (gdbarch, gdbarch_double_bit (gdbarch),
729 "real*8", gdbarch_double_format (gdbarch));
730 builtin_f_type->builtin_real_s16
731 = arch_float_type (gdbarch, gdbarch_long_double_bit (gdbarch),
732 "real*16", gdbarch_long_double_format (gdbarch));
733
734 builtin_f_type->builtin_complex_s8
735 = arch_complex_type (gdbarch, "complex*8",
736 builtin_f_type->builtin_real);
737 builtin_f_type->builtin_complex_s16
738 = arch_complex_type (gdbarch, "complex*16",
739 builtin_f_type->builtin_real_s8);
740 builtin_f_type->builtin_complex_s32
741 = arch_complex_type (gdbarch, "complex*32",
742 builtin_f_type->builtin_real_s16);
743
744 return builtin_f_type;
745 }
746
747 static struct gdbarch_data *f_type_data;
748
749 const struct builtin_f_type *
750 builtin_f_type (struct gdbarch *gdbarch)
751 {
752 return (const struct builtin_f_type *) gdbarch_data (gdbarch, f_type_data);
753 }
754
755 void
756 _initialize_f_language (void)
757 {
758 f_type_data = gdbarch_data_register_post_init (build_fortran_types);
759 }
760
761 /* See f-lang.h. */
762
763 struct value *
764 fortran_argument_convert (struct value *value, bool is_artificial)
765 {
766 if (!is_artificial)
767 {
768 /* If the value is not in the inferior e.g. registers values,
769 convenience variables and user input. */
770 if (VALUE_LVAL (value) != lval_memory)
771 {
772 struct type *type = value_type (value);
773 const int length = TYPE_LENGTH (type);
774 const CORE_ADDR addr
775 = value_as_long (value_allocate_space_in_inferior (length));
776 write_memory (addr, value_contents (value), length);
777 struct value *val
778 = value_from_contents_and_address (type, value_contents (value),
779 addr);
780 return value_addr (val);
781 }
782 else
783 return value_addr (value); /* Program variables, e.g. arrays. */
784 }
785 return value;
786 }
787
788 /* See f-lang.h. */
789
790 struct type *
791 fortran_preserve_arg_pointer (struct value *arg, struct type *type)
792 {
793 if (TYPE_CODE (value_type (arg)) == TYPE_CODE_PTR)
794 return value_type (arg);
795 return type;
796 }