3 * config/tc-mips.c (insn_uses_reg): Map register numbers in mips16
6 * cond.c (cond_finish_check): New function.
7 * as.h (cond_finish_check): Declare.
8 * as.c (main): Call cond_finish_check.
13 * config/tc-d30v.c (md_assemble): If two instructions
14 are supposed to be assembled in parallel and the first one is
15 long, print an error and stop.
16 (md_apply_fix3): Don't calculate absolute relocs. Just write
22 * config/tc-i386.h (iclrKludge): Define.
23 * config/tc-i386.c (md_assemble): Handle iclrKludge.
25 * config/tc-alpha.h (tc_frob_file_before_adjust): Define if
27 (alpha_frob_file_before_adjust): Declare if OBJ_ECOFF.
28 * config/tc-alpha.c (alpha_debug): New static variable.
29 (md_parse_option): Set alpha_debug if -g is seen.
30 (alpha_frob_file_before_adjust): New function if OBJ_ECOFF.
35 * config/tc-d30v.c (build_insn): Enable range-checking code.
36 (postfix): Stop at space or comma. Fixes PR 11862.
37 (md_assemble): Change error message.
42 * Makefile.in: Added automatic dependency building.
43 * dep-in.sed: New file.
47 * config/obj-ieee.c (segment_name): Don't define function if this
50 * config/obj-coff.h (DO_STRIP): Don't define.
51 * config/tc-h8300.h (DO_STRIP): Don't define.
52 * config/tc-h8500.h (DO_STRIP): Don't define.
53 * config/tc-w65.h (DO_STRIP): Don't define.
54 * config/tc-z8k.h (DO_STRIP): Don't define.
56 * symbols.c (colon): Call obj_frob_label if it is defined.
57 * config/obj-vms.h (obj_frob_label): Rename from tc_frob_label.
59 * configure.in: Don't set files and links. Don't call
60 AC_LINK_FILES. Substitute te_file. Create targ-cpu.h,
61 obj-format.h, targ-env.h, and itbl-cpu.h in AC_OUTPUT.
63 * Makefile.in (TARG_CPU_C): New variable.
64 (TARG_CPU_O, TARG_CPU_H): New variables.
65 (OBJ_FORMAT_C, OBJ_FORMAT_O, OBJ_FORMAT_H): New variables.
66 (TARG_ENV_H, ATOF_TARG_C, ATOF_TARG_O): New variables.
67 (SOURCES): Rename from REAL_SOURCES. Delete old definition.
68 (LINKED_SOURCES): Remove.
69 (HEADERS): Rename from REAL_HEADERS. Delete old definition.
70 (LINKED_HEADERS): Remove.
71 (OBJS): Use $(TARG_CPU_O), etc., rather than targ-cpu.o, etc.
72 ($(OBJS)): Depend upon $(TARG_ENV_H), etc., rather than
74 ($(TARG_CPU_O), $(OBJ_FORMAT_O) $(ATOF_TARG_O)): New targets.
75 (targ-cpu.o, obj-format.o, atof-targ.o): Remove targets.
76 (itbl-cpu.h): Remove target.
77 (DISTCLEAN_HERE): Remove targ-cpu.c, obj-format.c, atof-targ.c,
82 * doc/as.texinfo (Symbol Names): Don't use obsolete @ctrl macro.
86 * config/tc-m68k.c (mri_chip): Replace calls to get_symbol_end by
87 open coded loop that does not require the name to start with a
92 * frags.c (frag_var): Change offset parameter to offsetT.
93 (frag_variant): Likewise.
94 * frags.h (frag_variant, frag_var): Update declarations.
95 * config/tc-m68k.c (struct m68k_it): Change foff field to
97 (add_frag): Change off parameter to offsetT.
98 * Several files: Add casts to calls to frag_var.
100 * Makefile.in (m68k-parse.c): Depend upon itbl-parse.c, to
101 serialize a parallel make.
102 (itbl-parse.h): Split target out from itbl-parse.c.
106 * config/m68k-parse.y (motorola_operand): Allow (zdireg,EXPR).
108 * config/te-delta.h (COFF_COMMON_ADDEND): Define.
109 * config/obj-coff.c (fixup_segment): Check COFF_COMMON_ADDEND when
110 storing the value of a common symbol.
114 * config/obj-coff.c (glue_symbols): Unused variable symbolP
116 (crawl_symbols): Do not modify symbol_rootP and symbol_lastP here;
117 that is done by symbol_remove and symbol_insert.
119 * config/obj-coff.h (S_IS_LOCAL): Return 0 for a debugging
124 * config/tc-mips.c (load_register): In 32 bit mode, when not
125 dealing with a 64 bit number, permit the upper 32 bits to be set
126 even if bit 31 is not set.
130 * read.c (potable): Add "equiv".
131 (s_set): Handle .equiv based on argument.
132 * doc/as.texinfo (Equiv): New node to document .equiv.
133 (Err): New node to document .err.
137 * Many files: Add function prototypes.
138 * as.c (show_usage, parse_args): Make static.
139 * frags.h (frag_alloc): Declare.
140 * subsegs.c (subseg_set_rest): Don't declare frag_alloc.
141 * symbols.c (dollar_label_instance): Change return type to long.
142 * symbols.h (print_symbol_value): Declare.
143 (print_expr, print_expr_1, print_symbol_value_1): Declare.
144 * write.c (fix_new_exp): Don't declare make_expr_symbol.
145 (remove_subsegs, relax_frag): Make static.
146 * config/atof-vax.c (atof_vax_sizeof): Change letter to int.
147 (what_kind_of_float): Likewise.
148 (atof_vax): Make static. Change what_kind to int.
149 (md_atof): Change what_statement_type to int.
150 * config/obj-ecoff.h (obj_ecoff_set_ext): Declare.
151 * config/tc-alpha.c (vax_md_atof): Declare.
152 (md_atof): Don't declare atof_ieee and vax_md_atof.
153 * config/tc-i386.c (set_16bit_code_flag): Make static.
154 * config/tc-i386.h (tc_i386_fix_adjustable): Declare.
155 * config/tc-m68k.c (add_fix): Change width to int.
156 (insert_reg): Change regname to const.
157 (md_atof): Don't declare atof_ieee.
158 (demand_empty_rest_of_line): Don't declare.
159 * config/tc-m88k.c (md_atof): Don't declare atof_ieee.
160 * config/tc-sparc.c (cmp_reg_entry): Change args to const PTR.
161 (parse_keyword_arg): Change lookup_fn to take const arg.
162 (md_atof): Don't declare atof_ieee.
163 * config/tc-sparc.h: Add ifdef for multiple inclusion.
164 (tc_aout_pre_write_hook): Don't declare.
168 * as.h (bfd_alloc_by_size_t): Don't declare.
169 * Many files: Use xmalloc rather than bfd_alloc_by_size_t.
173 * symbols.c (symbol_new): Don't call debug_verify_symchain.
174 (symbol_append): Set sy_next and sy_previous when adding a single
175 symbol to an empty list. Call debug_verify_symchain.
176 (verify_symbol_chain): Use assert, not know.
180 * NEWS: Note BeOS support.
181 * configure.in: (ppc-*-beos): New target, use coff as object format.
182 * configure: Regenerate with autoconf.
186 * config/tc-mips.c (md_apply_fix): Improve error message for out
189 * Makefile.in: Add dependencies on obstack.h where needed.
193 * config/tc-mips.c (md_estimate_size_before_relax): Handle the
194 case of a symbol equated to another symbol when using SVR4_PIC.
196 * Makefile.in (TARG_CPU_DEP_sparc): Add opcode/sparc.h.
200 * read.c (read_a_source_file): Call LISTING_NEWLINE before
201 HANDLE_CONDITIONAL_ASSEMBLY when handling an MRI line label.
203 * config/obj-elf.c (obj_elf_data): Call md_flush_pending_output
204 and md_elf_section_change_hook if they are defined.
205 (obj_elf_text, obj_elf_previous): Likewise.
209 * config/obj-multi.h (struct elf_obj_sy): Define if
211 (OBJ_SYMFIELD_TYPE): Define as struct elf_obj_sy if
213 * config/obj-elf.h (struct elf_obj_sy): Don't define if
214 OBJ_SYMFIELD_TYPE is defined.
216 * doc/as.texinfo (bss): Improve description of .bss section. In
217 ELF or COFF, you are permitted to switch into the section.
218 (Comm): Rewrite description of common symbols.
219 (Lcomm): Mention that some targets permit a third argument.
223 * config/tc-ppc.c (ppc_elf_lcomm): Don't call S_CLEAR_EXTERNAL.
225 * symbols.c (colon): Change type of local to int. From Alan Modra
228 * config/tc-m88k.c (m88k_do_align): Don't use a special nop
229 alignment if a zero fill pattern was explicitly specified.
230 * config/tc-sh.c (sh_do_align): Likewise.
232 * read.c (equals): Always permit register names to be redefined.
234 * config/tc-mips.c (mips_fix_adjustable): Permit a reloc against a
235 mips16 symbol to be adjusted if a symbol is being subtracted from
239 * config/obj-elf.c (obj_elf_symver): Check for duplicate or
240 illegal symbol version names.
241 (elf_frob_symbol): Check for external default versions.
246 * config/obj-elf.h (struct elf_obj_sy): Define.
247 (OBJ_SYMFIELD_TYPE): Define to elf_obj_sy struct. Change all
249 * config/obj-elf.c (obj_elf_symver): Just record the name.
250 (obj_symbol_new_hook): Initialized versioned_name field.
251 (elf_frob_symbol): If there is a versioned_name, either rename the
252 symbol, or add an alias with that name.
256 * config/tc-mn10300.c (md_relax_table): Define.
257 (md_convert_frag): Implement.
258 (md_assemble): Handle relaxable operands/instructions correctly.
259 (md_estimate_size_before_relax): Implement.
260 * config/tc-mn10300.h (TC_GENERIC_RELAX_TABLE): Define.
262 * config/tc-mn10200.c (md_relax_table): Fix typos.
264 * config/tc-mn10300.c (md_assemble): Don't use any MN10300 specific
265 relocs anymore. Tweak fx_offset for pc-relative relocs.
269 * cond.c (s_ifc): Call mri_comment_field and mri_comment_end when
275 * config/tc-tic80.c (md_pseudo_table): Add "sect" and "section"
277 * config/tc-tic80.c (md_begin): Declare external variable
278 coff_flags and insert an F_AR32WR bit into it.
283 * read.c (equals): Add reassign parameter. Change all callers.
284 * read.h (equals): Update declaration.
288 * config/tc-mips.c (mips16_extended_frag): Don't assume that we
289 can rely on the frag address to determine whether a frag is
294 * write.h (LOCAL_LABEL): Only define if not BFD_ASSEMBLER.
295 (S_LOCAL_NAME): Likewise.
296 (FAKE_LABEL_NAME): Define unconditionally.
297 * symbols.c (colon): Call bfd_is_local_label, not LOCAL_LABEL, if
299 (S_IS_LOCAL): Call bfd_is_local_label_name, not LOCAL_LABEL.
300 * config/tc-*.h: Only define LOCAL_LABEL if not BFD_ASSEMBLER.
301 Don't define FAKE_LABEL_NAME.
302 * config/te-ic960.h: Likewise.
303 * config/tc-mips.h (tc_frob_file_before_adjust): Define.
304 (mips_frob_file_before_adjust): Declare.
305 * config/tc-mips.c (mips_frob_file_before_adjust): New function.
306 (mips_local_label): Remove.
308 * config/te-sco386.h: Remove; not used.
313 * config/tc-tic80 (md_pseudo_table): Add align pseudo op to do
314 byte alignment rather than power-of-two alignment that is the
320 * config/tc-m68k.c (md_assemble): Handle a reloc width of 'W'.
322 * gasp.c (hash_add_to_string_table): Correct misspelling in error
323 message, and add newline.
324 (process_file): Don't process assignments in the label if this is
325 a equ or assign pseudo-op.
326 (process_pseudo_op): Swap first argument to do_assign for K_ASSIGN
327 and K_EQU, to match documentation.
331 * config/obj-coff.c (obj_coff_section): Add 'r' section attribute
332 to denote read-only data sections.
336 * config/obj-elf.c (obj_elf_common): Set BSF_OBJECT in flags.
337 * config/tc-sparc.c (s_common): Likewise, if BFD_ASSEMBLER.
339 * expr.c (operand): Simplify 0b handling. Don't treat 0b as a
340 binary number if the next character is '+' or '-'.
345 * config/tc-tic80.c (FLT_CHARS): Change from "dD" to "fF".
346 (find_opcode): Match operands that can be floats.
347 (build_insn): Handle O_big (float) expressions and build
353 * configure.in (mips*-*-lnews*): New target, also make empty
354 emulation list for this target.
356 * tc-mips.c (ECOFF_LITTLE_FORMAT): Define.
357 (mips_target_format): Use.
358 * te-lnews.h: New file.
363 * config/tc-tic80.c (find_opcode, build_insn): Changes to match
364 operands with :m or :s modifiers and generate the right opcodes
370 * Makefile.in (itbl-parse.c itbl-parse.h): Use $(BISON) and
371 $(BISONFLAGS), not $(YACC) and $(YACCFLAGS).
375 * config/tc-m68k.c (instring): Useless local declaration of
376 crack_operand removed.
377 * expr.h (expressionS): Changed type of X_op field to operatorT if
383 * configure.in: Add i386coff and i386elf to emulation list.
384 * configure: Rebuild.
385 * as.c (i386coff, i386elf): Declare.
386 * obj.h (coff_format_ops): Declare.
387 * config/obj-coff.c (OBJ_HEADER): Define.
388 (coff_obj_symbol_new_hook): Rename from obj_symbol_new_hook.
389 (coff_obj_read_begin_hook): Rename from obj_read_begin_hook.
390 (obj_pseudo_table): Add "version".
391 (coff_pop_insert): New static function.
392 (coff_sec_sym_ok_for_reloc): New static function.
393 (no_func): New static function.
394 (coff_format_ops): New variable.
395 * config/obj-coff.h (coff_obj_symbol_new_hook): Declare.
396 (obj_symbol_new_hook): Define.
397 (coff_obj_read_begin_hook): Declare.
398 (obj_read_begin_hook): Define.
399 * config/tc-i386.h (i386_target_format): Declare.
400 * config/tc-i386.c: Check OBJ_MAYBE_ELF as well as OBJ_ELF; check
401 OUTPUT_FLAVOR when appropriate.
402 (i386_target_format): New function.
403 * Makefile.in (obj-coff.o): New target.
404 (e-i386coff.o, e-i386elf.o): New targets.
407 * config/tc-i960.h (TC_SYMFIELD_TYPE): Define if OBJ_COFF.
408 (_tc_get_bal_of_call): Don't declare.
409 (tc_get_bal_of_call): Declare as function, don't define as macro.
410 * config/tc-i960.c (tc_set_bal_of_call): If OBJ_COFF, store balP
411 in sy_tc field, not x_balntry field.
412 (tc_get_bal_of_call): Rename from _tc_get_bal_of_call. Change
413 return type to symbolS *. If OBJ_COFF, retrieve value from sy_tc
414 field, not x_balntry field.
416 * config/obj-elf.c (obj_elf_section): Permit a .note section to
417 have the SHF_ALLOC attribute.
419 * Makefile.in ($(OBJS)): Don't depend upon $(IT_HDRS).
420 (TARG_CPU_DEP_mips): Depend upon $(srcdir)/itbl-ops.h.
421 (itbl-lex.o): Depend upon itbl-parse.h.
423 * itbl-parse.y (yyerror): Change return type to int. Change to
424 use old style function declaration.
426 * Makefile.in (itbl-lex.o): Remove -Wall.
427 (itbl-parse.o): Likewise.
429 * cond.c (s_ifdef): If we should omit conditionals from listings,
431 (s_if, s_ifc, s_endif, s_else, s_ifeqs): Likewise.
432 * listing.c (list_info_struct): Add EDICT_NOLIST_NEXT.
433 (listing_listing): Handle EDICT_NOLIST_NEXT.
434 (listing_list): An argument of 2 means EDICT_NOLIST_NEXT.
435 * listing.h (LISTING_NOCOND): Define.
436 (LISTING_SKIP_COND): Define.
437 * as.c (show_usage): Mention c as a suboption of -a.
438 (parse_args): Handle c as a suboption of -a.
439 * doc/as.texinfo: Document -alc.
444 * config/tc-tic80.c (md_apply_fix): Handle R_ABS type fixups.
449 * doc/as.texinfo: Document .symver.
453 * config/tc-d10v.c: Change pre_defined_registers to
454 d10v_predefined_registers and reg_name_cnt to d10v_reg_name_cnt.
458 * config/obj-coff.c: Fix typo in comment section.
460 * config/tc-tic80.c (md_pseudo_table): Add entry for bss, which takes
461 an additional alignment argument.
462 (find_opcode): Allow O_symbol relocs for any 32 bit field, not just
464 (build_insn): Handle O_symbol relocs for any 32 bit field, not just
470 * Makefile.in: Remove dependancies on itbl-cpu.h.
471 * as.c: Define stubs for itbl_parse and itbl_init if HAVE_ITBL_CPU
476 * itbl-ops.h: Include as.h.
480 * as.c: Remove -t option.
481 * configure, configure.in: Move itbl-cpu.h to mips specific configure.
482 * itbl-ops.h: Include itbl-cpu.h only if HAVE_ITBL_CPU is defined.
483 * config/tc-mips.h: Define HAVE_ITBL_CPU.
487 * itbl-ops.c: Don't define DEBUG.
491 * Makefile.in: Update itbl-test.c to reflect its new location.
495 * itbl-ops.c: Add test for itbl_have_entries.
496 * config/tc-mips.c: Remove test for itbl_have_entries.
497 * config/tc-mips.h: Define tc_init_after_args to mips_init_after_args.
501 * Makefile.in (DISTSTUFF): Remove itbl-parse.y, itbl-lex.l, and
502 itbl-ops.c. Add itbl-parse.c and itbl-lex.c.
503 (LEX, LEXFLAGS): Define.
504 * itbl-ops.c (append_insns_as_macros): Remove bogus ASSERT.
508 * itbl-parse.y: Fix indentation mistakes from indent program.
509 * itbl-lex.l: Fix indentation mistakes from indent program.
510 * itbl-ops.h: Add include for ansidecl.h.
511 Add PARAMS around function arguments.
512 Add declaration for itbl_have_entries.
513 * itbl-ops.c: Add PARAMS around function arguments.
514 * Makefile.in: Add itbl build rules.
515 Add dependancies for itbl files to mips target.
516 * as.c: Add itbl support.
517 Add new option "--insttbl" for dynamically extending instruction set.
518 * as.h: Declare insttbl_file_name;
519 the name of file defining extensions to the basic instruction set
520 * configure.in, configure: Add itbl-parse.o, itbl-lex.o, and
521 itbl-ops.o to extra_objects for mips configuration.
522 Add include file link from itbl-cpu.h to
523 config/itbl-${target_cpu_type}.h.
524 * config/tc-mips.c: Allow copz instructions.
525 Add notes for future additions to the itbl support.
527 (macro): Call itbl_assemble to assemble itbl instructions.
528 See if an unknown register is specified in an itbl entry.
531 * doc/internals.texi (CPU backend): Fix typo in md_section_align
534 * config/tc-tic80.h (NEED_FX_R_TYPE): Define.
535 * config/tc-tic80.c (find_opcode): Add code to support O_symbol
537 (build_insn): Grab a frag early so we can use the address in
538 fixups. Take one's complement of BITNUM values before insertion
539 in opcode. Add code to support O_symbol operands.
540 (md_apply_fix): Replace unimplemented warning with implementation.
541 (md_pcrel_from): Ditto.
542 (tc_coff_fix2rtype): Ditto.
547 * config/tc-d30v.c (parallel_ok): New function.
548 * config/tc-d30v.h: Define TARGET_BYTES_BIG_ENDIAN.
550 * config/tc-d10v.c (md_pcrel_from_section): Return 0 if
551 relocation is in different section. Fixes PR11574.
555 * tc-mips.c (mips_ip): If configured for an embedded ELF system,
556 don't set the section alignment to 2**4.
560 * config/tc-m68k.c (line_comment_chars): Add '*'.
562 * app.c (LEX_IS_TWOCHAR_COMMENT_2ND): Don't define.
563 (do_scrub_begin): Don't set lex['*'].
564 (do_scrub_chars): When handling LEX_IS_TWOCHAR_COMMENT_1ST, don't
565 check for LEX_IS_TWOCHAR_COMMENT_2ND. Instead, just check for
568 * configure.in: Set em=svr4 for m68k-*-sysv4*.
569 * configure: Rebuild.
570 * config/te-svr4.h: New file.
571 * config/tc-m68k.c (m68k_comment_chars): Only include `#' if
576 * config/tc-mn10200.c (md_convert_frag): Create a fixup for the
577 short conditional branch around a long unconditional branch.
581 * config/obj-coff.c (obj_coff_ln [both versions]): Call
584 * config/tc-arm.c (fix_new_arm): Use make_expr_symbol to handle a
587 * symbols.c (resolve_symbol_value): If both left and right
588 operands are undefined, warn about both of them.
593 * config/obj-elf.c (elf_pseudo_table): Add "symver".
594 (obj_elf_symver): New static function.
595 * config/obj-elf.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Copy the st_other
598 * write.c (relax_segment): Make type and printf format agree.
600 * read.c (get_line_sb): Don't end the line on a semicolon inside a
606 * config/tc-d30v.c, config/tc-d30v.h: New files.
608 * configure: Rebuilt.
610 * configure.in: Add case for d30v.
615 * config/tc-alpha.h (md_operand): Define with a null expansion,
616 like all the other targets.
617 * doc/internals.texi (CPU backend): Add missing word in
618 md_flush_pending_output description. Fix typo in md_convert_frag
621 * config/tc-tic80: Minor comment additions/changes.
626 * config/tc-m68k.c (LOCAL_LABEL): Macro redefined if TE_DELTA.
627 (tc_canonicalize_symbol_name): Macro defined if TE_DELTA.
628 * config/obj-coff.c (obj_coff_def): Use
629 tc_canonicalize_symbol_name if defined.
630 (obj_coff_tag, obj_coff_val): Likewise.
631 * expr.c (operand): Reject '~' as operator if is_name_beginner.
636 does not actually work, though:
637 * configure.in (i386-sequent-bsd*): New target.
638 * configure: Rebuild.
639 * config/tc-dynix.h: New file.
640 * config/tc-i386.h: Define TARGET_FORMAT if TE_DYNIX.
642 * read.c (do_align): Add max parameter. Change all callers.
643 Remove useless static variables.
644 (s_align): New static function. Do common portion of
645 s_align_bytes and s_align_ptwo.
646 (s_align_bytes, s_align_ptwo): Just call s_align.
647 * frags.c (frag_align): Add max parameter. Change all callers.
648 (frag_align_pattern): Likewise.
649 * frags.h (frag_align, frag_align_pattern): Update declarations.
650 * write.c (relax_segment): Limit alignment change to fr_subtype.
651 Fix some types to be addressT.
652 * config/obj-coff.c (size_section): Likewise.
653 * config/obj-ieee.c (size_section): Likewise.
654 * config/tc-d10v.h (md_do_align): Add max parameter.
655 * config/tc-i386.h (md_do_align): Likewise.
656 * config/tc-m88k.h (md_do_align): Likewise.
657 * config/tc-m88k.c (m88k_do_align): Likewise.
658 * config/tc-sh.h (md_do_align): Likewise.
659 * config/tc-sh.c (sh_do_align): Likewise.
660 * as.h: Improve comments on rs_align and rs_align_code.
661 * doc/as.texinfo: Document new alignment arguments.
662 * doc/internals.texi (Frags): Document use of fr_subtype field for
663 rs_align and rs_align_code.
667 * config/tc-mips.c: Changed opcode parsing.
671 * config/{tc-alpha.h, tc-arc.h, tc-d10v.h, tc-generic.h, tc-i960.h,
672 tc-mn10200.h, tc-mn10300.h, tc-sh.h, tc-v850.h, tc-vax.h, tc-w65.h}:
673 Add default definition of zero for TARGET_BYTES_BIG_ENDIAN.
674 * config/{tc-arm.h, tc-hppa.h, tc-i386.h, tc-mips.h, tc-ns32k.h,
675 tc-ppc.h, tc-sparc.h}: Move definition of TARGET_BYTES_BIG_ENDIAN
676 to a location consistent with the rest of the target include files.
677 * config/tc-i386.c: Remove misleading comment.
678 * doc/internals.texi (CPU backend): Add description of function
681 * config/tc-tic80.c: Add code to insert predefined symbols into the
682 symbol table so they can be parsed by the standard expression parser.
683 Remove custom code that use to parse them.
684 * config/tc-tic80.h: Move definition of TARGET_BYTES_BIG_ENDIAN
685 to a location consistent with the rest of the target include files.
690 * as.h: GNU c provides unlink() function.
692 Unify section handling on openVMS/Alpha:
693 * config/tc-alpha.c(s_alpha_link): Remove.
694 (s_alpha_section): New function.
695 Remove case-hacking of symbols
696 Add .code_address pseudo-op.
697 (BFD_RELOC_ALPHA_CODEADDR): New relocation.
698 (s_alpha_code_address): New function.
699 (alpha_ctors_section, alpha_dtors_section): New sections for C++
700 static constructors/destructors.
701 Add debug code for crash debugs, to be removed when traceback code
702 is added to object code.
703 (s_alpha_name): New function for .name pseudo-op.
704 (alpha_print_token): New function to print token expressions with
705 alpha specific extensions.
707 * makefile.vms: Allow compilation with current gcc snapshot.
711 * doc/Makefile.in (TEXI2DVI): Set to just name of program.
712 (DVIPS): Set to dvips.
713 (ps, as.ps, gasp.ps): New targets.
714 (internals.info, gasp.dvi, internals.dvi): Set both TEXINPUTS
715 and MAKEINFO env variables.
716 (internals.ps): Use DVIPS macro.
717 (clean): Remove core and backup files.
718 (distclean): Remove temporary files from building internals.
720 * doc/internals.texi (Frags): Fix typo.
721 (GAS processing): Ditto.
722 (CPU backend): Ditto.
723 * ecoff.c (init_file): Use TARGET_BYTES_BIG_ENDIAN value directly.
724 * mpw-config.in: Define TARGET_BYTES_BIG_ENDIAN as 1.
725 * read.c: Remove ugly hack that dealt with config files not
726 correctly defining TARGET_BYTES_BIG_ENDIAN.
727 (target_big_endian): Use TARGET_BYTES_BIG_ENDIAN directly.
728 * config/arm-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
729 * config/arm-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
730 * config/mips-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
731 * config/mips-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
732 * config/ppc-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
733 * config/ppc-sol.mt: Replace TARGET_BYTES_LITTLE_ENDIAN
734 with TARGET_BYTES_BIG_ENDIAN defined to 0.
735 * config/tc-arm.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
736 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
737 whether it is defined or not.
738 * config/tc-mips.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN.
739 * config/tc-ppc.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
740 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
741 whether it is defined or not.
743 * config/tic80.h (TARGET_FORMAT): Define to coff-tic80.
744 (TARGET_BYTES_BIG_ENDIAN): Define to 0.
749 * write.c (write_relocs): Correct text in as_fatal error message,
750 bfd_perform_relocation -> bfd_install_relocation.
754 * config/tc-m68k.c (LEX_TILDE): Define if TE_DELTA.
755 * read.c (LEX_TILDE): Define if not defined.
756 (lex_type): Use LEX_TILDE.
757 * expr.c (get_symbol_end): Check first char with is_name_beginner,
762 * config/tc-sparc.c (md_show_usage): Add missing backslash at end
765 * config/tc-mips.c (mips16_extended_frag): Correct base address
766 for an extended PC relative instruction.
767 (md_convert_frag): Likewise.
769 * config/tc-mips.c (prev_nop_frag): New static variable.
770 (prev_nop_frag_holds): New static variable.
771 (prev_nop_frag_required): New static variable.
772 (prev_nop_frag_since): New static variable.
773 (append_insn): If we aren't reordering, and prev_nop_frag is not
774 NULL, and we don't need any nops, then decrease the size of
775 prev_nop_frag. Don't insert nops because of instructions in
776 noreorder sections. Remember whether the previous instructions
777 where in noreorder sections even when not reordering.
778 (mips_no_prev_insn): Add preserver parameter. Change all
779 callers. Refer prev_nop_frag variables when appropriate.
780 (mips_emit_delays): Set up prev_nop_frag.
781 (s_mipsset): Clear prev_nop_frag if reordering.
785 * config/tc-mips.c (append_insn): Remove useless code which
786 handled swapping a mips16 jump with a mips16 instruction with a
789 * config/tc-mips.c (md_parse_option): When debugging, set
790 mips_optimize to 1, not 0.
792 * config/tc-mips.c (mips16_ip): Handle an extend operand.
794 * config/tc-mips.c (my_getExpression): In mips16 mode, if it looks
795 like the expression was based on `.', adjust the value of the
798 * config/tc-mips.c (append_insn): Warn about an attempt to put an
799 extended instruction in a delay slot when not reordering.
800 (md_convert_frag): Warn if an extended instruction appears in a
803 * config/tc-mips.c (mips_pseudo_table): Add "insn".
804 (s_insn): New static function.
805 * doc/c-mips.texi: Document .insn.
807 * config/tc-mips.c (md_begin): Add the general registers to the
809 (mips16_ip): First parse the expression, and then see whether it
810 came up with a register, rather than trying to first see whether
811 we are looking at a register.
816 * config/tc-tic80.c: Numerous changes and additions to flesh
817 out functions that were previously just stubs, and fix some
818 problems found using the new TIc80 testsuite cases.
823 * config/tc-mips.c (mips16_ip): Handle %gprel modifier.
824 (md_apply_fix): Handle BFD_RELOC_MIPS16_GPREL.
826 * config/tc-mips.c (append_insn): Output jump instruction as a
827 pair of 2 byte instructions, rather than as a single 4 byte
831 * config/tc-mips.c (md_begin): Correct test of mips_5900.
832 (mips_ip): Don't check INSN_ISA for a macro.
838 * config/tc-mips.c: added r5900.
843 * itbl-ops.c, itbl-lex.l, itbl-parse.y, itbl-ops.h,
844 config/itbl-mips.h: Add copyright message and fix indentation.
848 * itbl-ops.c: New file. Add support for dynamically read
849 instruction registers, opcodes and formats. Build internal table
850 for new instructions and provide callbacks for assembler and
852 * itbl-lex.l, itbl-parse.y: Lex and yacc parsers for instruction
854 * itbl-ops.h: New file. Header file for itbl support.
855 * config/itbl-mips.h: New file. Mips specific definitions for
860 * config/tc-mn10200.c (md_assemble): If a constant operand won't
861 fit into the constant field of a relaxable operand, then it does
866 * config/tc-mn10200.c (md_estimate_size_before_relax): Treat
867 a jsr target in a different section just like a jsr to
872 * config/tc-mips.c (mips_fix_adjustable): Don't adjust relocations
873 against any mips16 symbols, not just externally visible ones.
874 (md_apply_fix): Corresponding change.
878 * config/tc-mips.c (mips16_ip): Accept floating point registers in
879 the operand of the exit instruction.
883 * symbols.c (resolve_symbol_value): If we leave an equated symbol
884 as O_symbol, copy over the segment.
888 * config/tc-mips.c (md_apply_fix): If we aren't adjusting this
889 fixup to be against the section symbol, adjust the value
892 * symbols.c (resolve_symbol_value): Don't change X_add_number for
894 * write.c (write_relocs): Avoid looping on equated symbols.
895 Adjust fx_offset by X_add_number for each symbol.
896 * config/obj-coff.c (do_relocs_for): Avoid looping on equated
898 (fixup_segment): Add a loop to track down equated symbols and
899 adjust fx_offset appropriately.
903 * config/tc-mn10200.c (md_relax_table): Add entries to allow
905 (md_convert_frag): Handle jmp->bra relaxing.
906 (md_assemble): Handle jmp->bra relaxing.
907 (md_estimate_size_before_relax): Likewise.
911 * config/tc-i386.c (i386_align_code): Add comments explaining the
916 * config/tc-sparc.c (enforce_aligned_data): New static variable.
917 (sparc_cons_align): Don't do anything unless enforce_aligned_data
919 (md_longopts): Add "enforce-aligned-data".
920 (md_show_usage): Mention --enforce-aligned-data.
921 * doc/c-sparc.texi (Sparc-Aligned-Data): New node; document
922 enforce-aligned-data.
924 * config/tc-ppc.c (md_pseudo_table): If OBJ_XCOFF, add "long",
926 (ppc_xcoff_cons): New static function.
928 * write.c (relax_segment): Give an error if a .space symbol is
931 * read.c (read_a_source_file): Don't handle mri_pending_align if
932 the handler is s_globl or s_ignore.
936 * config/tc-d10v.c (find_opcode): Remove unused variable "numops".
938 * config/tc-tic80.c: Many additions to previous placeholder file.
939 * config/tc-tic80.h: Ditto.
944 * config/tc-i386.c (i386_align_code): Improve the nop patterns.
948 * config/tc-mips.c (mips_fix_adjustable): New function.
949 * config/tc-mips.h (tc_fix_adjustable): Call mips_fix_adjustable.
950 (mips_fix_adjustable): Declare.
953 * read.c (read_a_source_file): After doing an mri_pending_align,
954 adjust the line_label if there is one.
955 (s_space): Set mri_pending_align if an odd number of bytes were
960 * config/tc-d10v.h (md_do_align): Add this hook to call
961 d10v_cleanup() when a ".align" is detected. Fixes PR11487.
963 * config/tc-d10v.c (find_opcode): Correctly calculate
964 branch displacement when .aligns are present.
968 * config/tc-mn10200.c (md_relax_table): Define.
969 (md_convert_frag): Implement.
970 (md_assemble): Handle relaxable operands/instructions correctly.
971 (md_estimate_size_before_relax): Implement.
972 * config/tc-mn10200.h (TC_GENERIC_RELAX_TABLE): Define.
976 * config/tc-mips.c (append_insn): Give an error for jumps to a
978 (md_apply_fix): Make a branch to an odd address an error rather
981 * config/tc-mips.c (md_convert_frag): If the user explicitly
982 requested an extended opcode, pass warn as true to mips16_immed.
984 * config/tc-mips.c (mips16_ip): Handle a missing expression like
985 an explicit 0, so that explicitly extended instructions work
990 * ecoff.c (ecoff_build_symbols): Don't generate a local ECOFF
991 symbol for a common symbol.
995 Patch presumed to have been checked in awhile ago but wasn't.
997 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
998 (relax_segment, case rs_org): Move code inside braces. Move locals
999 target,after inside too.
1000 (relax_segment, case rs_machine_dependent): Guts moved to ...
1001 (relax_frag): New function.
1002 Call md_prepare_relax_scan if defined.
1006 * config/tc-m68k.c (m68k_ip): Reject pc-relative addresses for the
1007 'p' operand specifier.
1011 * config/tc-m68k.c (HAVE_LONG_BRANCH): New macro, returns true for
1012 m68k family cpus which support long branch addressing modes.
1013 (m68k_ip, md_convert_frag_1, md_estimate_size_before_relax,
1014 md_create_long_jump): Use it.
1018 * config/tc-mips.c (md_begin): Don't set SEC_ALLOC or SEC_LOAD for
1019 the .reginfo or .MIPS.options section if configured for an
1022 * config/tc-mips.c (md_begin): Don't set interlocks for
1027 * read.c (read_a_source_file): Make sure the symbol ends with
1028 whitespace before checking whether the next character is '='.
1032 * config/tc-i386.c (sco_id): Moved from here...
1033 * config/obj-elf.c (sco_id): ...to here. Adding the identifier
1034 really is an SCO ELF specific thing, not just a SCO x86 specific
1037 start-sanitize-tic80
1040 * configure.in (tic80-*-*): Don't require 'coff'.
1041 * configure: Regenerate.
1046 * read.c (emit_expr): Check for overflow of a negative value
1048 * write.c (fixup_segment): Likewise.
1049 * config/obj-coff.c (fixup_segment): Likewise.
1051 * config/tc-m68k.c (struct label_line): Define.
1052 (labels, current_label): New static variables.
1053 (md_assemble): Mark current_label as text, and clear it.
1054 (m68k_frob_label): New function.
1055 (m68k_flush_pending_output): New function.
1056 (m68k_frob_symbol): New function.
1057 * config/tc-m68k.h (tc_frob_label): Define.
1058 (md_flush_pending_output): Define.
1059 (tc_frob_symbol): Don't warn, just call m68k_frob_symbol.
1060 (tc_frob_coff_symbol): Likewise.
1062 * read.c (read_a_source_file): When defining a macro in MRI mode,
1063 don't add the symbol to the symbol table.
1067 * config/tc-mn10300.c (tc_gen_reloc): Handle sym1-sym2 fixups
1068 here since fixup_segment doesn't (linkrelax is set).
1069 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
1073 * config/tc-mn10200.c (md_assemble): Tweak fx_offset for pc-relative
1078 * config/tc-hppa.c (struct hppa_fix_struct): Tweak fx_r_field's type
1079 to avoid warnings with the native HP compiler.
1080 (fix_new_hppa): Similarly for the r_type argument.
1081 (pa_build_unwind_subspace, hppa_elf_mark_end_of_function): Enclose
1082 in an #if OBJ_ELF to keep gcc -Wall quiet.
1083 (md_apply_fix): Always initialize "result".
1085 * config/tc-mn10200.c (md_assemble): Generate relocations.
1089 * config/tc-m68k.c (s_even): Adjust the alignment of the current
1094 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
1095 pass on the new file hook.
1097 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
1098 !alpha_force_relocation, as local LITERALs can be adjusted to be
1099 relative to the section.
1103 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
1105 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
1106 aligned to odd boundary.
1107 (tc_frob_coff_symbol): Define.
1109 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
1111 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
1113 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
1116 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
1118 (md_assemble): If the size is 'B', set fx_signed.
1119 (md_apply_fix_2): Use fx_signed when checking for overflow.
1121 * write.h (struct fix): Add fx_signed field.
1122 * write.c (fix_new_internal): Initialize fx_no_overflow and
1124 (fixup_segment): Use fx_signed when checking for overflow.
1125 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
1126 fx_signed when checking for overflow.
1130 * NOTES, NOTES.config: Removed. These are rarely, if ever,
1131 updated, and all the useful information is in doc/internals.texi.
1134 * read.c (read_a_source_file): Check for conditional operators
1135 before doing an MRI pending alignment.
1136 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
1137 (tc_conditional_pseudop): Define.
1138 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
1139 * doc/internals.texi (CPU backend): Describe
1140 tc_conditional_pseudoop.
1143 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
1145 (md_begin): Check m68k_rel32_from_cmdline before setting
1147 (m68k_mri_mode_change): Likewise.
1148 (md_longopts): Add --disp-size-default-16 and
1149 --disp-size-default-32.
1150 (md_parse_option): Handle new options.
1151 (md_show_usage): Mention new options.
1152 * doc/c-m68k.texi (M68K-Opts): Document new options.
1155 * config/tc-m68k.c (m68k_index_width_default): New static
1157 (m68k_ip): Use m68k_index_width_default to set the size of a base
1158 register whose size was not given.
1159 (md_longopts): Add --base-size-default-16 and
1160 --base-size-default-32.
1161 (md_parse_option): Handle new options.
1162 (md_show_usage): Mention new options.
1163 * doc/c-m68k.texi (M68K-Opts): Document new options.
1165 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
1167 * configure.in: Recognize mips-*-linux* target.
1168 * configure: Rebuild.
1170 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
1171 work if valueT is only 32 bits.
1173 * config/tc-mips.c: Throughout, check target_big_endian rather
1175 (byte_order): Remove.
1176 (mips_init_after_args): Remove.
1177 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
1178 (mips_init_after_args): Don't declare.
1179 (tc_init_after_args): Don't define.
1181 * config/tc-mips.h (tc_frob_after_relocs): Define if
1183 (mips_elf_final_processing): Likewise.
1184 (ELF_TC_SPECIAL_SECTIONS): Likewise.
1189 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
1190 sld.{h,w} and the operand is relocatable, adjust the adend by
1191 shifting it right one bit.
1196 * read.c (read_a_source_file): Check mri_pending_align after
1197 checking for a macro. From Ronald F. Guilmette
1200 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
1202 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
1203 add segment argument. If OBJ_ELF, treat a relocation against a
1204 symbol in a linkonce section like a relocation against an external
1206 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
1210 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
1214 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
1215 includes config.h instead of host.h, tc.h instead of tp.h, and
1216 targ-env.h instead of target-environment.h.
1217 Also, obj-format.h includes targ-cpu.h instead of
1219 start-sanitize-tic80
1220 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
1221 * configure: Rebuild with autoconf.
1222 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
1223 (TARGET_FORMAT): Define to "coff-tic80".
1224 * config/tc-tic80.c: New file for TIc80 support.
1225 * config/tc-tic80.h: New file for TIc80 support.
1230 * doc/as.texinfo (M): Mention explicitly that -M changes macro
1235 * write.c (adjust_reloc_syms): If the fixup symbol has been
1236 equated to an undefined symbol, convert the fixup to being against
1237 the target symbol. Remove obsolete code handling a special case
1242 * mpw-make.sed: Use NewFolderRecursive for installation.
1246 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
1250 * config/tc-d10v.c (md_assemble): Fix bug which caused
1251 second instruction in a line to be case sensitize. PR11312
1255 * config/tc-mn10200.c (mn10200_insert_operand): Don't
1256 range check operands with MN10200_OPERAND_NOCHECK set.
1257 (check_operand): Likewise.
1261 * config/tc-mips.c: Undo part of last Friday's alignment changes.
1262 (md_begin): Always align the text section to a four byte
1264 (append_insn): Remove call to record_align.
1266 * config/tc-mips.c (insn_label): Remove.
1267 (struct insn_label_list): Define.
1268 (insn_labels, free_insn_labels): New static variables.
1269 (mips_clear_insn_labels): New static function.
1270 (append_insn): Mark all mips16 text labels, and make them odd.
1271 Handle all labels after emitting a nop, not just one. Call
1272 mips_clear_insn_labels rather than just clearing insn_label.
1273 (mips_emit_delays): Add insns parameter, and use it to decide
1274 whether to mark mips16 labels. Handle all labels, not just one.
1275 Force mips16 labels to be odd. Change all callers.
1276 (mips16_immed): Don't check for an odd branch target.
1277 (md_apply_fix): Don't check mips16 mode for a branch reloc.
1278 (mips16_extended_frag): Ignore the low bit in a branch target.
1279 (md_convert_frag): Likewise.
1280 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
1281 clearing insn_label.
1282 (mips_align, mips_flush_pending_output, s_cons): Likewise.
1283 (s_float_cons, s_gpword): Likewise.
1284 (s_align): Use insn_labels rather than insn_label.
1285 (s_cons, s_float_cons, s_gpword): Likewise.
1286 (mips_frob_file_after_relocs): New function.
1287 (mips_define_label): Rewrite to add to insn_labels list.
1288 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
1289 * ecoff.c (ecoff_build_symbols): If the size of a function comes
1290 out odd, increment it.
1292 * config/tc-mips.c (append_insn): Only update prev_insn when not
1293 reordering if place is NULL.
1295 * config/tc-mips.c (mips16_ip): Check for a missing expression
1296 when using the register indirect addressing mode.
1300 * config/tc-mn10200.c (mn10200_insert_operand): Don't
1301 check 24bit operands for overflow.
1302 (check_operand): Likewise.
1306 * doc/as.texinfo (Section): Document how to use the .section
1307 pseudo-op for COFF and ELF.
1311 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
1315 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
1316 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
1317 store them. Adjust other RELAX_MIPS16 macros.
1318 (RELAX_MIPS16_DSLOT): Define.
1319 (RELAX_MIPS16_JAL_DSLOT): Define.
1320 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
1321 handling of whether previous instruction has a fixup. Set
1322 prev_insn_reloc_type.
1323 (mips_no_prev_insn): Clear prev_insn_reloc_type.
1324 (mips16_extended_frag): Use the right base address for a PC
1325 relative add or load.
1326 (md_convert_frag): Likewise. If a PC relative add or load is
1327 used, record the alignment for the section.
1331 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
1332 linkonce section into a reloc against the section symbol.
1334 * config/tc-mips.c (mips16_macro): Remove nop instructions after
1335 branch instructions.
1337 * config/tc-mips.c (md_begin): If configured for an embedded ELF
1338 system, don't set the section alignment to 2**4.
1339 (s_change_sec): Likewise.
1340 (append_insn): Call record_alignment for the section.
1341 (md_section_align): Don't align the section size for an embedded
1346 * write.c (adjust_reloc_syms): Make sure that symbols are
1347 resolved; expression symbols may have been skipped.
1348 * config/obj-coff.c (fixup_segment): Likewise.
1352 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
1353 BFD_RELOC_24_PLT_PCREL relocation.
1354 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
1358 * config/tc-d10v.c (write_2_short): Remove code that called
1359 parallel_ok() when the programmer specified parallel instructions.
1363 * config/tc-mn10300.c (md_assemble): Update to handle endianness
1366 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
1367 * config/tc-mn10300.c (md_assemble): Likewise.
1371 * config/tc-mips.c (append_insn): Make sure there is enough room
1372 in a frag after a mips16 instruction to switch it with a jump
1375 * config/tc-mips.c (mips16_extended_frag): Give an error for an
1376 attempt to use a non absolute symbol in an extending frag.
1380 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
1381 * config/tc-mn10200.h: Likewise.
1385 * app.c (do_scrub_chars): At the end of a C comment, pass space to
1386 UNGET rather than PUT. Set old_state before setting state to -2.
1388 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
1389 when extending because the value is exactly maxtiny + 1.
1391 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
1392 arguments, and store them. Adjust other RELAX_MIPS16 macros.
1393 (RELAX_MIPS16_USER_SMALL): Define.
1394 (RELAX_MIPS16_USER_EXT): Define.
1395 (mips16_small, mips16_ext): New static variables.
1396 (append_insn): Pass mips16_small and mips16_ext to
1397 RELAX_MIPS16_ENCODE.
1398 (mips16_ip): Set mips16_small and mips16_ext.
1399 (mips16_immed): Don't check mips16_autoextend.
1400 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
1402 * write.c (write_relocs): Print an error for an out of range
1403 fixup, rather than calling abort.
1405 * as.c (main): Unlink the output file if there are errors while
1406 generating the fixups.
1410 * config/tc-mips.c (mips16_extended_frag): Don't call
1412 (md_convert_frag): Call resolve_symbol_value before calling
1413 S_GET_VALUE, and don't add in the frag address.
1415 * config/tc-mips.c (mips16_immed): Add file and line parameters,
1416 and use them when reporting errors. Change all callers.
1420 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
1421 Remove '$' prefixing for registers.
1425 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
1426 is initialized before using it.
1427 (d10v_cleanup): No longer uses its argument, so make it void.
1429 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1433 * write.c (fixup_segment): Don't discard the symbol for a PC
1434 relative fixup to an absolute symbol.
1438 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
1439 with multiple sections.
1443 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
1444 no-mips-16 to no-mips16.
1445 (s_mipsset): Accept .set mips16 and .set nomips16.
1449 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
1450 argument, and check for +/- constant following the suffix, folding
1451 it into the expression.
1452 (ppc_elf_cons): Change ppc_elf_suffix calls.
1453 (md_assemble): Ditto.
1454 (shlib): Replace boolean mrelocatable with enumeration shlib.
1455 (md_parse_option): Discriminate between PIC style shared libraries
1457 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
1462 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
1463 change the comment characters.
1465 * config/tc-ppc.c (comment_chars): Delete in favor of
1467 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
1469 (ppc_comment_chars): Select appropriate comment chars by default.
1470 (msolaris): New flag for -m{,no-}solaris.
1471 (md_parse_option): Recognize -K pic. Add support for
1473 (md_show_usage): Update.
1474 (md_begin): Do not set ELF flags if Solaris.
1475 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
1476 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
1480 * as.c (main): Correct handling of flag_always_generate_output.
1484 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
1485 fx_offset, not fx_addnumber.
1487 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
1492 * config/tc-i386.c: Correct misspelling: balenced to balanced.
1496 * config/tc-mips.c (md_section_align): Check for an alignment of
1497 4, not an alignment of 16. Corrects August 7 patch.
1501 * configure, conf.in: Rebuild with autoconf 2.12.
1503 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
1504 alignment is specified.
1506 Add support for mips16 (16 bit MIPS implementation):
1507 * config/tc-mips.c: Extensive additions for mips16 support, not
1509 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
1510 (insn_uses_reg): Change last parameter to an enum.
1511 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
1512 (md_relax_frag): Define as mips_relax_frag.
1513 (mips_relax_frag): Declare.
1514 (struct mips_cl_insn): Add use_extend and extend fields.
1515 (tc_fix_adjustable): Define.
1516 * config/obj-elf.h (S_GET_OTHER): Define.
1517 (S_SET_OTHER): Define.
1521 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
1522 o and p operand specifiers.
1526 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
1527 (relax_segment, case rs_org): Move code inside braces. Move locals
1528 target,after inside too.
1529 (relax_segment, case rs_machine_dependent): Guts moved to ...
1530 (relax_frag): New function.
1531 Call md_prepare_relax_scan if defined.
1532 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
1537 * config/tc-mn10300.c (address_registers): Use '$' as register
1538 prefix instead of '%'.
1539 (data_registers, other_registers, md_assemble): Likewise.
1541 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
1542 (data_registers, other_registers, md_assemble): Likewise.
1544 * config/tc-mn10300.c (md_assemble): Correctly determine the
1545 correct location and type for each relocation.
1546 (md_pcrel_from): Simplify.
1550 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
1551 converted into branch around branch.
1555 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
1556 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
1558 * config/tc-mn10300.c (md_begin): Set linkrelax.
1559 (md_assemble): Create fixups as needed.
1560 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
1564 * config/tc-d10v.c (parallel_ok): When automatically converting
1565 serial ops to parallel, do not consider a branch as the first
1570 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
1574 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
1576 (check_operand): Likewise.
1580 * config/tc-d10v.c (parallel_ok): Branch and link instructions
1582 (write_2_short): Call parallel_ok to check whether two short
1583 instructions the user requested execute in parallel, can be
1588 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
1589 allow a branch and link in parallel with an exe instruction.
1593 * doc/c-d10v.texi: Add info on @word modifier.
1597 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
1598 operands are assumed to be 32bits. Use "bits" field to hold the
1599 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
1600 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
1603 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
1604 of a MN10300_OPERAND_SPLIT operand by operand->shift.
1606 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1607 MN10300_OPERAND_SPLIT.
1611 * config/tc-mn10300.c (md_assemble): Insert operands into
1612 the extension part of the instruction if necessary.
1613 (mn10300_insert_operand): Accept pointer to extension word
1614 argument. Make insn a pointer argument too. Return type
1615 is now void. All callers changed.
1619 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1620 repeated register operands.
1624 * doc/as.texinfo: Added section on reporting bugs.
1626 * config/tc-alpha.c: Change uses of void * to PTR. Change the
1627 alpha_macro emit field to expect a const argument, and change the
1628 arg field to be const. Fix some spacing to follow the GNU
1633 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
1634 (pca56) and 21264 (ev6) cpus.
1635 (md_apply_fix): Private relocation types are now negative.
1636 (alpha_force_relocation): Likewise.
1637 (tc_gen_reloc): Likewise.
1638 (emit_insn): Likewise.
1639 (emit_ldXu): Do the right thing when the hardware can do byte insns.
1640 (emit_stX): Likewise.
1641 (emit_sextX): Likewise.
1645 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
1646 a symbol found in a reloc.
1648 * symbols.c (resolve_symbol_value): Improve the error message if
1649 an undefined symbol is used in an expression.
1653 * doc/internals.texi: Rewrite, and add a lot of documentation.
1654 * doc/Makefile.in (internals.info): New target.
1659 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
1660 against weak symbols.
1664 * config/tc-v850.c (md_assemble): Don't lose for relaxable
1667 * config/tc-v850.c (md_convert_frag): Make sure we insert the
1668 fixup at the right address within the frag.
1670 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
1671 to an absolute value, instead increment it as needed.
1673 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
1674 * config/tc-v850.c: Fix some indention problems.
1675 (md_relax_table): Define for D9->D99 branch displacement
1677 (md_convert_frag): Do something useful instead of aborting.
1678 (md_estimate_size_before_relax): Likewise.
1679 (md_assemble): Note if the matching instruction has a relaxable
1680 operand. If it does, allocate frag with frag_var and don't
1686 * config/tc-d10v.h (md_cleanup): New function. This is needed to
1687 write out any buffered instructions when a ".end" is found.
1691 * read.c (read_a_source_file): New hook md_cleanup().
1695 * write.c (fix_new_exp): Use make_expr_symbol to build an
1696 expression symbol for a complex fixup.
1701 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
1702 hacks to improve parsing of complex hi, lo, zda, etc
1704 (md_assemble): Don't demand and eat a trailing ')' after finding
1705 a v850 relocation prefix. Sign extend the constant in a
1706 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
1708 (parse_cons_expression_v850): Don't eat a trailing ')' after
1709 finding a v850 relocation prefix.
1711 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
1712 (TC_CONS_FIX_NEW): Likewise.
1713 * config/tc-v850.c (parse_cons_expression_v850): New function.
1714 (cons_fix_new_v850): Likewise.
1716 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
1721 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
1722 for an unknown relocation type.
1727 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
1732 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
1733 differently for movea & sst/sld insns.
1738 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
1743 * obj.h (struct format_ops): Add frob_file_after_relocs field.
1744 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
1745 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
1746 frob_file_after_relocs field.
1747 * config/obj-elf.c (elf_format_ops): Likewise.
1748 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
1749 including obj-elf.h.
1753 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
1754 specially if not ELF.
1755 (md_apply_fix): Handle BFD_RELOC_64.
1756 (tc_gen_reloc): Handle BFD_RELOC_64.
1758 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
1759 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
1762 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
1763 Don't check S_GET_OTHER.
1767 * config/tc-mips.c (mips_ip): Accept an odd floating point
1768 register with l.s or s.s.
1770 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
1772 (obj_aout_type): New static function.
1776 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
1781 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
1782 and sdaoff expressions.
1784 * write.c (fixup_segment): Don't add symbol value to addend if
1785 TC_V850 and OBJ_ELF.
1786 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
1789 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
1790 (md_pcrel_from_section): Likewise.
1791 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
1795 * config/tc-v850.c (md_pcrel_from): Delete unused function.
1796 (md_pcrel_from_section): New function.
1797 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
1802 * config/tc-mips.c (load_register): Add cast to offsetT when using
1807 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
1808 (elf_frob_file_after_relocs): ... here. New function.
1809 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
1810 * write.c (write_object_file): Call *frob_after_relocs after the
1811 call to write_relocs.
1813 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
1815 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
1816 symbol, as this messes up linking. Instead, expand the recursive call
1817 inline and change up the appropriate bits to get the 0x8000 offset
1818 in the reloc addend.
1822 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
1823 two symbols in the same segment to be adjusted.
1825 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
1826 * configure: Rebuild.
1830 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
1832 (ppc_elf_validate_fix): Ditto.
1833 (md_assemble): Ditto.
1835 (ppc_pe_section): Ditto.
1836 (ppc_frob_symbol): Ditto.
1840 * config/mn10300.c (md_assemble): Pass an extra shift count
1841 to mn10300_insert_operand based on the opcode format.
1842 (mn10300_insert_operand): Accept and use extra shift count
1845 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
1846 formats rather than hard-coded constants.
1848 * config/tc-mn10300.c (md_assemble): Format D5 instructions
1849 are 7 bytes long. Write out instructions in big-endian format.
1853 * config/tc-mn10300.c (md_assemble): Tweak further so
1854 that all instructions are parsed correctly.
1858 * as.h: Include libiberty.h.
1859 (xmalloc, xrealloc): Don't declare.
1860 * as.c: Don't include libiberty.h.
1861 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
1862 * config/tc-mips.c: Likewise.
1863 * messages.c: Likewise.
1864 (xstrerror): Don't declare.
1865 * xmalloc.c: Remove.
1869 * config/tc-mn10300.h (pre_defined_registers) Remove.
1870 (system_registers, cc_names): Likewise.
1871 (address_registers, data_registers, other_registers): New register
1873 (register_name, system_register_name, cc_name): Remove.
1874 (mn10300_reloc_prefix): Likewise.
1875 (data_register_name): New function.
1876 (address_register_name, other_register_name): Likewise.
1877 (md_assemble): Rough cut at parsing operands. Remove lots of
1879 (md_apply_fix3): Disable for now.
1883 * config/tc-m68k.c (select_control_regs): New function, extracted
1884 out of m68k_init_after_args.
1885 (m68k_init_after_args): Use it.
1886 (mri_chip): Use it here as well to update set of allowed control
1891 * config/obj-elf.c (elf_begin): New function.
1892 (obj_elf_section): Add the section symbol to the symbol table.
1893 * config/obj-elf.h (obj_begin): Define.
1894 (elf_begin): Declare.
1895 * as.c (perform_an_assembly_pass): Call obj_begin if it is
1900 * config/obj-coff.c (fixup_segment): Subtract the section address
1901 from a PC relative reloc if TC_M68K.
1905 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
1906 .uaxword available even if not OBJ_ELF.
1907 (md_atof): Remove unused local variable wordP.
1911 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
1912 for Matsushita MN10x00 support.
1913 * configure.in: Recognize mn10x00-*-*
1914 * configure: Rebuilt.
1918 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
1919 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
1920 (load_expression): track clobbering of base reg before jmp/jsr.
1921 (s_alpha_file): pass case_hack flags and source filename via
1922 symbol table to bfd.
1923 * tc-alpha.h (TC_CONS_FIX_NEW): define
1927 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
1928 * configure: Rebuild.
1932 * read.c (s_macro): Warn if a macro has the same name as a
1934 (s_space): In m68k MRI mode, align to a word boundary.
1935 * macro.c (define_macro): Add namep parameter. Change all
1937 * macro.h (define_macro): Update declaration.
1939 * as.c (show_usage): Print bug report address.
1940 (parse_args): Change version printing to match current GNU
1942 * gasp.c (show_usage): Print bug report address.
1943 (main): Change version printing to match current GNU standards.
1945 * config/tc-m68k.c (init_table): Correct access control unit
1948 * config/tc-alpha.c: Add some static function prototypes.
1949 (alpha_macros): Move to top of file. Make static.
1950 (alpha_num_macros): Move to top of file.
1955 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
1961 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
1962 and just call sprintf_vma.
1966 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
1970 * read.c (float_cons): Call md_flush_pending_output if it is
1975 * config/tc-d10v.c (md_operand): Created. Allows operands to
1977 * config/tc-d10v.h (md_operand): Undefined.
1981 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
1982 (md_assemble): A fixup width of '3' means a 1 byte reloc.
1986 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
1987 reloc for the i960 for a reloc in the same section. This undoes
1988 one of the two changes made Aug 19.
1992 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
1993 symbols to the position of the debugging information.
1997 * expr.c (expr): Always use unsigned right shifts for >>.
2001 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
2002 slots when processing BL fixups.
2004 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
2005 on the first half of the instruction.
2009 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
2010 stabs expression, rather than giving an error.
2012 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
2017 * config/tc-mips.c (append_insn): Fill in the value for a constant
2018 jump, rather than creating a reloc.
2022 * config/tc-mips.c (append_insn): Don't swap an instruction which
2023 sets a condition code with an instruction which uses a condition
2025 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
2028 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
2029 (mips_ip): Give a better error message if the ISA level is wrong.
2030 (md_parse_option): Recognize -mcpu=[v][r]5000.
2034 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
2036 (load_register): Ensure hi32 bits are not lost during lo32bit
2037 processing. Fix shift offset that was overflowing into the next
2038 instruction field. Add code to generate shorter sequences for
2039 constants with a single contiguous seqeuence of ones.
2043 * config/tc-d10v.c (d10v_dot_word): New function to support
2044 "@word" with the word pseudo-op.
2045 (md_apply_fix3): Cleanup and changes to support correct sizes
2046 for 16 and 18-bit relocs.
2050 * configure.in (sparc-*-aout): Set `em'.
2051 * configure: Regenerated.
2052 * config/te-sparcaout.h: New file.
2053 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
2054 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
2055 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
2056 (SPECIAL_CASE_{SETSW,SETX}): Define.
2057 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
2058 (md_begin): Delete setting of `target_big_endian'.
2059 (output_insn): New function.
2060 (md_assemble): Rewrite. Add `setx' support.
2061 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
2063 (md_atof): Add little endian support.
2064 (md_number_to_chars): Likewise.
2065 (md_apply_fix): Likewise.
2066 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
2067 (md_parse_option): Likewise.
2068 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
2072 * ecoff.c (ecoff_new_file): New function.
2073 * ecoff.h (ecoff_new_file): Declare.
2074 * config/obj-ecoff.h (obj_app_file): Define.
2078 * config/tc-alpha.c (load_expression): Bias the .lit8 section
2079 symbol by 32k so that our 16-bit signed offset can address the
2084 * config/tc-mips.c (load_register): Remove unused variable tmp.
2088 * config/tc-mips.c (load_register): Remove unnecessary code that
2089 was causing the high 32bits of 64bit constants to be lost.
2093 * config/tc-d10v.c: Added changes to support function
2094 pointers and "@word" syntax.
2099 * config/tc-v850.c: Remove commented out and #if 0'd code.
2100 (v850_reloc_prefix): Provide prototype.
2101 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
2103 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
2104 (md_assemble): Add default to case statement.
2108 * config/tc-v850.c (md_assemble): Compute size of the instrction
2111 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
2116 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
2117 routines to fetch/store the updated instruction from/to memory.
2118 (v850_insert_operand): If the operand has a specialized insert
2123 * config/tc-v850.c (reg_name_search): Align calling convention to
2124 be like identical function found in tc-ppc.c.
2125 (get_reloc): Removed.
2126 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
2127 (md_assemble): emit fixups.
2128 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
2130 (md_apply_fix3): handle fixups/relocs.
2131 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
2137 * configure.in (sh-*-elf*): New target.
2138 * config/tc-sh.h (TARGET_ARCH): Define.
2139 (WORKING_DOT_WORD): Define.
2140 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
2141 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
2142 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
2143 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
2144 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
2145 (SUB_SEGMENT_ALIGN): Likewise.
2146 (RELOC_32): Don't define.
2147 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
2148 (target_big_endian): Declare if OBJ_ELF.
2149 (TARGET_FORMAT): Define if OBJ_ELF.
2150 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
2152 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
2153 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
2154 (struct sh_count_relocs): Define.
2155 (sh_count_relocs): New static function, broken out of
2156 sh_frob_file. Add BFD_ASSEMBLER code.
2157 (sh_frob_section): Likewise.
2158 (sh_frob_file): Call sh_frob_section.
2159 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
2160 call section_symbol rather than seg_info (seg)->dot.
2161 (md_section_align): Add OBJ_ELF version.
2162 (SWITCH_TABLE_CONS): Define.
2163 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
2164 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
2165 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
2167 (struct reloc_map): Define if not BFD_ASSEMBLER.
2168 (coff_reloc_map): Likewise.
2169 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
2170 (tc_gen_reloc): New function if BFD_ASSEMBLER.
2171 * write.c (write_relocs): Ifdef out fx_where test which triggers
2172 inappropriately for SH ELF.
2173 (write_object_file): Call tc_frob_file_before_adjust and
2174 obj_frob_file_before_adjust if they are defined.
2176 * write.c (write_object_file): Use BFD_RELOC_16, not
2177 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
2179 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
2184 * config/tc-d10v.c (find_opcode): Fix problem with calculating
2185 branch sizes in across sections.
2190 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
2195 * config/tc-v850.c (md_begin): Deal with end of opcode
2201 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
2202 the wrong opcode for cases like st2w where there are many forms
2203 of the same instruction.
2207 * expr.c (operand): If md_parse_name is defined, call it before
2208 calling symbol_find_or_make.
2209 * config/tc-ppc.h (md_parse_name): Define.
2210 (ppc_parse_name): Declare.
2211 * config/tc-ppc.c (reg_name_search): Add regs and regcount
2213 (register_name): Update call to reg_name_search.
2214 (cr_operand): New static variable.
2215 (cr_names): New static const array.
2216 (ppc_parse_name): New function.
2217 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
2218 cr_operand before calling expression.
2222 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
2223 hppa_gen_reloc_type call.
2227 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
2228 on addresses. Improved error messages.
2229 * doc/c-d10v.texi: Added docs for register pairs.
2233 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
2238 * ecoff.c (init_file): Initialize fMerge to 1.
2239 (add_file): Restore old file merging code, but only merge files if
2241 (ecoff_directive_loc): Clear fMerge field of current file.
2242 (ecoff_generate_asm_lineno): Likewise.
2246 * doc/c-d10v.texi: Fix typo.
2251 * config/tc-v850.c (md_assemble): Correct bit masking for
2252 hi and lo expressions.
2254 * config/tc-v850.c (md_assemble): Rough cut at demanding
2255 "ep" or "r30" in sst and sld instructions.
2256 (md_apply_fix3): Don't abort. Just warn that we don't
2259 * config/tc-v850.c (CC_NAME_CNT): Define.
2260 (cc_name): New function.
2261 (md_assemble): Handle V850_OPERAND_CC correctly.
2263 * config/tc-v850.c (md_assemble): Don't forget to initialize
2266 * config/tc-v850.c (reg_name_search): Generalize to search
2267 any given register table.
2268 (register_name): Pass appropriate table and size to reg_name_search.
2269 (system_register_name): New function.
2270 (SYSREG_NAME_CNT): Define.
2271 (md_assemble): Handle operands which are system registers.
2273 * config/tc-v850.c (md_assemble): If we find a register, but the
2274 opcode doesn't want a register, then we don't have a match.
2275 (md_assemble): Get size of the instruction from the opcode table.
2280 * configure.in: Set and substitute HLDENV.
2281 * configure: Rebuild.
2282 * Makefile.in (HLDENV): New variable.
2283 (as.new): Use $(HLDENV).
2285 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
2286 if an array dimension is not known.
2290 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
2291 * doc/c-d10v.texi: Cleanup.
2296 * config/tc-v850.c: New file.
2297 * config/tc-v850.h: New file.
2298 * configure (v850-*-elf): New target.
2299 * configure.in (v850-*-elf): New target.
2304 * doc/c-d10v.texi: New file.
2305 * doc/all.texi: Added D10V stuff.
2306 * doc/as.texinfo: Added D10V stuff.
2310 * config/tc-d10v.c: All references to defined symbols should
2311 now use the optimal instruction. .float and .double now work.
2315 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
2316 section address for the i960 as is done for the i386.
2320 * mpw-config.in: Add wildcards for config matching, add mips-*-*
2321 case, forward-include bfd/elf-bfd.h.
2325 * config/tc-d10v.c: Add additional information to the opcode
2326 table to help determinine which instructions can be done
2331 * config/tc-arm.c: Major changes to add Thumb support, with lots
2333 Reverted to INSN_SIZE macro, rather than insn_size variable.
2334 (insns): Added ARM "bx" instruction support.
2335 (tinsns): Added Thumb instruction definition structure.
2336 (arm_tops_hsh): Added hash structure for Thumb opcodes.
2337 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
2338 (opcode_select,s_arm,s_thumb,s_code): Added.
2339 (decode_shift): Allow upper-case RRX.
2340 (do_ldst): Simpler halfword support.
2341 (do_ldmstm): Improved.
2342 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
2343 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
2344 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
2345 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
2346 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
2348 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
2349 (md_parse_option): Add support for -mthumb.
2350 (md_show_usage): Updated to reflect new command line option.
2351 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
2352 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
2353 fragment information to be held.
2357 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
2358 non-null, but is a constant.
2359 (fix_new_arm): Call make_expr_symbol to make the expression symbol
2360 so that error reporting will work correctly.
2364 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
2365 against weak symbols.
2369 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
2370 (ppc_force_relocation): Declare if OBJ_XCOFF.
2371 * config/tc-ppc.c (ppc_force_relocation): New function if
2376 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
2380 * config/tc-d10v.c: Fix problem with relocs.
2384 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
2385 with nops if not in data_section or bss_section.
2389 Add support for openVMS/Alpha.
2390 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
2391 * config/obj-evax.c: New file.
2392 * config/obj-evax.h: New file.
2393 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
2395 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
2396 defined. Add case for bfd_target_evax_flavour.
2397 * config/vms-a-conf.h: New file.
2398 * conf-a-gas.com: New file.
2399 * configure.in: Add target alpha-*-*vms*.
2400 * configure: Rebuild.
2401 * makefile.vms: New file.
2402 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
2403 Don't call ffs on openVMS/Alpha.
2407 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
2408 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
2410 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
2411 (NEED_DECLARATION_SBRK): New macro.
2412 * configure, conf.in: Rebuild.
2413 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
2414 (strstr): Declare if NEED_DECLARATION_STRSTR.
2415 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
2419 * symbols.c (resolve_symbol_value): Handle addition or subtraction
2420 by a constant before entering the main switch. Reject attempts to
2421 apply an arithmetic function to non-absolute symbols, except for
2422 the special case of subtraction of two symbols in the same
2425 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
2426 not to more than a 16 byte boundary.
2428 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
2430 (tc_gen_reloc): If out of memory call as_fatal rather than
2431 assert. If no howto found, call as_bad_where rather than
2432 as_fatal. Change the error message slightly. Set howto to a
2433 non-NULL value in order to keep going.
2437 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
2441 * config/tc-h8300.c (get_specific): New operand "size" derived
2442 from ".b", ".w" and ".l" extensions. All callers changed. If
2443 the base instruction has no operands, then use the size to
2444 determine which specific instruction to use.
2448 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
2452 * config/tc-mips.c (md_section_align): Don't change addr if
2457 * config/tc-hppa.c: Revert yesterday's changes.
2461 * config/tc-d10v.c: Disable range checking on 16-bit values.
2465 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
2466 not just the ones that don't match.
2470 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
2474 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
2475 pre-cursor to adding Thumb support. Also added cpu_variant flag
2476 information to each of the asm_flg structures.
2477 (md_parse_option): Updated ARM7 parsing to allow 't' for
2478 thumb/halfword support, aswell as 'm' for long multiply.
2479 (md_show_usage): Updated help message.
2480 (md_assemble): Check that instruction flags are applicated to the
2481 current cpu variant.
2482 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
2483 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
2484 signextension instructions.
2485 (do_ldst): Generate halfword and signextension variants if
2486 mnemonic flags match.
2487 (ldst_extend): Do not allow shifts in the offset field of halfword
2488 or signextension instructions.
2489 (validate_offset_imm): Provide check on halfword and signextension
2491 (add_to_lit_pool): Merge identical literal pool values.
2495 * config/tc-hppa.c (selector_table): Add 'E' selector.
2496 (cons_fix_new_hppa): Don't coke on e_esel.
2497 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
2498 to help generate exception handling tables.
2499 (md_apply_fix): Don't try to apply fixups with an e_esel
2501 (hppa_fix_adjustable): Fixups with e_esel selectors
2506 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
2511 * config/tc-d10v.c: Added lots of error checking. Added hacks
2512 to support accumulator shifts.
2516 * symbols.c (S_SET_EXTERNAL): Let .weak override.
2517 (S_CLEAR_EXTERNAL): Likewise.
2518 (S_SET_WEAK): Remove error; just let .weak override.
2522 * config/tc-d10v.c (md_assemble): Now handles multiline
2527 * config/tc-d10v.c: Fix packaging bug. Added range checking.
2528 Added kludge for divs instruction. Fixed minor problem with
2529 multiple text sections.
2530 * config/tc-d10v.h (d10v_cleanup): Change prototype.
2534 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
2535 addresses to be right-shifted by 2.
2539 * config/tc-d10v.c: Many changes to get relocs working.
2540 (register_name): No longer creates a symbol for register names.
2541 (pre_defined_registers): moved to opcodes/d10v-opc.c.
2542 (d10v_insert_operand): Now works correctly for either container.
2543 * config/tc-d10v.h (d10v_cleanup): Declare.
2547 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
2548 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
2552 * tc-alpha.c: Patches to track current minimum alignment to reduce
2553 the number of fragments created with frag_align.
2554 (alpha_current_align): New static variable.
2555 (s_alpha_text): Reset alignment to 0.
2556 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
2557 (s_alpha_stringer, s_alpha_space): New functions.
2558 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
2559 (alpha_cons_align): New function to replace both of them.
2560 (emit_insn): Only align if alpha_current_align is less than 2;
2561 reset alpha_current_align to 2.
2562 (s_alpha_gprel32): Likewise.
2563 (s_alpha_section): New function. Basically duplicate the other
2564 alpha section change hooks. Only define for ELF.
2565 (s_alpha_float_cons): Simplify alignment handling.
2566 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
2567 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
2568 Don't define the s_alpha_cons pseudo-ops. Do define
2569 s_alpha_stringer and s_alpha_space pseudo-ops.
2570 (alpha_align): Skip if less than current default alignment. Set
2572 * tc-alpha.h (md_flush_pending_output): Remove.
2573 (md_cons_align): Add.
2575 * tc-alpha.c: Add oodles of function description comments.
2576 (md_bignum_to_chars): Remove; there are no callers.
2577 (md_show_usage): Mention some more variants.
2582 * configure.in (sparc-*-sysv4*): New target.
2583 * configure: Rebuild.
2585 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
2586 uaxword to use s_uacons.
2587 (sparc_no_align_cons): New static variable.
2588 (s_uacons): New static function.
2589 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
2592 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
2594 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
2595 5 with regard to shared libraries.
2596 * configure: Rebuild.
2598 * config/tc-m68k.c (m68k_ip): Use the correct length when
2599 allocating space for the unsupported architecture error message.
2603 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
2607 * config/tc-d10v.c: New file.
2608 * config/tc-d10v.h: New file.
2609 * configure (d10v-*-elf): New target.
2610 * configure.in (d10v-*-elf): New target.
2614 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
2618 * config/tc-alpha.c (alpha_align): Change fill parameter
2619 to a pointer. Take NULL as 0 or nop depending on section. Change
2621 (s_alpha_align): Rename local variables.
2623 * doc/as.texinfo (.align): Document action of omitted
2628 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
2629 when an unsupported PC relative reloc is seen, rather than calling
2632 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
2633 Instead, check state when '#' comment is seen.
2637 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
2639 (tc_gen_reloc): If fixup was changed to be PC relative, change
2640 reloc type accordingly. Use name of reloc in error message.
2642 * as.h: Don't define const or volatile.
2643 * flonum.h: Don't define const.
2645 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
2646 if fx_pcrel is set. Correct setting the addend case in the
2647 OBJ_ELF case (from Andreas Schwab
2649 (md_show_usage): Correct -mfc5200 to -m5200.
2653 * doc/c-m68k.texi: Document -m5200 flag.
2654 * doc/as.texinfo: Likewise.
2656 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
2661 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
2662 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
2666 * Makefile.in (VERSION): Set to cygnus-2.7.1.
2668 * Released binutils 2.7.
2672 * config/tc-mips.c (mips_ip): Only perform range check when
2673 dealing with O_constant expressions.
2677 * m68k-parse.h (m68k_register): Add new coldfile control
2680 * config/tc-m68k.c (mcf5200_control_regs): New variable,
2681 array of control registers for the coldfire.
2682 (cpu_of_arch): Added mcf5200.
2683 (archs): Added mcf5200.
2684 (init_table): Add new control registers.
2685 (m68k_ip): Added support for new control registers.
2686 (m68k_init_after_args): Likewise.
2688 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
2692 * read.h (is_it_end_of_statement): Declare.
2693 * read.c (is_it_end_of_statement): Remove declaration.
2695 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
2697 (md_assemble): Fix handling of @l with an unsigned constant. Add
2698 default case to reloc switch.
2700 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
2703 * ecoff.c (last_lineno): New static variable.
2704 (add_procedure): Set last_lineno.
2705 (ecoff_directive_loc): Likewise.
2706 (ecoff_generate_asm_lineno): Likewise.
2707 (ecoff_fix_loc): New function.
2708 * ecoff.h (ecoff_fix_loc): Declare.
2709 * config/tc-mips.c (append_insn): When inserting nops, and using
2710 ECOFF debugging, call ecoff_fix_loc.
2714 * config/tc-h8300.c (build_bytes): If an operand type is
2715 marked as SRC_IN_DST retrieve it from the "destination" op.
2719 * configure.in (arm-*-riscix*): Set emulation to riscix.
2720 * configure: Rebuild.
2721 * config/te-riscix.h: New file to define TE_RISCIX.
2723 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
2727 * Makefile.in (config.status): Just run config.status as other
2732 * mpw-config.in (TARGET_OS): Add definition to conf.
2736 * config/tc-mips.c (append_insn): Parenthesize
2737 cop_interlocks expressions.
2741 * listing.c (listing_print): Close the listing file if it is not
2742 stdout. Close the other files opened for the listing.
2744 * config/tc-sparc.h (md_cons_align): Define.
2745 (sparc_cons_align): Declare.
2746 (HANDLE_ALIGN): Define.
2747 (sparc_handle_align): Declare.
2748 * config/tc-sparc.c (sparc_cons_align): New function.
2749 (sparc_handle_align): New function.
2750 * read.c (cons_worker): Call md_cons_align if it is defined.
2752 * as.h (struct frag): Add fr_file and fr_line fields.
2753 * frags.c (frag_new): Set fr_file and fr_line.
2754 (frag_var): Likewise.
2755 (frag_variant): Likewise.
2757 * as.h (struct frag): Remove unused align_mask and align_offset
2760 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
2765 * configure.in (mips-*-osf*): New target.
2766 * configure: Rebuild.
2768 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
2772 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
2773 between co-processor comparisons and branches for the VR4300.
2777 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
2778 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
2780 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
2781 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
2782 includedir): Use autoconf set values.
2787 * listing.c (listing_eject): Don't do anything if listing is 0.
2788 (listing_list): Likewise.
2789 (listing_source_line): Likewise.
2790 (listing_title): Don't save title if listing is 0.
2791 (listing_source_file): Check listing rather than listing_tail.
2793 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
2794 using shared libraries.
2795 * configure: Rebuild.
2799 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
2800 absolute expression if a relocation type was specified.
2804 * configure.in: Add support for *-*-rtems* configurations.
2805 * configure: Rebuild.
2809 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
2810 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
2811 opcodes library for alpha.
2812 * configure: Rebuild with autoconf 2.10.
2813 * config/tc-alpha.c: Substantial rewrite to add ELF support and
2814 use new opcode table.
2815 * config/tc-alpha.h (md_undefined_symbol): Don't define.
2816 (LOCAL_LABEL): Define differently if OBJ_ELF.
2817 (FAKE_LABEL_NAME): Define if OBJ_ELF.
2818 * config/alpha-opcode.h: Remove.
2819 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
2820 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
2821 include/opcode/alpha.h rather than config/alpha-opcode.h.
2825 * config/obj-aout.c (obj_emit_relocations): Give an error if the
2826 relocation symbol was not resolved.
2827 * config/obj-coff.c (do_relocs_for): Likewise.
2829 * write.c (adjust_reloc_syms): Refetch the symbol section after
2830 calling S_GET_VALUE, since it may have changed.
2832 * expr.c (struct expr_symbol_line): Define.
2833 (expr_symbol_lines): New static variable.
2834 (make_expr_symbol): Add entry to expr_symbol_lines.
2835 (expr_symbol_where): New function.
2836 * expr.h: Use extern on function declarations.
2837 (expr_symbol_where): Declare.
2838 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
2839 rather than printing the meaningless name of an expression
2844 * config/tc-i386.c (md_number_to_chars): Deleted.
2845 * config/tc-i386.h (md_number_to_chars): New macro.
2847 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
2849 (load_symbol_address, load_expression): Use build_mem.
2850 (build_operate): New function.
2851 (emit_addq_r): Use it.
2855 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
2859 * config/tc-vax.h (NOP_OPCODE): Define.
2863 * config/obj-vms.h (S_IS_COMMON): Define.
2864 (S_IS_LOCAL): Check for \002 as well as \001.
2865 (LONGWORD_ALIGNMENT): New macro.
2866 (SUB_SEGMENT_ALIGN): Use it.
2870 * config/vms-conf.h: Reconcile with conf.in.
2874 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
2877 * doc/as.texinfo: Document that any number of hex digits can
2880 * as.c (struct defsym_list): Define.
2881 (defsyms): New static variable.
2882 (parse_args): Just put --defsym arguments on defsyms list, rather
2884 (main): Define defsyms after output file is created.
2886 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
2889 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
2890 insn_error rather than calling check_absolute_expr.
2892 * as.c (emulation_name): Remove unused static variable.
2893 (default_emul_bfd_name): Add return NULL to avoid warning.
2894 * ecoff.c (ecoff_stab): Remove unused variables name and
2896 * frags.c (frag_new): Remove unused variable tmp.
2897 * hash.c (hash_grow): Parenthesize + within <<.
2898 (hash_print_statistics): Use %lu, not %d, to print unsigned
2900 * messages.c: Include "libiberty.h".
2901 (fprint_value): Add cast to avoid printf warning.
2902 (sprint_value): Likewise.
2903 * read.c: Include "ecoff.h".
2904 (emit_expr): Add casts to avoid printf warnings.
2905 * read.h: Use extern for function declarations.
2906 (pop_insert): Declare.
2907 * stabs.c: Include "ecoff.h".
2908 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
2909 former_last_fragP, and new_fragP.
2910 * subsegs.h (subsegs_print_statistics): Declare.
2911 * symbols.c (debug_verify_symchain): Change macro to discard
2913 * write.c (dump_section_relocs): Likewise.
2914 * write.h: Use extern for function declarations.
2915 (write_print_statistics): Declare.
2916 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
2918 * config/e-mipself.c (mipself_bfd_name): Likewise.
2919 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
2921 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
2922 emit relocations against external symbols.
2924 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
2925 message if bfd_reloc_type_lookup fails, rather than calling
2928 * config/tc-alpha.c (alpha_force_relocation): Add
2929 BFD_RELOC_12_PCREL to switch.
2933 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
2934 TE_PE (Lfoo, not .Lfoo).
2938 * read.c (s_fill): Don't warn about a zero repeat count.
2940 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
2941 coprocessor register.
2943 * config/tc-i386.c (md_assemble): When checking the size of a
2944 register to set the size of an instruction, do a bitwise and with
2945 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
2950 * config/tc-h8300.c (parse_reg): Tweak error messages.
2951 (build_bytes): Likewise.
2952 (skip_colonthing): Handle :32 suffix.
2953 (get_specific): Promote L_24 to L_32 if it makes a match.
2954 Don't always promote L_8 to L_16.
2955 (do_a_fix_imm): Clean up L_32 and L_24 handling.
2957 * config/tc-h8300.c (Smode): New variable.
2958 (h8300hmode): Turn off Hmode.
2959 (h8300smode): New function. Turn on Smode and Hmode.
2960 (md_pseudo_table): New ".h8300s" pseudo-op.
2961 (parse_reg): Handle "exr" register.
2962 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
2963 Handle "mach" and "machl" operands for ldmac.
2964 (get_specific): Handle "stm.l" and "ldm.l".
2965 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
2966 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
2971 * doc/as.texinfo: Reorder chapter of machine dependent options so
2972 that it is sorted by chip name.
2974 * doc/as.texinfo: Use consistant spelling of Vax.
2975 * doc/c-vax.texi: Likewise.
2979 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
2981 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
2982 (md_apply_fix): Likewise.
2983 (pa_try): New function.
2984 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
2988 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
2989 (pa_level): New function.
2993 * listing.c (listing_newline): Don't do anything if listing is 0.
2997 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
2998 symbol_create, not symbol_new.
3002 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
3003 absolute symbol unless TC_FORCE_RELOCATION returns true.
3005 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
3007 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
3008 the symbol to a pointer. Don't set previous_file_symbol.
3009 Simplify symbol list rearrangement.
3010 (coff_frob_symbol): Don't do anything with C_FILE symbols.
3011 (coff_adjust_symtab): Don't check previous_file_symbol.
3015 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
3017 (md_pseudo_table): Add ppc_elf_lcomm.
3021 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
3022 d0{24:1},d0' works without an immediate prefix on the bit numbers.
3023 (md_begin): Add digits to alt_notend_table.
3024 (md_parse_option): Make s a const pointer.
3026 * config/tc-sparc.c (md_pseudo_table): Add "empty".
3027 (s_empty): New static function.
3029 * config/obj-coff.c (struct filename_list): Only define if not
3031 (filename_list_head, filename_list_tail): Likewise.
3032 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
3033 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
3034 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
3035 have a section of N_ABS rather than N_DEBUG. If we do a merge,
3036 remove the new symbol from the list.
3037 (obj_coff_endef, both versions): Call tag_insert even if there is
3038 an old symbol with the same name, if the old symbol does not
3040 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
3041 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
3042 symbol that will be moved to the end of the symbol list.
3043 (coff_adjust_section_syms): Always call section_symbol for .text,
3045 (coff_frob_section): Likewise. Also, remove unused variable
3048 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
3049 manipulating frags directly.
3050 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
3053 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
3054 whether the next character is '*' before checking whether we are
3055 at the start of a line. Permit LINE_COMMENT_START to start a
3056 comment in state 1 (seen some whitespace) as well, to match the
3059 * gasp.c (do_align): Permit a fill value for .align.
3063 * read.c (next_char_of_string): Warn if a newline is seen in the
3064 middle of a string. Call bump_line_counters when appropriate.
3068 * symbols.c (colon): Use LOCAL_LABEL.
3072 * Makefile.in (install): Don't check to see if tooldir exists.
3073 Make $(tooldir) and $(tooldir)/bin.
3077 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
3082 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
3083 insn_error. Don't put the string "ERROR" in insn_error. Set
3084 insn_error rather than calling as_warn for an unsupported opcode.
3088 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
3089 before permitting -64.
3090 * output-file.c (output_file_create): Remove duplicate
3095 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
3096 section rather than a .reginfo section.
3097 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
3100 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
3101 bit value with the high bit set.
3105 * read.c (s_lcomm): Set section flags for .sbss section.
3107 * config/tc-mips.c (mips_64): New static variable.
3108 (mips_target_format): If mips_64, return elf64 targets rather than
3110 (md_longopts): Add "32" and "64".
3111 (md_parse_option): Handle -32 and -64.
3112 (md_show_usage): Mention -32 and -64.
3113 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
3118 * config/tc-ppc.c (comment_chars): Make '!' a comment character
3119 for Solaris compatibility.
3121 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
3122 .stabd with 4 arguments into a .stabn.
3126 * config/tc-mips.c (macro): When passing X_add_number to
3127 macro_build, cast it to int first.
3131 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
3132 created by emit_expr.
3134 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
3135 as_perror rather than assert.
3139 * config/tc-mips.c (mips_ip): Mark sections created to hold
3140 floating point information as read only.
3144 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
3148 * read.c (potable): Add .skip as a synonym for .space.
3150 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
3151 4 arguments, providing the 4th argument is 0, to allow
3152 compatibility with the Solaris assembler.
3156 * config/tc-sh.h (struct sh_segment_info_type): Define.
3157 (TC_SEGMENT_INFO_TYPE): Define.
3158 (sh_frob_label): Declare.
3159 (tc_frob_label): Define.
3160 (sh_flush_pending_output): Declare.
3161 (md_flush_pending_output): Define.
3162 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
3163 reloc before the instruction if necessary.
3164 (sh_frob_label): New function.
3165 (sh_flush_pending_output): New function.
3166 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
3167 when looking for the reloc for the target of .uses.
3168 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
3169 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
3171 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
3172 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
3173 the absolute symbol.
3175 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
3176 TC_SEGMENT_INFO_TYPE is defined.
3180 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
3181 matches the register size.
3185 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
3187 (do_relocs_for): Likewise.
3188 (fixup_segment): Don't just quit if linkrelax is set. Try to
3189 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
3194 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
3195 relative relocations with -mrelocatable. Also allow unfixed
3196 relocs in .ex_shared.
3200 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
3201 non-zero before assuming this is a long file name.
3202 (w_strings): Likewise.
3203 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
3205 * config/obj-coff.c (w_strings): Move declaration of i inside
3206 #ifdef block which uses it.
3210 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
3211 "relaxmode". Output relocs which identify various relaxing
3212 possibilities for mov.[bwl] instructions.
3213 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
3217 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
3218 (MD_APPLY_FIX3): Define.
3219 * config/tc-arm.c (my_get_expression): Only watch for bad segments
3221 (md_apply_fix3): Renamed from md_apply_fix.
3222 If pcrel reloc and symbol is in different section, undo effects
3227 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
3228 any reloc with an LR% or RR% field selector for SOM.
3232 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
3237 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
3239 (yank_symbols): Likewise.
3243 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
3244 BFD_ASSEMBLER, add name field.
3245 * config/obj-coff.c: Include "libiberty.h".
3246 (coff_header_append): Handle long section names.
3247 (crawl_symbols): Just use the name field for the symbol name,
3248 without worrying about null byte termination.
3249 (w_strings): Handle long section names.
3250 (write_object_file): Likewise. Also, use the name field, rather
3252 (obj_coff_add_segment): Permit long section names.
3253 (obj_coff_init_stab_section): Use the name field, rather than
3255 (adjust_stab_section): Likewise.
3256 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
3258 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
3259 (mem_fmt): Likewise.
3260 (md_apply_fix): Don't check fx_im_disp field.
3264 * configure.in: Add * after sparc*-*-vxworks.
3265 * configure: Rebuild.
3267 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
3268 define it to comment_chars. Use tc_comment_chars rather than
3270 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
3271 * config/tc-m68k.h (tc_comment_chars): Define.
3272 (m68k_comment_chars): Declare.
3273 * config/tc-m68k.c (m68k_comment_chars): Rename from
3274 comment_chars. Change into a pointer rather than an array.
3275 (md_longopts): Add "bitwise-or".
3276 (md_parse_option): Handle OPTION_BITWISE_OR.
3277 (md_show_usage): Mention --bitwise-or.
3278 * doc/c-m68k.texi: Document --bitwise-or.
3282 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
3285 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
3286 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
3287 correct handling of input line pointer, and call
3288 demand_empty_rest_of_line.
3292 * config/tc-sparc.c (in_bitfield_range): New static function.
3293 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
3294 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
3295 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
3296 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
3297 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
3301 * config/obj-coff.c: BFD_ASSEMBLER:
3302 (coff_last_bf): New static variable.
3303 (coff_frob_symbol): Set endndx of a .bf symbol.
3305 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
3306 (last_bfP): New static variable.
3307 (yank_symbols): Set endndx of a .bf symbol.
3311 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
3312 and -s options. Add -be for good measure.
3316 * read.c (s_space): Support non-constant fill value. Handle fill
3317 value correctly for a size other than 1.
3321 * config/tc-arm.c (my_get_float_expression): Update call to
3322 gen_to_words, X_PRECISION changed from 6 to 5.
3326 * config/tc-ppc.c (register_name,reg_name_search): Move register
3327 name lookup from PE specific code to all targets. Add support for
3328 -mregnames/-mno-regnames to control whether register names are
3330 (md_assemble): Call register_name for all platforms.
3331 (md_parse_option): Add support for -mregnames/-mno-regnames.
3333 * configure.in (powerpcle*-*-solaris): Add support.
3334 (powerpc*-*-linux): Ditto.
3335 * configure: Regenerate.
3337 * config/ppc-sol.mt: New config file for PowerPC Solaris.
3341 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
3342 be associated with a single %lo reloc.
3344 * config/tc-mips.c (load_address): Cast X_add_number to valueT
3345 before comparing against MAX_GPREL_OFFSET, so that negative
3346 numbers are handled correctly.
3351 * config/tc-sparc.c (last_insn): New static variable.
3352 (md_assemble): Warn about putting floating point branches in a
3353 delay slot. If architecture is less than v9, insert NOP
3354 instructions between floating point instructions and floating
3355 point branches. (The SunOS assembler does both these operations.)
3356 Save the last instruction opcode.
3357 (sparc_ip): Add pinsn parameter. Change caller.
3359 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
3360 for byte jump to next instruction to skip empty frags.
3364 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
3365 for addq with a 16 bit signed value, just emit a lda instruction
3370 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
3375 * configure.in: Permit --enable-shared to specify a list of
3377 * configure: Rebuild.
3381 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
3382 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
3386 * config/atof-ieee.c: Fix handling of denormalized extended
3387 precision numbers and overflow/underflow detection.
3388 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
3389 not include the 16 bit gap in the m68k extended precision format.
3393 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
3395 * configure: Rebuild.
3399 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
3400 be accessed using the $gp register.
3401 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
3405 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
3406 bfd_mach_sparc_sparclet if current_architecture is sparclet.
3410 * read.c (get_line_sb): Bump line counters based on
3411 input_line_pointer[-1], not *input_line_pointer. Don't bother to
3412 call LISTING_NEWLINE.
3413 (s_macro): Don't call demand_empty_rest_of_line.
3414 * app.c (do_scrub_chars): When handling C style comments, unget
3419 * read.h (enum linkonce_type): Define.
3420 (s_linkonce): Declare.
3421 * read.c (potable): Add "linkonce".
3422 (s_linkonce): New function.
3423 * subsegs.h (segment_info_type): Add linkonce field to
3424 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
3425 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
3426 (obj_coff_pe_handle_link_once): Declare if TE_PE.
3427 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
3429 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
3430 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
3431 entry based on the linkonce field in segment_info.
3432 * doc/as.texinfo: Document .linkonce.
3436 * doc/as.1: Changed to be recognized by catman -w on Solaris.
3440 * stabs.c (s_stab_generic): Call the listing functions before
3441 doing the rest of the processing, which may involve freeing the
3442 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
3443 SEPARATE_STAB_SECTIONS case.
3445 * config/tc-hppa.c: Remove nested comment.
3446 (tc_gen_reloc): Move label done inside the ifdef in which it is
3448 (md_apply_fix): Pass pointers to correct types to libhppa.h
3449 functions. Always return a value.
3451 * config/tc-mips.h (tc_frob_file): Define.
3452 (mips_frob_file): Declare.
3453 * config/tc-mips.c (struct mips_hi_fixup): Define.
3454 (mips_hi_fixup_list): New static variable.
3455 (imm_unmatched_hi): New static variable.
3456 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
3457 offset_reloc. Pass imm_unmatched_hi to append_insn.
3458 (append_insn): Add unmatched_hi parameter. If it is set, add the
3459 new fixup to mips_hi_fixup_list. Change all callers.
3460 (mips_ip): Set imm_unmatched_hi when appropriate.
3461 (mips_frob_file): New function.
3465 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
3466 * configure: Regenerated.
3470 * as.c (main): Call bfd_set_error_program_name.
3474 * as.h (strdup): Don't declare.
3475 * stabs.c: Include libiberty.h
3476 (get_stab_string_offset): Use xstrdup rather than strdup.
3477 (s_stab_generic): Likewise.
3478 * as.c (parse_args): Likewise.
3479 * read.c (s_mri_sect): Likewise.
3481 * gasp.c (change_base): Recognize \(...) construct documented to
3482 pass through enclosed characters literally through to the output.
3483 (process_assigns): Likewise. Also, be more careful to avoid
3484 looking past the end of the buffer.
3488 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
3493 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
3497 * config/alpha-opcode.h: Added cvtst instruction.
3501 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
3502 string. Extract the addend from the result of expression.
3506 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
3507 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
3508 mode, change the state to 3 rather than 1.
3512 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
3513 correspond to 11 March change.
3517 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
3518 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
3522 * configure: Rebuild with autoconf 2.8.
3526 * config/atof-ieee.c (gen_to_words): Improve handling of
3527 X_PRECISION numbers. Based on patches from Andreas Schwab
3532 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
3534 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
3535 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
3536 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
3537 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
3542 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
3546 * config/tc-hppa.c (SUBSEG_MILLI): Define.
3547 (pa_def_subspaces): Add $MILLICODE$.
3548 (pa_spaces_begin): Set section flags for $MILLICODE$.
3552 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
3553 not SHT_NOBITS. Don't tamper with flags based on type if a
3554 special section was found (revert Feb 29 change).
3556 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
3557 if aligning to a longword boundary or greater.
3561 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
3565 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
3566 * configure: Regenerated.
3570 * config/tc-hppa.c: Add default definitions for R_N0SEL and
3571 R_N1SEL since they're not defined for old versions of hpux.
3573 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
3574 Set "sym_ptr_ptr" and "addend" fields to dummy values for
3575 R_N0SEL and R_N1SEL.
3579 * flonum-konst.c: Add two more constants for 1e+-2048 and
3580 1e+-4096, and correct the other constants.
3582 * symbols.c (resolve_symbol_value): Handle O_logical_not.
3586 * config/obj-elf.c (obj_elf_section): Allow predefined section
3587 types to set the nobits type. Avoid a shadowed declaration.
3591 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
3592 reduce relocs using e_nlrsel field selectors.
3594 * write.c (fix_new_exp): Don't use #elif. Some compilers
3597 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
3599 (pa_chk_field_selector): Handle new field selectors for SOM.
3603 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
3604 * configure: Rebuild.
3605 * config/te-linux.h (LOCAL_LABELS_FB): Define.
3606 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
3608 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
3613 * mpw-make.sed: Update to handle shared library support.
3617 * config/tc-sparc.c (sparc_ip): Print all architectures that support
3618 the insn on mismatch.
3622 * configure.in: Add support for a29-coff.
3623 * configure: Rebuild.
3627 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
3628 of the section when looking for the R_SH_USES fixup, because the
3629 frag addresses have not yet been adjusted.
3631 * gdbinit.in: Set a breakpoint on as_warn_where.
3633 * config/tc-mips.c (macro): Add missing arguments to macro_build
3638 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
3639 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
3643 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
3647 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
3648 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
3649 Recognize [uU] format args as sparclet cpregs.
3653 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
3658 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
3659 well as S_IS_EXTERNAL.
3660 (tc_fix_adjustable): Likewise.
3661 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
3662 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
3664 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
3665 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
3669 * config/tc-sparc.c (max_architecture): Change to sparclite for
3671 (default_compatible): Delete.
3672 (sparc_ffs): New function.
3673 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
3674 (sparc_ip): Rewrite architecture match and bump logic.
3678 * configure.in: Check for 'do not mix' from native linker before
3679 trying to use -rpath.
3680 * configure: Rebuild.
3684 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3685 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3686 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3687 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
3689 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
3690 to the symbol lnnoptr field, to get the correct file offset.
3694 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
3695 symbols that are not function, file, or section symbols to be
3700 * configure: Set and substitute RPATH_ENVVAR.
3701 * configure: Rebuild.
3702 * Makefile.in (RPATH_ENVVAR): New variable.
3703 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
3706 doesn't need opcodes. If configuring shared, opcodes needs bfd.
3707 * configure: Rebuild.
3711 * read.c (s_mri_sect): Don't return '\0' in type. Set all
3712 appropriate flags in BFD section.
3714 * configure.in (m68k-*-psos*): New target.
3715 * configure: Rebuild.
3716 * config/te-psos.h: New file.
3717 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
3723 * configure.in: Remove duplicate setting of cpu_type. Check
3724 whether opcodes library is required for on all targets, not just
3726 * configure: Rebuild.
3728 * config/tc-mips.c (mips_big_got): New static variable.
3729 (s_extern): Don't declare.
3730 (reg_needs_delay): New static function.
3731 (macro_build): Permit GOT/CALL_HI/LO relocs.
3732 (macro_build_lui): If place is not NULL, use the number in the
3734 (load_address): Handle mips_big_got case.
3735 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
3737 (OPTION_XGOT): Define.
3738 (md_longopts): Add "xgot" if OBJ_ELF.
3739 (md_parse_option): Handle -xgot.
3740 (md_show_usage): Mention -xgot.
3741 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
3742 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
3746 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
3747 the byte relocation, point it to the low byte of the word.
3751 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
3752 * configure: Rebuild.
3756 * configure.in: Recognize any sparc* cpu.
3757 * configure: Regenerated.
3761 * read.c (potable): Add "mri" and ".mri".
3762 (s_mri): New function.
3763 * read.h (s_mri): Declare.
3764 * app.c (scrub_m68k_mri): New static variable.
3765 (mri_pseudo): New static variable.
3766 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
3767 flag_m68k_mri. Initialize scrub_m68k_mri.
3768 (mri_state, mri_last_ch): New static variables.
3769 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
3771 (app_push): Save new fields.
3772 (app_pop): Restore new fields.
3773 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
3774 If TC_M68K, use a trivial state machine to look for occurrences of
3775 the .mri pseudo-op, and change the mode appropriately.
3776 * as.h (do_scrub_begin): Update prototype.
3777 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
3779 * config/tc-m68k.c (reg_prefix_optional_seen): New static
3781 (m68k_mri_mode_change): New function.
3782 (md_parse_option): Set reg_prefix_optional_seen.
3783 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
3784 (MRI_MODE_CHANGE): Define.
3785 * doc/as.texinfo: Document .mri pseudo-op.
3787 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
3792 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
3793 * config/m68k-parse.h (enum pic_relocation): Define.
3794 (struct m68k_exp): Add pic_reloc field.
3795 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
3796 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
3797 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
3799 * config/tc-m68k.c: Delete definition of NO_RELOC.
3800 (struct m68k_it): Add pic_reloc field.
3801 (add_fix): Copy over pic_reloc field.
3802 (md_pseudo_table): Interpret .align parameter as byte count.
3803 (mote_pseudo_table): Likewise.
3804 (tc_m68k_fix_adjustable): New function.
3805 (get_reloc_code): New function.
3806 (md_assemble): Use it as last argument to fix_new_exp.
3807 (md_apply_fix_2): For a relocation against a symbol don't put the
3808 addend into the data.
3809 (tc_gen_reloc): Different addend computation for OBJ_ELF.
3810 (m68k_ip): Don't relax an operand that requires pic relocation.
3811 (md_begin): Align .text, .data and .bss on 4 byte boundary by
3813 * write.c (fixup_segment): Don't add symbol value to addend if
3814 TC_M68K and OBJ_ELF.
3815 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
3816 (motorola_operand): Add rule for `(zapc, EXPR)'.
3820 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
3821 *really* is permissible before the comma.
3825 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
3826 inserting nop instructions.
3830 * config/te-aux.h: Change include of aux.h to aux-coff.h.
3834 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
3835 class setting for a CALLNAME symbol in COFF.
3837 * read.c (potable): Pass negative numbers for new .balign[wl] and
3838 .p2align[wl] pseudo-ops.
3839 (s_align_bytes): Treat a negative argument as specifying the fill
3841 (s_align_ptwo): Likewise.
3845 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
3846 (do_align): Take new len parameter. Change all callers. Pass it
3848 (s_align_bytes): Arg now indicates the length of the fill pattern.
3849 (s_align_ptwo): Likewise.
3850 * config/obj-coff.c (write_object_file): Pass length to
3852 * config/tc-i386.h (md_do_align): Take new len parameter.
3853 * config/tc-m88k.h (md_do_align): Likewise.
3854 * config/tc-m88k.c (m88k_do_align): Likewise.
3855 * config/tc-sh.h (md_do_align): Likewise.
3856 * config/tc-sh.c (sh_do_align): Likewise.
3857 * doc/as.texinfo: Document new pseudo-ops.
3859 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
3860 done in cvt_frag_to_fill.
3862 * config/tc-sh.h (sh_do_align): Declare.
3863 (md_do_align): Define.
3864 * config/tc-sh.c (sh_do_align): New function.
3866 * ecoff.c (ecoff_build_lineno): Don't try to store the address
3867 difference if the next address is before the current one.
3869 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
3870 (archs): Initialize new field.
3871 (m68k_ip): Don't list alias names when listing CPUs which support
3874 * as.c (main): Call parse_args before read_begin.
3875 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
3876 front of generated pseudo-ops.
3877 * read.c (potable): Ignore "name".
3878 (s_app_file): Permit a single quote after the string, since one
3879 may appear in m68k MRI mode.
3881 * configure.in: Check for --enable-shared. If linking against
3882 shared BFD and opcodes, fix library name on SunOS, and try to set
3884 * configure: Rebuild.
3888 * as.h (flag_m68k_mri): Declare.
3889 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
3890 * Many files: For MRI syntax that is specific to the m68k MRI
3891 assembler, check flag_m68k_mri rather than flag_mri or
3892 MRI_MODE_NEEDS_PSEUDO_DOT.
3896 * config/tc-i960.c (ARCH_HX): Define.
3898 (targ_has_sfr): Handle ARCH_HX.
3899 (targ_has_iclass): Handle ARCH_HX.
3900 (tc_coff_fix2rtype): Add return 0 to avoid warning.
3901 (tc_headers_hook): If the architecture was specified explicitly,
3902 use it when setting the flags. Set the extern variable coff_flags
3903 rather than headers->filehdr.f_flags, since the latter is set
3904 unconditionally in obj-coff.c.
3905 (i960_handle_align): Remove unused variable fixp.
3907 Support for building bfd and opcodes as shared libraries, based on
3909 * configure.in: Set OPCODES and BFD to search directories.
3910 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
3911 * configure: Rebuild.
3912 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
3913 (LIBDEPS): New variable.
3914 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
3915 $(HLDFLAGS) in link.
3916 (check): Set LD_LIBRARY_PATH in the environment.
3920 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
3921 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
3922 read-only, not read/write.
3926 * Makefile.in (INSTALL_XFORM): Remove -e.
3930 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
3931 than comparing S_GET_SEGMENT to undefined_section.
3932 (write_object_file): Skip symbols which were equated to an
3933 undefined or common symbol.
3934 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
3935 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3937 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
3938 or S_IS_COMMON call S_GET_VALUE.
3939 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3940 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
3941 to an undefined or common symbol, convert the reloc to be against
3943 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3944 undefined or common symbol.
3945 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3946 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
3947 to an undefined or common symbol, convert the reloc to be against
3949 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3950 undefined or common symbol.
3951 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
3952 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3954 (yank_symbols): Skip symbols which were equated to an undefined or
3959 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
3960 * config/obj-bout.h (S_IS_LOCAL): Likewise.
3962 * configure.in: Make sure we only add m68k-parse.o to
3963 ${extra_objects} once, no matter how many m68k targets have been
3965 * configure: Rebuild.
3969 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
3970 * configure: Rebuild.
3974 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
3975 * doc/c-m68k.texi: Document .extend and .ldouble.
3977 * configure.in (m68*-apple-aux*): New target.
3978 * config/te-aux.h: New file.
3979 * config/obj-coff.c (compare_external_relocs): New static function
3981 (do_relocs_for): Sort relocs if TE_AUX.
3982 (fixup_segment): If TE_AUX, store common symbol value in segment.
3983 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
3987 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
3989 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
3991 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
3992 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
3994 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
3995 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
3996 an undefined symbol and the size of a function symbol.
3997 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
3998 for all common symbols.
4002 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
4005 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
4010 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
4013 * ecoff.c (localsym_t): Add addend field.
4014 (add_ecoff_symbol): Add addend argument. Change all callers.
4015 (coff_sym_value): Make static.
4016 (coff_sym_addend): New static variable.
4017 (ecoff_directive_def): Initialize coff_sym_addend.
4018 (ecoff_directive_val): Accept symbol + constant.
4019 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
4020 (ecoff_build_symbols): Include the addend in the symbol value.
4024 * config/tc-ppc.c (md_assemble): Ignore overflow on
4025 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
4029 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
4030 bytes in size, so offset appropriately in big endian mode when
4031 writing the bottom 2 bytes.
4035 * config/tc-sparc.c (default_compatible): New static local.
4036 (md_begin): Initialize it. Rewrite warn_on_bump handling.
4037 (sparc_ip): If no architecture or -bump specified, don't mark as
4038 mismatched those in default_compatible.
4043 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
4044 * configure: Rebuild.
4045 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
4046 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
4047 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
4048 (sco_id): Declare if SCO_ELF.
4052 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
4053 ({max,warn_after}_architecture): New static locals.
4054 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
4055 If both architecture and -bump requested, set max_architecture to max.
4056 (sparc_md_end): Simplify.
4057 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
4058 Rewrite code to bump architecture and check for conflicts.
4059 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
4060 with Solaris assembler.
4061 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
4062 (md_show_usage): Update.
4066 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
4070 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
4071 #ifdef sparcv9 when choosing value.
4074 (sparc_md_end): Declare.
4075 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
4076 (initial_architecture): New static local.
4077 (can_bump_v9_p): Likewise.
4078 (NO_V9): Delete all occurrences.
4079 (sparc_md_end): New function.
4080 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
4081 Don't bump architecture to v9 unless can_bump_v9_p set.
4082 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
4083 * configure.in (sparc64 target cpu): Don't set obj_format here.
4084 (SPARC_V9): Renamed from sparcv9.
4085 (sparc64-*-elf*): Define SPARC_ARCH64.
4086 * configure: Regenerated.
4087 * acconfig.h (SPARC_V9): Renamed from sparcv9.
4088 (SPARC_ARCH64): Add.
4089 * conf.in: Regenerated.
4090 * config/vmsconf.h: Update.
4094 * config/tc-mips.c (load_register): Optimise "dli" loads.
4095 (md_show_usage): add "-mcpu=vr4100" to help text.
4099 * symbols.c (resolve_symbol_value): If a symbol is equated to an
4100 undefined symbol, preserve the X_op of O_symbol.
4101 (S_GET_VALUE): Fix check to permit this case.
4102 * write.c (write_relocs): If a reloc is against an undefined
4103 symbol equated to another symbol, change the reloc to be against
4105 * config/obj-coff.c (do_relocs_for): Likewise.
4107 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
4112 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
4113 computer is PowerPC.
4115 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
4117 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
4118 mentioned in the eabi.
4122 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
4123 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
4127 * config/tc-ppc.c (mapping): Add more relocation suffixes.
4131 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
4132 as a section it is ok to have unadorned -mrelocatable pointers in.
4136 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
4141 * config/obj-elf.c (obj_elf_section): Add hooks so machine
4142 dependent section attributes can be handled.
4144 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
4145 macros to add support for exclude section flag and ordered section
4148 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
4149 functions to add support for exclude section flag and ordered
4154 * subsegs.c (section_symbol): Don't try to look up the section
4155 symbol in the hash table. It should be possible to have a symbol
4156 with the same name as a section, but no connection to it.
4158 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
4161 * expr.c (operand): Skip whitespace after a close parenthesis.
4166 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
4167 as_warn_where for relocation overflow.
4168 (parse_reg): Accept register name only if next character is
4173 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
4174 addresses for static variables.
4178 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
4179 and xgot give the old toc16 relocation.
4180 (md_apply_fix3): Support all GOT relocations.
4184 * config/tc-mips.c (load_address): Correctly handle a constant in
4189 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
4190 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
4191 (build_Mbytes): Handle REG_B.
4195 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
4199 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
4203 * config/obj-coff.c (write_object_file): Set the s_align field to
4204 the number of bytes, rather than to the power of 2.
4208 * Makefile.in (DISTCLEAN_HERE): New variable.
4209 (distclean): Use it.
4210 (maintainer-clean): Depend upon clean-here rather than clean,
4211 distclean, and clean-info. Run make maintainer-clean in doc.
4212 Remove files listed in DISTCLEAN_HERE.
4213 * doc/Makefile.in (maintainer-clean realclean): Split out from
4214 distclean. Depend upon clean-info and distclean.
4218 * mac-as.r: Fix copyright and version strings.
4219 (cfrg): Use PROG_NAME instead of literal name.
4223 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
4225 * config/tc-a29k.h (tc_unrecognized_line): Define.
4226 * config/tc-a29k.c (a29k_unrecognized_line): New function.
4227 (md_operand): Handle a29k style local dollar labels.
4231 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
4235 * read.c (s_fill): If md_flush_pending_output is defined, call
4240 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
4241 Treat rs_align_code like rs_align.
4245 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
4246 converting an absolute address into a PC-relative one.
4250 * mpw-config.in: Don't always use te-generic.h for emulation.
4251 (powerpc-apple-macos): Use emulation te-macos.h.
4253 (mips-gm-magic): New configuration.
4255 * mpw-make.sed (install, install-only): Edit in Mac-specific
4260 * configure.in: Improve message about unsupported ELF targets.
4261 * configure: Rebuild.
4263 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
4264 pattern is zero. From Manfred Hollstein.
4268 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
4269 the various idata sections, we check the name on the .section pseudo.
4273 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
4274 add section's paddr to add_number; compatibility to native as and
4279 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
4281 * hash.c (struct hash_entry): Moved here...
4282 * hash.h (struct hash_entry): ...from here.
4284 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
4285 if it's already known to be null.
4289 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
4290 bit relocs converted to PC relative relocs. This turned out to
4291 add the offset from the beginning of .text twice.
4295 * stabs.c (s_stab_generic): In 's' case, free string from
4298 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
4300 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
4303 * subsegs.h (segment_info_type): Make bitfields unsigned.
4305 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
4306 and move them together. On most systems this will reduce the
4307 structure size by one word.
4308 (operatorT): Define O_max.
4309 * expr.c (expr_begin): Verify that X_op is wide enough to hold
4312 * read.c (pop_insert): Print error returned by hash table
4315 * as.c (dump_statistics): Split out from main; dump some hash
4316 table stats and target-specific stats.
4317 (start_time): No longer automatic to main.
4318 (main): Set file-level start_time and call dump_statistics at
4319 exit. Exit by calling xexit.
4320 (show_usage): Make --statistics description less specific.
4321 * subsegs.c (subsegs_print_statistics): New function.
4322 * write.c (write_print_statistics): New function.
4323 (n_fixups): New static variable.
4324 (fix_new_internal): Increment it.
4325 * read.c (read_print_statistics): New function.
4326 * read.h (read_print_statistics): Declare.
4327 * symbol.c (symbol_print_statistics): New function.
4328 * symbol.h (symbol_print_statistics): Declare.
4329 * hash.c (hash_print_statistics): New function.
4330 * hash.h (hash_print_statistics): Declare.
4331 * config/tc-i386.c (i386_print_statistics): New function.
4332 * config/tc-i386.h (i386_print_statistics): Declare.
4333 (tc_print_statistics): New macro.
4334 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
4337 * hash.c (DELETED): Rewrite to use a valid but unique address.
4338 (START_POWER): Reduce to 10.
4339 (enum stat_enum): New enumerator, replacing STAT_* index macros.
4340 Add new values for counting strcmp calls.
4341 (GROW_FACTOR): New macro.
4342 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
4344 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
4345 (hash_new): Use FULL_VALUE.
4346 (struct hash_control): Definition moved here.
4347 (hash_code): Don't mask to low bits.
4348 (hash_ask): Mask returned hash code here. Check hash value before
4349 calling strcmp; count strcmp calls.
4350 * hash.h (struct hash_control): Declare, don't define, here.
4351 (HASH_STATLENGTH): Deleted.
4352 (struct hash_entry): Add field for hash code.
4353 (hash_say, hash_apply): Don't declare.
4355 * hash.c (destroy): Return void.
4356 (applicatee): Ditto.
4357 (main): Fix declarations.
4358 (hash_apply): Return void. Argument `function' returns void. Put
4359 inside "#ifdef TEST".
4360 (hash_say): Define only if TEST is defined.
4361 * hash.h (hash_apply, hash_say): Declarations deleted.
4365 * configure: Rebuild with autoconf 2.7.
4369 * aclocal.m4 (AC_PROG_CC): Remove local definition.
4370 * configure: Rebuild with autoconf 2.6.
4374 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
4375 (ppc_stabx): Don't increment ppc_debug_name_section_size.
4377 (ppc_frob_file): Remove.
4378 * config/tc-ppc.h (tc_frob_file): Don't define.
4379 (ppc_frob_file): Don't declare.
4383 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
4384 * config/alpha-opcode.h: Include one-operand variants of jmp and
4387 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
4388 with te-delta.h in 8.3 file systems.
4389 * configure.in: Adjusted.
4393 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
4395 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
4397 * expr.c (operand): Handle MRI suffixes after unadorned 0.
4401 Version 2.6 released.
4402 * Makefile.in (VERSION): Updated to 2.6.
4404 * config/obj-coff.c (write_object_file): Change use of md_do_align
4405 to pass a pointer rather than a fill value, to match other uses.
4409 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
4411 * as.c (main): Move md_end call to just after call to
4412 perform_an_assembly_pass. Delete cpu-specific code here.
4413 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
4414 * config/tc-arm.c (md_end): Unused function deleted.
4415 * config/tc-ns32k.c (md_end): Ditto.
4417 * config/tc-i386.c (i386_align_code): New function, moved here
4418 from HANDLE_ALIGN macro.
4419 * config/tc-i386.h (HANDLE_ALIGN): Call it.
4423 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
4428 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
4429 multi-byte nops to avoid changing gcc. The necessary gcc change
4430 might break old assemblers.
4434 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
4435 generate optimal multi-byte nop instructions for ".align n"
4436 ".align n,0x90", and aligns requiring more than 15 bytes of
4437 padding still generate multiple 0x90's as before.
4441 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
4445 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
4446 * configure: Update.
4447 * mpw-make.sed: Reorder commands to make sed happier.
4448 * config/te-macos.h: New file.
4449 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
4453 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
4454 conditional branches.
4455 (md_apply_fix): Likewise.
4459 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
4460 generate a reloc which the linker will not handle correctly. Fix
4461 overflow checking--R_IREL is 18 bits, not 17.
4465 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
4466 from a PC relative reloc if TC_A29K.
4468 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
4469 code depends upon it.
4471 * config/tc-a29k.c (md_operand): Handle $float, $double, and
4472 $extend. Based on code from Eric Freudenthal
4474 * config/tc-a29k.h (LEX_DOLLAR): Define.
4475 * read.c (LEX_DOLLAR): Define if not defined.
4476 (lex_type): Use LEX_DOLLAR.
4480 * configure.in (a29k-nyu-sym1): New target, just like other a29k
4485 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
4489 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
4490 used only by "dlltool.c".
4494 * config/tc-alpha.c: Undefine inline if not __GNUC__.
4495 (md_pseudo_table): Don't define "extern".
4499 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
4501 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
4505 * macro.c (macro_expand_body): Don't warn about == with a
4506 nonexistent parameter, in case it is in a comment field.
4508 * as.c (main): On TC_A29K, call macro_init with macro_alternate
4510 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
4511 set, even if macro_alternate is set.
4512 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
4513 (sub_actual): If macro_strip_at, and kind is '@', don't look up
4514 the token unless it ended in '@'.
4515 * config/tc-a29k.c (line_separator_chars): Remove '@'.
4516 * doc/c-a29k.texi: Document macro usage on A29K.
4520 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
4522 (install_operand): Likewise.
4524 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
4528 * configure.in (m88k-motorola-sysv*): New target.
4529 * configure: Rebuild.
4530 * config/te-delta88.h: New file.
4531 * config/obj-coff.c (write_object_file): Use md_do_align if it is
4533 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
4534 (md_do_align): Define.
4535 * config/tc-m88k.c: Include "subsegs.h".
4536 (m88k_do_align): New function.
4538 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
4539 (COFF_NOLOAD_PROBLEM): Define.
4540 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
4544 * config/tc-i386.c (md_assemble): For a jump instruction with
4545 non-constant target, require 7 available bytes in the current
4550 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
4555 * configure.in (alpha-*-linux*): Use ecoff.
4556 * configure: Rebuild.
4557 * ecoff.c (ecoff_directive_extern): New function.
4558 (ecoff_directive_weakext): New function.
4559 (ecoff_build_symbols): Handle weak symbols.
4560 (ecoff_setup_ext): Likewise.
4561 (ecoff_frob_symbol): Warn about weak common symbols.
4562 * ecoff.h (ecoff_directive_extern): Declare.
4563 (ecoff_directive_weakext): Declare.
4564 * symbols.c (S_IS_WEAK): New function.
4565 * symbols.h (S_IS_WEAK): Declare.
4566 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
4568 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
4573 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
4575 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
4576 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
4577 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
4578 * config/obj-coff.c (coff_frob_symbol): Don't call
4579 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
4580 one. If OBJ_XCOFF, try to figure out whether the symbol is going
4583 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
4584 (ppc_stab_symbol): New static variable.
4585 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
4586 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
4587 sy_tc.real_name to the stab string.
4588 (ppc_bc, ppc_ec): New static functions.
4589 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
4591 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
4593 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
4596 * input-scrub.c (input_scrub_push): Reset sb_index.
4600 * listing.c (listing_newline): Don't create a frag in the absolute
4605 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
4606 (ppc_csect): Move most of the code to ppc_change_csect, and call
4608 (ppc_change_csect): New static function, taken from ppc_csect.
4609 (ppc_section): New static function.
4610 (ppc_saw_abs): New static varable.
4611 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
4612 a symbol has no csect.
4613 (ppc_adjust_symtab): New function.
4614 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
4615 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
4617 * write.c (write_object_file): If tc_adjust_symtab is defined,
4618 call it just before the call to obj_adjust_symtab.
4620 * symbols.c (symbol_find_or_make): Change name to be const.
4621 * symbols.h (symbol_find_or_make): Update declaration.
4625 * doc/as.texinfo (Align): Mention SH.
4626 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
4628 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
4629 (Z8000 Directives, even): Don't give numeric align value, instead
4634 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
4639 * Makefile.in (diststuff): Also make info.
4640 (maintainer-clean realclean): Also make clean-info.
4644 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
4645 (pa_subspace): For ".nsubspa", always create a new subspace
4646 with the given attributes, even if one already exists with the
4651 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
4652 TC_FORCE_RELOCATION, taking an additional section argument. If
4653 the section of the target symbol is not the same as the current
4654 section, always force the relocation to be used.
4655 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
4657 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
4658 md_pcrel_from function, taking an additional section argument.
4659 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
4661 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
4662 older TC_FORCE_RELOCATION if not defined.
4663 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
4664 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
4665 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
4666 TC_FORCE_RELOCATION.
4670 * input-scrub.c (as_where): Set name to null pointer if we don't
4672 * messages.c (identify): Only print filename if non-null.
4673 (as_show_where): Ditto, for line number too.
4674 (as_warn_internal, as_bad_internal): Ditto.
4676 * input-file.c (input_file_open): If the input file can't be
4677 opened, consider it an error.
4681 * config/tc-mips.c: Added mips_4100 control, and support for
4682 accepting the 4100 as a MIPS architecture variant (md_begin,
4683 macro_build, mips_ip, md_parse_option). Adding suitable
4684 command-line OPTIONs, and updating the help text (md_show_usage).
4688 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
4689 BFD_ASSEMBLER is defined.
4691 * Use one active frag and one obstack per frag chain:
4692 * frags.c (frags): Variable deleted.
4693 (frag_alloc): New function.
4694 (frag_grow, frag_more, frag_variant, frag_now_fix,
4695 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
4697 (frag_new): Ditto. Verify that frch_last and frag_now match on
4698 entry and exit, and that old frag_now has non-zero type. Replace
4699 "know" uses with "assert". Use frag_alloc instead of mucking with
4701 * frags.h (frags): Declaration deleted.
4702 * subsegs.h (struct frchain): Add new field frch_frag_now.
4703 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
4705 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
4706 give it any stricter alignment than frchainS structures need. Do
4707 not initialize frags obstack. Set frag_now to point to
4708 dummy_obstack. Initialize absolute_frchain.
4709 (subseg_set_rest): Save and restore frag_now in frch_frag_now
4710 field of frchainS. Don't create new frags on section switch, and
4711 use frag_alloc when creating a new frag chain. For absolute
4712 section, set frchain_now to absolute_frchain. Verify that
4713 frch_last and frag_now match on entry and exit. Initialize
4714 per-chain obstack, and under gcc, set required alignment to that
4715 needed by fragS structure.
4717 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
4719 * stabs.c (get_stab_string_offset): Only copy input string if a
4720 fresh copy is needed, not if the section already exists.
4721 (s_stab_generic): Cache stab section name to bypass lookups, since
4722 usually it will match. Could be made faster still by changing the
4723 memory allocation rules.
4724 (s_xstab): Cache section name to bypass repeated string
4727 * frags.c (frag_new): Deleted register declarations.
4729 * listing.c (frag_now): Don't declare.
4731 * as.c (chunksize): New variable.
4732 (debug_memory): New variable.
4733 (main): If debug_memory is set, reduce chunksize and
4735 * as.h (chunksize): Declare it.
4736 * read.c (read_begin): Use it.
4738 * config/tc-alpha.c (md_shortopts): Include 'g'.
4739 (md_parse_option): Ignore it.
4741 * Makefile.in (distclean): Remove Makefile and config.status from
4742 testsuite directory.
4743 (clean-here): Don't delete testsuite. Instead, delete only the
4744 files within it that would be generated by running tests.
4746 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
4747 frag_now_fix instead of accessing obstack info directly.
4748 * config/tc-arm.c (s_ltorg): Ditto.
4749 (md_assemble): Ditto.
4751 * config/tc-i386.c (md_assemble): Call frag_grow instead of
4756 * stabs.c (aout_process_stab): Insert debug symbol into symbol
4757 chain after parsing value expression, if any, to avoid separating
4758 continued .stabs lines.
4762 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
4763 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
4767 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
4768 -mcom support and PPC_OPCODE_ANY for -many.
4769 (md_parse_option): Ditto.
4771 (md_begin): For duplicate instructions, print all duplicates
4776 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
4777 common mode operation.
4778 (md_show_usage): Add -mcom to usage message.
4782 * expr.c (op_rank): Add O_symbol_rva.
4783 * expr.h (operatorT): Add O_symbol_rva.
4784 * read.c (cons_worker): Set O_symbol_rva when necessary.
4785 * write.c (fix_new_exp): Understand O_symbol_rva.
4789 * config/tc-mips.c: Correct s_cons arguments. From Michael
4794 * config/tc-ppc.c (ppc_macro): Make count unsigned.
4795 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
4796 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
4798 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
4799 make non-static. Change all users.
4800 * config/obj-coff.h (coff_n_line_nos): Declare.
4806 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
4808 * config/obj-elf.c (obj_elf_common): Convert specified byte
4809 alignment to power of two. Set size of local bss symbol.
4811 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
4815 * sb.c, macro.c: Decide whether to include <string.h> or
4816 <strings.h> just as as.h does.
4820 * Makefile.in (site.exp): Fix setting of $srcdir.
4822 * config/tc-arm.c (md_atof): Fix little-endian output.
4823 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
4828 * doc/as.texinfo: Split out the NS32k family documentation,
4829 despite its being commented out for now.
4830 * doc/c-ns32k.texi: New file.
4832 * sb.c, macro.c: Include string.h.
4834 * Makefile.in (comparison): Only check *.o; we don't care if
4835 timestamps inserted by the native linker differ.
4837 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
4838 if alignment stricter than 4 bytes is requested; in that case,
4839 align to a 4-byte boundary first.
4843 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
4844 uninitialized" warning for `Max_Source_Offset'.
4848 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
4849 qualifiers and detect errors if present.
4850 (md_assemble): Add call to parse_toc_entry. Also added some support
4851 for the [tocv] qualifier.
4852 (ppc_pe_tocd): New function to support data in the toc section.
4856 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
4861 * config/tc-m68k.c (isword): Accept all values from -65536 to
4862 +65535, so ~VAL will not be rejected.
4864 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
4865 skip characters after the pseudo-op.
4867 * read.c (get_line_sb): Don't look past buffer_limit.
4868 (s_include): In MRI mode, skip characters after the file name.
4872 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
4873 periods in register names.
4875 For older changes see ChangeLog.1.