3 * configure.in: Don't set files and links. Don't call
4 AC_LINK_FILES. Substitute te_file. Create targ-cpu.h,
5 obj-format.h, targ-env.h, and itbl-cpu.h in AC_OUTPUT.
7 * Makefile.in (TARG_CPU_C): New variable.
8 (TARG_CPU_O, TARG_CPU_H): New variables.
9 (OBJ_FORMAT_C, OBJ_FORMAT_O, OBJ_FORMAT_H): New variables.
10 (TARG_ENV_H, ATOF_TARG_C, ATOF_TARG_O): New variables.
11 (SOURCES): Rename from REAL_SOURCES. Delete old definition.
12 (LINKED_SOURCES): Remove.
13 (HEADERS): Rename from REAL_HEADERS. Delete old definition.
14 (LINKED_HEADERS): Remove.
15 (OBJS): Use $(TARG_CPU_O), etc., rather than targ-cpu.o, etc.
16 ($(OBJS)): Depend upon $(TARG_ENV_H), etc., rather than
18 ($(TARG_CPU_O), $(OBJ_FORMAT_O) $(ATOF_TARG_O)): New targets.
19 (targ-cpu.o, obj-format.o, atof-targ.o): Remove targets.
20 (itbl-cpu.h): Remove target.
21 (DISTCLEAN_HERE): Remove targ-cpu.c, obj-format.c, atof-targ.c,
26 * doc/as.texinfo (Symbol Names): Don't use obsolete @ctrl macro.
30 * config/tc-m68k.c (mri_chip): Replace calls to get_symbol_end by
31 open coded loop that does not require the name to start with a
36 * frags.c (frag_var): Change offset parameter to offsetT.
37 (frag_variant): Likewise.
38 * frags.h (frag_variant, frag_var): Update declarations.
39 * config/tc-m68k.c (struct m68k_it): Change foff field to
41 (add_frag): Change off parameter to offsetT.
42 * Several files: Add casts to calls to frag_var.
44 * Makefile.in (m68k-parse.c): Depend upon itbl-parse.c, to
45 serialize a parallel make.
46 (itbl-parse.h): Split target out from itbl-parse.c.
50 * config/m68k-parse.y (motorola_operand): Allow (zdireg,EXPR).
52 * config/te-delta.h (COFF_COMMON_ADDEND): Define.
53 * config/obj-coff.c (fixup_segment): Check COFF_COMMON_ADDEND when
54 storing the value of a common symbol.
58 * config/obj-coff.c (glue_symbols): Unused variable symbolP
60 (crawl_symbols): Do not modify symbol_rootP and symbol_lastP here;
61 that is done by symbol_remove and symbol_insert.
63 * config/obj-coff.h (S_IS_LOCAL): Return 0 for a debugging
68 * config/tc-mips.c (load_register): In 32 bit mode, when not
69 dealing with a 64 bit number, permit the upper 32 bits to be set
70 even if bit 31 is not set.
74 * read.c (potable): Add "equiv".
75 (s_set): Handle .equiv based on argument.
76 * doc/as.texinfo (Equiv): New node to document .equiv.
77 (Err): New node to document .err.
81 * Many files: Add function prototypes.
82 * as.c (show_usage, parse_args): Make static.
83 * frags.h (frag_alloc): Declare.
84 * subsegs.c (subseg_set_rest): Don't declare frag_alloc.
85 * symbols.c (dollar_label_instance): Change return type to long.
86 * symbols.h (print_symbol_value): Declare.
87 (print_expr, print_expr_1, print_symbol_value_1): Declare.
88 * write.c (fix_new_exp): Don't declare make_expr_symbol.
89 (remove_subsegs, relax_frag): Make static.
90 * config/atof-vax.c (atof_vax_sizeof): Change letter to int.
91 (what_kind_of_float): Likewise.
92 (atof_vax): Make static. Change what_kind to int.
93 (md_atof): Change what_statement_type to int.
94 * config/obj-ecoff.h (obj_ecoff_set_ext): Declare.
95 * config/tc-alpha.c (vax_md_atof): Declare.
96 (md_atof): Don't declare atof_ieee and vax_md_atof.
97 * config/tc-i386.c (set_16bit_code_flag): Make static.
98 * config/tc-i386.h (tc_i386_fix_adjustable): Declare.
99 * config/tc-m68k.c (add_fix): Change width to int.
100 (insert_reg): Change regname to const.
101 (md_atof): Don't declare atof_ieee.
102 (demand_empty_rest_of_line): Don't declare.
103 * config/tc-m88k.c (md_atof): Don't declare atof_ieee.
104 * config/tc-sparc.c (cmp_reg_entry): Change args to const PTR.
105 (parse_keyword_arg): Change lookup_fn to take const arg.
106 (md_atof): Don't declare atof_ieee.
107 * config/tc-sparc.h: Add ifdef for multiple inclusion.
108 (tc_aout_pre_write_hook): Don't declare.
112 * as.h (bfd_alloc_by_size_t): Don't declare.
113 * Many files: Use xmalloc rather than bfd_alloc_by_size_t.
117 * symbols.c (symbol_new): Don't call debug_verify_symchain.
118 (symbol_append): Set sy_next and sy_previous when adding a single
119 symbol to an empty list. Call debug_verify_symchain.
120 (verify_symbol_chain): Use assert, not know.
124 * NEWS: Note BeOS support.
125 * configure.in: (ppc-*-beos): New target, use coff as object format.
126 * configure: Regenerate with autoconf.
130 * config/tc-mips.c (md_apply_fix): Improve error message for out
133 * Makefile.in: Add dependencies on obstack.h where needed.
137 * config/tc-mips.c (md_estimate_size_before_relax): Handle the
138 case of a symbol equated to another symbol when using SVR4_PIC.
140 * Makefile.in (TARG_CPU_DEP_sparc): Add opcode/sparc.h.
144 * read.c (read_a_source_file): Call LISTING_NEWLINE before
145 HANDLE_CONDITIONAL_ASSEMBLY when handling an MRI line label.
147 * config/obj-elf.c (obj_elf_data): Call md_flush_pending_output
148 and md_elf_section_change_hook if they are defined.
149 (obj_elf_text, obj_elf_previous): Likewise.
153 * config/obj-multi.h (struct elf_obj_sy): Define if
155 (OBJ_SYMFIELD_TYPE): Define as struct elf_obj_sy if
157 * config/obj-elf.h (struct elf_obj_sy): Don't define if
158 OBJ_SYMFIELD_TYPE is defined.
160 * doc/as.texinfo (bss): Improve description of .bss section. In
161 ELF or COFF, you are permitted to switch into the section.
162 (Comm): Rewrite description of common symbols.
163 (Lcomm): Mention that some targets permit a third argument.
167 * config/tc-ppc.c (ppc_elf_lcomm): Don't call S_CLEAR_EXTERNAL.
169 * symbols.c (colon): Change type of local to int. From Alan Modra
172 * config/tc-m88k.c (m88k_do_align): Don't use a special nop
173 alignment if a zero fill pattern was explicitly specified.
174 * config/tc-sh.c (sh_do_align): Likewise.
176 * read.c (equals): Always permit register names to be redefined.
178 * config/tc-mips.c (mips_fix_adjustable): Permit a reloc against a
179 mips16 symbol to be adjusted if a symbol is being subtracted from
183 * config/obj-elf.c (obj_elf_symver): Check for duplicate or
184 illegal symbol version names.
185 (elf_frob_symbol): Check for external default versions.
190 * config/obj-elf.h (struct elf_obj_sy): Define.
191 (OBJ_SYMFIELD_TYPE): Define to elf_obj_sy struct. Change all
193 * config/obj-elf.c (obj_elf_symver): Just record the name.
194 (obj_symbol_new_hook): Initialized versioned_name field.
195 (elf_frob_symbol): If there is a versioned_name, either rename the
196 symbol, or add an alias with that name.
200 * config/tc-mn10300.c (md_relax_table): Define.
201 (md_convert_frag): Implement.
202 (md_assemble): Handle relaxable operands/instructions correctly.
203 (md_estimate_size_before_relax): Implement.
204 * config/tc-mn10300.h (TC_GENERIC_RELAX_TABLE): Define.
206 * config/tc-mn10200.c (md_relax_table): Fix typos.
208 * config/tc-mn10300.c (md_assemble): Don't use any MN10300 specific
209 relocs anymore. Tweak fx_offset for pc-relative relocs.
213 * cond.c (s_ifc): Call mri_comment_field and mri_comment_end when
219 * config/tc-tic80.c (md_pseudo_table): Add "sect" and "section"
221 * config/tc-tic80.c (md_begin): Declare external variable
222 coff_flags and insert an F_AR32WR bit into it.
227 * read.c (equals): Add reassign parameter. Change all callers.
228 * read.h (equals): Update declaration.
232 * config/tc-mips.c (mips16_extended_frag): Don't assume that we
233 can rely on the frag address to determine whether a frag is
238 * write.h (LOCAL_LABEL): Only define if not BFD_ASSEMBLER.
239 (S_LOCAL_NAME): Likewise.
240 (FAKE_LABEL_NAME): Define unconditionally.
241 * symbols.c (colon): Call bfd_is_local_label, not LOCAL_LABEL, if
243 (S_IS_LOCAL): Call bfd_is_local_label_name, not LOCAL_LABEL.
244 * config/tc-*.h: Only define LOCAL_LABEL if not BFD_ASSEMBLER.
245 Don't define FAKE_LABEL_NAME.
246 * config/te-ic960.h: Likewise.
247 * config/tc-mips.h (tc_frob_file_before_adjust): Define.
248 (mips_frob_file_before_adjust): Declare.
249 * config/tc-mips.c (mips_frob_file_before_adjust): New function.
250 (mips_local_label): Remove.
252 * config/te-sco386.h: Remove; not used.
257 * config/tc-tic80 (md_pseudo_table): Add align pseudo op to do
258 byte alignment rather than power-of-two alignment that is the
264 * config/tc-m68k.c (md_assemble): Handle a reloc width of 'W'.
266 * gasp.c (hash_add_to_string_table): Correct misspelling in error
267 message, and add newline.
268 (process_file): Don't process assignments in the label if this is
269 a equ or assign pseudo-op.
270 (process_pseudo_op): Swap first argument to do_assign for K_ASSIGN
271 and K_EQU, to match documentation.
275 * config/obj-coff.c (obj_coff_section): Add 'r' section attribute
276 to denote read-only data sections.
280 * config/obj-elf.c (obj_elf_common): Set BSF_OBJECT in flags.
281 * config/tc-sparc.c (s_common): Likewise, if BFD_ASSEMBLER.
283 * expr.c (operand): Simplify 0b handling. Don't treat 0b as a
284 binary number if the next character is '+' or '-'.
289 * config/tc-tic80.c (FLT_CHARS): Change from "dD" to "fF".
290 (find_opcode): Match operands that can be floats.
291 (build_insn): Handle O_big (float) expressions and build
297 * configure.in (mips*-*-lnews*): New target, also make empty
298 emulation list for this target.
300 * tc-mips.c (ECOFF_LITTLE_FORMAT): Define.
301 (mips_target_format): Use.
302 * te-lnews.h: New file.
307 * config/tc-tic80.c (find_opcode, build_insn): Changes to match
308 operands with :m or :s modifiers and generate the right opcodes
314 * Makefile.in (itbl-parse.c itbl-parse.h): Use $(BISON) and
315 $(BISONFLAGS), not $(YACC) and $(YACCFLAGS).
319 * config/tc-m68k.c (instring): Useless local declaration of
320 crack_operand removed.
321 * expr.h (expressionS): Changed type of X_op field to operatorT if
327 * configure.in: Add i386coff and i386elf to emulation list.
328 * configure: Rebuild.
329 * as.c (i386coff, i386elf): Declare.
330 * obj.h (coff_format_ops): Declare.
331 * config/obj-coff.c (OBJ_HEADER): Define.
332 (coff_obj_symbol_new_hook): Rename from obj_symbol_new_hook.
333 (coff_obj_read_begin_hook): Rename from obj_read_begin_hook.
334 (obj_pseudo_table): Add "version".
335 (coff_pop_insert): New static function.
336 (coff_sec_sym_ok_for_reloc): New static function.
337 (no_func): New static function.
338 (coff_format_ops): New variable.
339 * config/obj-coff.h (coff_obj_symbol_new_hook): Declare.
340 (obj_symbol_new_hook): Define.
341 (coff_obj_read_begin_hook): Declare.
342 (obj_read_begin_hook): Define.
343 * config/tc-i386.h (i386_target_format): Declare.
344 * config/tc-i386.c: Check OBJ_MAYBE_ELF as well as OBJ_ELF; check
345 OUTPUT_FLAVOR when appropriate.
346 (i386_target_format): New function.
347 * Makefile.in (obj-coff.o): New target.
348 (e-i386coff.o, e-i386elf.o): New targets.
351 * config/tc-i960.h (TC_SYMFIELD_TYPE): Define if OBJ_COFF.
352 (_tc_get_bal_of_call): Don't declare.
353 (tc_get_bal_of_call): Declare as function, don't define as macro.
354 * config/tc-i960.c (tc_set_bal_of_call): If OBJ_COFF, store balP
355 in sy_tc field, not x_balntry field.
356 (tc_get_bal_of_call): Rename from _tc_get_bal_of_call. Change
357 return type to symbolS *. If OBJ_COFF, retrieve value from sy_tc
358 field, not x_balntry field.
360 * config/obj-elf.c (obj_elf_section): Permit a .note section to
361 have the SHF_ALLOC attribute.
363 * Makefile.in ($(OBJS)): Don't depend upon $(IT_HDRS).
364 (TARG_CPU_DEP_mips): Depend upon $(srcdir)/itbl-ops.h.
365 (itbl-lex.o): Depend upon itbl-parse.h.
367 * itbl-parse.y (yyerror): Change return type to int. Change to
368 use old style function declaration.
370 * Makefile.in (itbl-lex.o): Remove -Wall.
371 (itbl-parse.o): Likewise.
373 * cond.c (s_ifdef): If we should omit conditionals from listings,
375 (s_if, s_ifc, s_endif, s_else, s_ifeqs): Likewise.
376 * listing.c (list_info_struct): Add EDICT_NOLIST_NEXT.
377 (listing_listing): Handle EDICT_NOLIST_NEXT.
378 (listing_list): An argument of 2 means EDICT_NOLIST_NEXT.
379 * listing.h (LISTING_NOCOND): Define.
380 (LISTING_SKIP_COND): Define.
381 * as.c (show_usage): Mention c as a suboption of -a.
382 (parse_args): Handle c as a suboption of -a.
383 * doc/as.texinfo: Document -alc.
388 * config/tc-tic80.c (md_apply_fix): Handle R_ABS type fixups.
393 * doc/as.texinfo: Document .symver.
397 * config/tc-d10v.c: Change pre_defined_registers to
398 d10v_predefined_registers and reg_name_cnt to d10v_reg_name_cnt.
402 * config/obj-coff.c: Fix typo in comment section.
404 * config/tc-tic80.c (md_pseudo_table): Add entry for bss, which takes
405 an additional alignment argument.
406 (find_opcode): Allow O_symbol relocs for any 32 bit field, not just
408 (build_insn): Handle O_symbol relocs for any 32 bit field, not just
414 * Makefile.in: Remove dependancies on itbl-cpu.h.
415 * as.c: Define stubs for itbl_parse and itbl_init if HAVE_ITBL_CPU
420 * itbl-ops.h: Include as.h.
424 * as.c: Remove -t option.
425 * configure, configure.in: Move itbl-cpu.h to mips specific configure.
426 * itbl-ops.h: Include itbl-cpu.h only if HAVE_ITBL_CPU is defined.
427 * config/tc-mips.h: Define HAVE_ITBL_CPU.
431 * itbl-ops.c: Don't define DEBUG.
435 * Makefile.in: Update itbl-test.c to reflect its new location.
439 * itbl-ops.c: Add test for itbl_have_entries.
440 * config/tc-mips.c: Remove test for itbl_have_entries.
441 * config/tc-mips.h: Define tc_init_after_args to mips_init_after_args.
445 * Makefile.in (DISTSTUFF): Remove itbl-parse.y, itbl-lex.l, and
446 itbl-ops.c. Add itbl-parse.c and itbl-lex.c.
447 (LEX, LEXFLAGS): Define.
448 * itbl-ops.c (append_insns_as_macros): Remove bogus ASSERT.
452 * itbl-parse.y: Fix indentation mistakes from indent program.
453 * itbl-lex.l: Fix indentation mistakes from indent program.
454 * itbl-ops.h: Add include for ansidecl.h.
455 Add PARAMS around function arguments.
456 Add declaration for itbl_have_entries.
457 * itbl-ops.c: Add PARAMS around function arguments.
458 * Makefile.in: Add itbl build rules.
459 Add dependancies for itbl files to mips target.
460 * as.c: Add itbl support.
461 Add new option "--insttbl" for dynamically extending instruction set.
462 * as.h: Declare insttbl_file_name;
463 the name of file defining extensions to the basic instruction set
464 * configure.in, configure: Add itbl-parse.o, itbl-lex.o, and
465 itbl-ops.o to extra_objects for mips configuration.
466 Add include file link from itbl-cpu.h to
467 config/itbl-${target_cpu_type}.h.
468 * config/tc-mips.c: Allow copz instructions.
469 Add notes for future additions to the itbl support.
471 (macro): Call itbl_assemble to assemble itbl instructions.
472 See if an unknown register is specified in an itbl entry.
475 * doc/internals.texi (CPU backend): Fix typo in md_section_align
478 * config/tc-tic80.h (NEED_FX_R_TYPE): Define.
479 * config/tc-tic80.c (find_opcode): Add code to support O_symbol
481 (build_insn): Grab a frag early so we can use the address in
482 fixups. Take one's complement of BITNUM values before insertion
483 in opcode. Add code to support O_symbol operands.
484 (md_apply_fix): Replace unimplemented warning with implementation.
485 (md_pcrel_from): Ditto.
486 (tc_coff_fix2rtype): Ditto.
491 * config/tc-d30v.c (parallel_ok): New function.
492 * config/tc-d30v.h: Define TARGET_BYTES_BIG_ENDIAN.
494 * config/tc-d10v.c (md_pcrel_from_section): Return 0 if
495 relocation is in different section. Fixes PR11574.
499 * tc-mips.c (mips_ip): If configured for an embedded ELF system,
500 don't set the section alignment to 2**4.
504 * config/tc-m68k.c (line_comment_chars): Add '*'.
506 * app.c (LEX_IS_TWOCHAR_COMMENT_2ND): Don't define.
507 (do_scrub_begin): Don't set lex['*'].
508 (do_scrub_chars): When handling LEX_IS_TWOCHAR_COMMENT_1ST, don't
509 check for LEX_IS_TWOCHAR_COMMENT_2ND. Instead, just check for
512 * configure.in: Set em=svr4 for m68k-*-sysv4*.
513 * configure: Rebuild.
514 * config/te-svr4.h: New file.
515 * config/tc-m68k.c (m68k_comment_chars): Only include `#' if
520 * config/tc-mn10200.c (md_convert_frag): Create a fixup for the
521 short conditional branch around a long unconditional branch.
525 * config/obj-coff.c (obj_coff_ln [both versions]): Call
528 * config/tc-arm.c (fix_new_arm): Use make_expr_symbol to handle a
531 * symbols.c (resolve_symbol_value): If both left and right
532 operands are undefined, warn about both of them.
537 * config/obj-elf.c (elf_pseudo_table): Add "symver".
538 (obj_elf_symver): New static function.
539 * config/obj-elf.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Copy the st_other
542 * write.c (relax_segment): Make type and printf format agree.
544 * read.c (get_line_sb): Don't end the line on a semicolon inside a
550 * config/tc-d30v.c, config/tc-d30v.h: New files.
552 * configure: Rebuilt.
554 * configure.in: Add case for d30v.
559 * config/tc-alpha.h (md_operand): Define with a null expansion,
560 like all the other targets.
561 * doc/internals.texi (CPU backend): Add missing word in
562 md_flush_pending_output description. Fix typo in md_convert_frag
565 * config/tc-tic80: Minor comment additions/changes.
570 * config/tc-m68k.c (LOCAL_LABEL): Macro redefined if TE_DELTA.
571 (tc_canonicalize_symbol_name): Macro defined if TE_DELTA.
572 * config/obj-coff.c (obj_coff_def): Use
573 tc_canonicalize_symbol_name if defined.
574 (obj_coff_tag, obj_coff_val): Likewise.
575 * expr.c (operand): Reject '~' as operator if is_name_beginner.
580 does not actually work, though:
581 * configure.in (i386-sequent-bsd*): New target.
582 * configure: Rebuild.
583 * config/tc-dynix.h: New file.
584 * config/tc-i386.h: Define TARGET_FORMAT if TE_DYNIX.
586 * read.c (do_align): Add max parameter. Change all callers.
587 Remove useless static variables.
588 (s_align): New static function. Do common portion of
589 s_align_bytes and s_align_ptwo.
590 (s_align_bytes, s_align_ptwo): Just call s_align.
591 * frags.c (frag_align): Add max parameter. Change all callers.
592 (frag_align_pattern): Likewise.
593 * frags.h (frag_align, frag_align_pattern): Update declarations.
594 * write.c (relax_segment): Limit alignment change to fr_subtype.
595 Fix some types to be addressT.
596 * config/obj-coff.c (size_section): Likewise.
597 * config/obj-ieee.c (size_section): Likewise.
598 * config/tc-d10v.h (md_do_align): Add max parameter.
599 * config/tc-i386.h (md_do_align): Likewise.
600 * config/tc-m88k.h (md_do_align): Likewise.
601 * config/tc-m88k.c (m88k_do_align): Likewise.
602 * config/tc-sh.h (md_do_align): Likewise.
603 * config/tc-sh.c (sh_do_align): Likewise.
604 * as.h: Improve comments on rs_align and rs_align_code.
605 * doc/as.texinfo: Document new alignment arguments.
606 * doc/internals.texi (Frags): Document use of fr_subtype field for
607 rs_align and rs_align_code.
611 * config/tc-mips.c: Changed opcode parsing.
615 * config/{tc-alpha.h, tc-arc.h, tc-d10v.h, tc-generic.h, tc-i960.h,
616 tc-mn10200.h, tc-mn10300.h, tc-sh.h, tc-v850.h, tc-vax.h, tc-w65.h}:
617 Add default definition of zero for TARGET_BYTES_BIG_ENDIAN.
618 * config/{tc-arm.h, tc-hppa.h, tc-i386.h, tc-mips.h, tc-ns32k.h,
619 tc-ppc.h, tc-sparc.h}: Move definition of TARGET_BYTES_BIG_ENDIAN
620 to a location consistent with the rest of the target include files.
621 * config/tc-i386.c: Remove misleading comment.
622 * doc/internals.texi (CPU backend): Add description of function
625 * config/tc-tic80.c: Add code to insert predefined symbols into the
626 symbol table so they can be parsed by the standard expression parser.
627 Remove custom code that use to parse them.
628 * config/tc-tic80.h: Move definition of TARGET_BYTES_BIG_ENDIAN
629 to a location consistent with the rest of the target include files.
634 * as.h: GNU c provides unlink() function.
636 Unify section handling on openVMS/Alpha:
637 * config/tc-alpha.c(s_alpha_link): Remove.
638 (s_alpha_section): New function.
639 Remove case-hacking of symbols
640 Add .code_address pseudo-op.
641 (BFD_RELOC_ALPHA_CODEADDR): New relocation.
642 (s_alpha_code_address): New function.
643 (alpha_ctors_section, alpha_dtors_section): New sections for C++
644 static constructors/destructors.
645 Add debug code for crash debugs, to be removed when traceback code
646 is added to object code.
647 (s_alpha_name): New function for .name pseudo-op.
648 (alpha_print_token): New function to print token expressions with
649 alpha specific extensions.
651 * makefile.vms: Allow compilation with current gcc snapshot.
655 * doc/Makefile.in (TEXI2DVI): Set to just name of program.
656 (DVIPS): Set to dvips.
657 (ps, as.ps, gasp.ps): New targets.
658 (internals.info, gasp.dvi, internals.dvi): Set both TEXINPUTS
659 and MAKEINFO env variables.
660 (internals.ps): Use DVIPS macro.
661 (clean): Remove core and backup files.
662 (distclean): Remove temporary files from building internals.
664 * doc/internals.texi (Frags): Fix typo.
665 (GAS processing): Ditto.
666 (CPU backend): Ditto.
667 * ecoff.c (init_file): Use TARGET_BYTES_BIG_ENDIAN value directly.
668 * mpw-config.in: Define TARGET_BYTES_BIG_ENDIAN as 1.
669 * read.c: Remove ugly hack that dealt with config files not
670 correctly defining TARGET_BYTES_BIG_ENDIAN.
671 (target_big_endian): Use TARGET_BYTES_BIG_ENDIAN directly.
672 * config/arm-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
673 * config/arm-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
674 * config/mips-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
675 * config/mips-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
676 * config/ppc-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
677 * config/ppc-sol.mt: Replace TARGET_BYTES_LITTLE_ENDIAN
678 with TARGET_BYTES_BIG_ENDIAN defined to 0.
679 * config/tc-arm.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
680 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
681 whether it is defined or not.
682 * config/tc-mips.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN.
683 * config/tc-ppc.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
684 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
685 whether it is defined or not.
687 * config/tic80.h (TARGET_FORMAT): Define to coff-tic80.
688 (TARGET_BYTES_BIG_ENDIAN): Define to 0.
693 * write.c (write_relocs): Correct text in as_fatal error message,
694 bfd_perform_relocation -> bfd_install_relocation.
698 * config/tc-m68k.c (LEX_TILDE): Define if TE_DELTA.
699 * read.c (LEX_TILDE): Define if not defined.
700 (lex_type): Use LEX_TILDE.
701 * expr.c (get_symbol_end): Check first char with is_name_beginner,
706 * config/tc-sparc.c (md_show_usage): Add missing backslash at end
709 * config/tc-mips.c (mips16_extended_frag): Correct base address
710 for an extended PC relative instruction.
711 (md_convert_frag): Likewise.
713 * config/tc-mips.c (prev_nop_frag): New static variable.
714 (prev_nop_frag_holds): New static variable.
715 (prev_nop_frag_required): New static variable.
716 (prev_nop_frag_since): New static variable.
717 (append_insn): If we aren't reordering, and prev_nop_frag is not
718 NULL, and we don't need any nops, then decrease the size of
719 prev_nop_frag. Don't insert nops because of instructions in
720 noreorder sections. Remember whether the previous instructions
721 where in noreorder sections even when not reordering.
722 (mips_no_prev_insn): Add preserver parameter. Change all
723 callers. Refer prev_nop_frag variables when appropriate.
724 (mips_emit_delays): Set up prev_nop_frag.
725 (s_mipsset): Clear prev_nop_frag if reordering.
729 * config/tc-mips.c (append_insn): Remove useless code which
730 handled swapping a mips16 jump with a mips16 instruction with a
733 * config/tc-mips.c (md_parse_option): When debugging, set
734 mips_optimize to 1, not 0.
736 * config/tc-mips.c (mips16_ip): Handle an extend operand.
738 * config/tc-mips.c (my_getExpression): In mips16 mode, if it looks
739 like the expression was based on `.', adjust the value of the
742 * config/tc-mips.c (append_insn): Warn about an attempt to put an
743 extended instruction in a delay slot when not reordering.
744 (md_convert_frag): Warn if an extended instruction appears in a
747 * config/tc-mips.c (mips_pseudo_table): Add "insn".
748 (s_insn): New static function.
749 * doc/c-mips.texi: Document .insn.
751 * config/tc-mips.c (md_begin): Add the general registers to the
753 (mips16_ip): First parse the expression, and then see whether it
754 came up with a register, rather than trying to first see whether
755 we are looking at a register.
760 * config/tc-tic80.c: Numerous changes and additions to flesh
761 out functions that were previously just stubs, and fix some
762 problems found using the new TIc80 testsuite cases.
767 * config/tc-mips.c (mips16_ip): Handle %gprel modifier.
768 (md_apply_fix): Handle BFD_RELOC_MIPS16_GPREL.
770 * config/tc-mips.c (append_insn): Output jump instruction as a
771 pair of 2 byte instructions, rather than as a single 4 byte
775 * config/tc-mips.c (md_begin): Correct test of mips_5900.
776 (mips_ip): Don't check INSN_ISA for a macro.
782 * config/tc-mips.c: added r5900.
787 * itbl-ops.c, itbl-lex.l, itbl-parse.y, itbl-ops.h,
788 config/itbl-mips.h: Add copyright message and fix indentation.
792 * itbl-ops.c: New file. Add support for dynamically read
793 instruction registers, opcodes and formats. Build internal table
794 for new instructions and provide callbacks for assembler and
796 * itbl-lex.l, itbl-parse.y: Lex and yacc parsers for instruction
798 * itbl-ops.h: New file. Header file for itbl support.
799 * config/itbl-mips.h: New file. Mips specific definitions for
804 * config/tc-mn10200.c (md_assemble): If a constant operand won't
805 fit into the constant field of a relaxable operand, then it does
810 * config/tc-mn10200.c (md_estimate_size_before_relax): Treat
811 a jsr target in a different section just like a jsr to
816 * config/tc-mips.c (mips_fix_adjustable): Don't adjust relocations
817 against any mips16 symbols, not just externally visible ones.
818 (md_apply_fix): Corresponding change.
822 * config/tc-mips.c (mips16_ip): Accept floating point registers in
823 the operand of the exit instruction.
827 * symbols.c (resolve_symbol_value): If we leave an equated symbol
828 as O_symbol, copy over the segment.
832 * config/tc-mips.c (md_apply_fix): If we aren't adjusting this
833 fixup to be against the section symbol, adjust the value
836 * symbols.c (resolve_symbol_value): Don't change X_add_number for
838 * write.c (write_relocs): Avoid looping on equated symbols.
839 Adjust fx_offset by X_add_number for each symbol.
840 * config/obj-coff.c (do_relocs_for): Avoid looping on equated
842 (fixup_segment): Add a loop to track down equated symbols and
843 adjust fx_offset appropriately.
847 * config/tc-mn10200.c (md_relax_table): Add entries to allow
849 (md_convert_frag): Handle jmp->bra relaxing.
850 (md_assemble): Handle jmp->bra relaxing.
851 (md_estimate_size_before_relax): Likewise.
855 * config/tc-i386.c (i386_align_code): Add comments explaining the
860 * config/tc-sparc.c (enforce_aligned_data): New static variable.
861 (sparc_cons_align): Don't do anything unless enforce_aligned_data
863 (md_longopts): Add "enforce-aligned-data".
864 (md_show_usage): Mention --enforce-aligned-data.
865 * doc/c-sparc.texi (Sparc-Aligned-Data): New node; document
866 enforce-aligned-data.
868 * config/tc-ppc.c (md_pseudo_table): If OBJ_XCOFF, add "long",
870 (ppc_xcoff_cons): New static function.
872 * write.c (relax_segment): Give an error if a .space symbol is
875 * read.c (read_a_source_file): Don't handle mri_pending_align if
876 the handler is s_globl or s_ignore.
880 * config/tc-d10v.c (find_opcode): Remove unused variable "numops".
882 * config/tc-tic80.c: Many additions to previous placeholder file.
883 * config/tc-tic80.h: Ditto.
888 * config/tc-i386.c (i386_align_code): Improve the nop patterns.
892 * config/tc-mips.c (mips_fix_adjustable): New function.
893 * config/tc-mips.h (tc_fix_adjustable): Call mips_fix_adjustable.
894 (mips_fix_adjustable): Declare.
897 * read.c (read_a_source_file): After doing an mri_pending_align,
898 adjust the line_label if there is one.
899 (s_space): Set mri_pending_align if an odd number of bytes were
904 * config/tc-d10v.h (md_do_align): Add this hook to call
905 d10v_cleanup() when a ".align" is detected. Fixes PR11487.
907 * config/tc-d10v.c (find_opcode): Correctly calculate
908 branch displacement when .aligns are present.
912 * config/tc-mn10200.c (md_relax_table): Define.
913 (md_convert_frag): Implement.
914 (md_assemble): Handle relaxable operands/instructions correctly.
915 (md_estimate_size_before_relax): Implement.
916 * config/tc-mn10200.h (TC_GENERIC_RELAX_TABLE): Define.
920 * config/tc-mips.c (append_insn): Give an error for jumps to a
922 (md_apply_fix): Make a branch to an odd address an error rather
925 * config/tc-mips.c (md_convert_frag): If the user explicitly
926 requested an extended opcode, pass warn as true to mips16_immed.
928 * config/tc-mips.c (mips16_ip): Handle a missing expression like
929 an explicit 0, so that explicitly extended instructions work
934 * ecoff.c (ecoff_build_symbols): Don't generate a local ECOFF
935 symbol for a common symbol.
939 Patch presumed to have been checked in awhile ago but wasn't.
941 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
942 (relax_segment, case rs_org): Move code inside braces. Move locals
943 target,after inside too.
944 (relax_segment, case rs_machine_dependent): Guts moved to ...
945 (relax_frag): New function.
946 Call md_prepare_relax_scan if defined.
950 * config/tc-m68k.c (m68k_ip): Reject pc-relative addresses for the
951 'p' operand specifier.
955 * config/tc-m68k.c (HAVE_LONG_BRANCH): New macro, returns true for
956 m68k family cpus which support long branch addressing modes.
957 (m68k_ip, md_convert_frag_1, md_estimate_size_before_relax,
958 md_create_long_jump): Use it.
962 * config/tc-mips.c (md_begin): Don't set SEC_ALLOC or SEC_LOAD for
963 the .reginfo or .MIPS.options section if configured for an
966 * config/tc-mips.c (md_begin): Don't set interlocks for
971 * read.c (read_a_source_file): Make sure the symbol ends with
972 whitespace before checking whether the next character is '='.
976 * config/tc-i386.c (sco_id): Moved from here...
977 * config/obj-elf.c (sco_id): ...to here. Adding the identifier
978 really is an SCO ELF specific thing, not just a SCO x86 specific
984 * configure.in (tic80-*-*): Don't require 'coff'.
985 * configure: Regenerate.
990 * read.c (emit_expr): Check for overflow of a negative value
992 * write.c (fixup_segment): Likewise.
993 * config/obj-coff.c (fixup_segment): Likewise.
995 * config/tc-m68k.c (struct label_line): Define.
996 (labels, current_label): New static variables.
997 (md_assemble): Mark current_label as text, and clear it.
998 (m68k_frob_label): New function.
999 (m68k_flush_pending_output): New function.
1000 (m68k_frob_symbol): New function.
1001 * config/tc-m68k.h (tc_frob_label): Define.
1002 (md_flush_pending_output): Define.
1003 (tc_frob_symbol): Don't warn, just call m68k_frob_symbol.
1004 (tc_frob_coff_symbol): Likewise.
1006 * read.c (read_a_source_file): When defining a macro in MRI mode,
1007 don't add the symbol to the symbol table.
1011 * config/tc-mn10300.c (tc_gen_reloc): Handle sym1-sym2 fixups
1012 here since fixup_segment doesn't (linkrelax is set).
1013 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
1017 * config/tc-mn10200.c (md_assemble): Tweak fx_offset for pc-relative
1022 * config/tc-hppa.c (struct hppa_fix_struct): Tweak fx_r_field's type
1023 to avoid warnings with the native HP compiler.
1024 (fix_new_hppa): Similarly for the r_type argument.
1025 (pa_build_unwind_subspace, hppa_elf_mark_end_of_function): Enclose
1026 in an #if OBJ_ELF to keep gcc -Wall quiet.
1027 (md_apply_fix): Always initialize "result".
1029 * config/tc-mn10200.c (md_assemble): Generate relocations.
1033 * config/tc-m68k.c (s_even): Adjust the alignment of the current
1038 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
1039 pass on the new file hook.
1041 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
1042 !alpha_force_relocation, as local LITERALs can be adjusted to be
1043 relative to the section.
1047 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
1049 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
1050 aligned to odd boundary.
1051 (tc_frob_coff_symbol): Define.
1053 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
1055 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
1057 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
1060 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
1062 (md_assemble): If the size is 'B', set fx_signed.
1063 (md_apply_fix_2): Use fx_signed when checking for overflow.
1065 * write.h (struct fix): Add fx_signed field.
1066 * write.c (fix_new_internal): Initialize fx_no_overflow and
1068 (fixup_segment): Use fx_signed when checking for overflow.
1069 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
1070 fx_signed when checking for overflow.
1074 * NOTES, NOTES.config: Removed. These are rarely, if ever,
1075 updated, and all the useful information is in doc/internals.texi.
1078 * read.c (read_a_source_file): Check for conditional operators
1079 before doing an MRI pending alignment.
1080 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
1081 (tc_conditional_pseudop): Define.
1082 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
1083 * doc/internals.texi (CPU backend): Describe
1084 tc_conditional_pseudoop.
1087 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
1089 (md_begin): Check m68k_rel32_from_cmdline before setting
1091 (m68k_mri_mode_change): Likewise.
1092 (md_longopts): Add --disp-size-default-16 and
1093 --disp-size-default-32.
1094 (md_parse_option): Handle new options.
1095 (md_show_usage): Mention new options.
1096 * doc/c-m68k.texi (M68K-Opts): Document new options.
1099 * config/tc-m68k.c (m68k_index_width_default): New static
1101 (m68k_ip): Use m68k_index_width_default to set the size of a base
1102 register whose size was not given.
1103 (md_longopts): Add --base-size-default-16 and
1104 --base-size-default-32.
1105 (md_parse_option): Handle new options.
1106 (md_show_usage): Mention new options.
1107 * doc/c-m68k.texi (M68K-Opts): Document new options.
1109 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
1111 * configure.in: Recognize mips-*-linux* target.
1112 * configure: Rebuild.
1114 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
1115 work if valueT is only 32 bits.
1117 * config/tc-mips.c: Throughout, check target_big_endian rather
1119 (byte_order): Remove.
1120 (mips_init_after_args): Remove.
1121 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
1122 (mips_init_after_args): Don't declare.
1123 (tc_init_after_args): Don't define.
1125 * config/tc-mips.h (tc_frob_after_relocs): Define if
1127 (mips_elf_final_processing): Likewise.
1128 (ELF_TC_SPECIAL_SECTIONS): Likewise.
1133 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
1134 sld.{h,w} and the operand is relocatable, adjust the adend by
1135 shifting it right one bit.
1140 * read.c (read_a_source_file): Check mri_pending_align after
1141 checking for a macro. From Ronald F. Guilmette
1144 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
1146 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
1147 add segment argument. If OBJ_ELF, treat a relocation against a
1148 symbol in a linkonce section like a relocation against an external
1150 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
1154 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
1158 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
1159 includes config.h instead of host.h, tc.h instead of tp.h, and
1160 targ-env.h instead of target-environment.h.
1161 Also, obj-format.h includes targ-cpu.h instead of
1163 start-sanitize-tic80
1164 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
1165 * configure: Rebuild with autoconf.
1166 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
1167 (TARGET_FORMAT): Define to "coff-tic80".
1168 * config/tc-tic80.c: New file for TIc80 support.
1169 * config/tc-tic80.h: New file for TIc80 support.
1174 * doc/as.texinfo (M): Mention explicitly that -M changes macro
1179 * write.c (adjust_reloc_syms): If the fixup symbol has been
1180 equated to an undefined symbol, convert the fixup to being against
1181 the target symbol. Remove obsolete code handling a special case
1186 * mpw-make.sed: Use NewFolderRecursive for installation.
1190 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
1194 * config/tc-d10v.c (md_assemble): Fix bug which caused
1195 second instruction in a line to be case sensitize. PR11312
1199 * config/tc-mn10200.c (mn10200_insert_operand): Don't
1200 range check operands with MN10200_OPERAND_NOCHECK set.
1201 (check_operand): Likewise.
1205 * config/tc-mips.c: Undo part of last Friday's alignment changes.
1206 (md_begin): Always align the text section to a four byte
1208 (append_insn): Remove call to record_align.
1210 * config/tc-mips.c (insn_label): Remove.
1211 (struct insn_label_list): Define.
1212 (insn_labels, free_insn_labels): New static variables.
1213 (mips_clear_insn_labels): New static function.
1214 (append_insn): Mark all mips16 text labels, and make them odd.
1215 Handle all labels after emitting a nop, not just one. Call
1216 mips_clear_insn_labels rather than just clearing insn_label.
1217 (mips_emit_delays): Add insns parameter, and use it to decide
1218 whether to mark mips16 labels. Handle all labels, not just one.
1219 Force mips16 labels to be odd. Change all callers.
1220 (mips16_immed): Don't check for an odd branch target.
1221 (md_apply_fix): Don't check mips16 mode for a branch reloc.
1222 (mips16_extended_frag): Ignore the low bit in a branch target.
1223 (md_convert_frag): Likewise.
1224 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
1225 clearing insn_label.
1226 (mips_align, mips_flush_pending_output, s_cons): Likewise.
1227 (s_float_cons, s_gpword): Likewise.
1228 (s_align): Use insn_labels rather than insn_label.
1229 (s_cons, s_float_cons, s_gpword): Likewise.
1230 (mips_frob_file_after_relocs): New function.
1231 (mips_define_label): Rewrite to add to insn_labels list.
1232 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
1233 * ecoff.c (ecoff_build_symbols): If the size of a function comes
1234 out odd, increment it.
1236 * config/tc-mips.c (append_insn): Only update prev_insn when not
1237 reordering if place is NULL.
1239 * config/tc-mips.c (mips16_ip): Check for a missing expression
1240 when using the register indirect addressing mode.
1244 * config/tc-mn10200.c (mn10200_insert_operand): Don't
1245 check 24bit operands for overflow.
1246 (check_operand): Likewise.
1250 * doc/as.texinfo (Section): Document how to use the .section
1251 pseudo-op for COFF and ELF.
1255 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
1259 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
1260 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
1261 store them. Adjust other RELAX_MIPS16 macros.
1262 (RELAX_MIPS16_DSLOT): Define.
1263 (RELAX_MIPS16_JAL_DSLOT): Define.
1264 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
1265 handling of whether previous instruction has a fixup. Set
1266 prev_insn_reloc_type.
1267 (mips_no_prev_insn): Clear prev_insn_reloc_type.
1268 (mips16_extended_frag): Use the right base address for a PC
1269 relative add or load.
1270 (md_convert_frag): Likewise. If a PC relative add or load is
1271 used, record the alignment for the section.
1275 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
1276 linkonce section into a reloc against the section symbol.
1278 * config/tc-mips.c (mips16_macro): Remove nop instructions after
1279 branch instructions.
1281 * config/tc-mips.c (md_begin): If configured for an embedded ELF
1282 system, don't set the section alignment to 2**4.
1283 (s_change_sec): Likewise.
1284 (append_insn): Call record_alignment for the section.
1285 (md_section_align): Don't align the section size for an embedded
1290 * write.c (adjust_reloc_syms): Make sure that symbols are
1291 resolved; expression symbols may have been skipped.
1292 * config/obj-coff.c (fixup_segment): Likewise.
1296 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
1297 BFD_RELOC_24_PLT_PCREL relocation.
1298 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
1302 * config/tc-d10v.c (write_2_short): Remove code that called
1303 parallel_ok() when the programmer specified parallel instructions.
1307 * config/tc-mn10300.c (md_assemble): Update to handle endianness
1310 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
1311 * config/tc-mn10300.c (md_assemble): Likewise.
1315 * config/tc-mips.c (append_insn): Make sure there is enough room
1316 in a frag after a mips16 instruction to switch it with a jump
1319 * config/tc-mips.c (mips16_extended_frag): Give an error for an
1320 attempt to use a non absolute symbol in an extending frag.
1324 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
1325 * config/tc-mn10200.h: Likewise.
1329 * app.c (do_scrub_chars): At the end of a C comment, pass space to
1330 UNGET rather than PUT. Set old_state before setting state to -2.
1332 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
1333 when extending because the value is exactly maxtiny + 1.
1335 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
1336 arguments, and store them. Adjust other RELAX_MIPS16 macros.
1337 (RELAX_MIPS16_USER_SMALL): Define.
1338 (RELAX_MIPS16_USER_EXT): Define.
1339 (mips16_small, mips16_ext): New static variables.
1340 (append_insn): Pass mips16_small and mips16_ext to
1341 RELAX_MIPS16_ENCODE.
1342 (mips16_ip): Set mips16_small and mips16_ext.
1343 (mips16_immed): Don't check mips16_autoextend.
1344 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
1346 * write.c (write_relocs): Print an error for an out of range
1347 fixup, rather than calling abort.
1349 * as.c (main): Unlink the output file if there are errors while
1350 generating the fixups.
1354 * config/tc-mips.c (mips16_extended_frag): Don't call
1356 (md_convert_frag): Call resolve_symbol_value before calling
1357 S_GET_VALUE, and don't add in the frag address.
1359 * config/tc-mips.c (mips16_immed): Add file and line parameters,
1360 and use them when reporting errors. Change all callers.
1364 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
1365 Remove '$' prefixing for registers.
1369 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
1370 is initialized before using it.
1371 (d10v_cleanup): No longer uses its argument, so make it void.
1373 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1377 * write.c (fixup_segment): Don't discard the symbol for a PC
1378 relative fixup to an absolute symbol.
1382 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
1383 with multiple sections.
1387 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
1388 no-mips-16 to no-mips16.
1389 (s_mipsset): Accept .set mips16 and .set nomips16.
1393 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
1394 argument, and check for +/- constant following the suffix, folding
1395 it into the expression.
1396 (ppc_elf_cons): Change ppc_elf_suffix calls.
1397 (md_assemble): Ditto.
1398 (shlib): Replace boolean mrelocatable with enumeration shlib.
1399 (md_parse_option): Discriminate between PIC style shared libraries
1401 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
1406 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
1407 change the comment characters.
1409 * config/tc-ppc.c (comment_chars): Delete in favor of
1411 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
1413 (ppc_comment_chars): Select appropriate comment chars by default.
1414 (msolaris): New flag for -m{,no-}solaris.
1415 (md_parse_option): Recognize -K pic. Add support for
1417 (md_show_usage): Update.
1418 (md_begin): Do not set ELF flags if Solaris.
1419 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
1420 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
1424 * as.c (main): Correct handling of flag_always_generate_output.
1428 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
1429 fx_offset, not fx_addnumber.
1431 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
1436 * config/tc-i386.c: Correct misspelling: balenced to balanced.
1440 * config/tc-mips.c (md_section_align): Check for an alignment of
1441 4, not an alignment of 16. Corrects August 7 patch.
1445 * configure, conf.in: Rebuild with autoconf 2.12.
1447 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
1448 alignment is specified.
1450 Add support for mips16 (16 bit MIPS implementation):
1451 * config/tc-mips.c: Extensive additions for mips16 support, not
1453 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
1454 (insn_uses_reg): Change last parameter to an enum.
1455 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
1456 (md_relax_frag): Define as mips_relax_frag.
1457 (mips_relax_frag): Declare.
1458 (struct mips_cl_insn): Add use_extend and extend fields.
1459 (tc_fix_adjustable): Define.
1460 * config/obj-elf.h (S_GET_OTHER): Define.
1461 (S_SET_OTHER): Define.
1465 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
1466 o and p operand specifiers.
1470 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
1471 (relax_segment, case rs_org): Move code inside braces. Move locals
1472 target,after inside too.
1473 (relax_segment, case rs_machine_dependent): Guts moved to ...
1474 (relax_frag): New function.
1475 Call md_prepare_relax_scan if defined.
1476 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
1481 * config/tc-mn10300.c (address_registers): Use '$' as register
1482 prefix instead of '%'.
1483 (data_registers, other_registers, md_assemble): Likewise.
1485 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
1486 (data_registers, other_registers, md_assemble): Likewise.
1488 * config/tc-mn10300.c (md_assemble): Correctly determine the
1489 correct location and type for each relocation.
1490 (md_pcrel_from): Simplify.
1494 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
1495 converted into branch around branch.
1499 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
1500 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
1502 * config/tc-mn10300.c (md_begin): Set linkrelax.
1503 (md_assemble): Create fixups as needed.
1504 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
1508 * config/tc-d10v.c (parallel_ok): When automatically converting
1509 serial ops to parallel, do not consider a branch as the first
1514 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
1518 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
1520 (check_operand): Likewise.
1524 * config/tc-d10v.c (parallel_ok): Branch and link instructions
1526 (write_2_short): Call parallel_ok to check whether two short
1527 instructions the user requested execute in parallel, can be
1532 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
1533 allow a branch and link in parallel with an exe instruction.
1537 * doc/c-d10v.texi: Add info on @word modifier.
1541 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
1542 operands are assumed to be 32bits. Use "bits" field to hold the
1543 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
1544 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
1547 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
1548 of a MN10300_OPERAND_SPLIT operand by operand->shift.
1550 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1551 MN10300_OPERAND_SPLIT.
1555 * config/tc-mn10300.c (md_assemble): Insert operands into
1556 the extension part of the instruction if necessary.
1557 (mn10300_insert_operand): Accept pointer to extension word
1558 argument. Make insn a pointer argument too. Return type
1559 is now void. All callers changed.
1563 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1564 repeated register operands.
1568 * doc/as.texinfo: Added section on reporting bugs.
1570 * config/tc-alpha.c: Change uses of void * to PTR. Change the
1571 alpha_macro emit field to expect a const argument, and change the
1572 arg field to be const. Fix some spacing to follow the GNU
1577 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
1578 (pca56) and 21264 (ev6) cpus.
1579 (md_apply_fix): Private relocation types are now negative.
1580 (alpha_force_relocation): Likewise.
1581 (tc_gen_reloc): Likewise.
1582 (emit_insn): Likewise.
1583 (emit_ldXu): Do the right thing when the hardware can do byte insns.
1584 (emit_stX): Likewise.
1585 (emit_sextX): Likewise.
1589 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
1590 a symbol found in a reloc.
1592 * symbols.c (resolve_symbol_value): Improve the error message if
1593 an undefined symbol is used in an expression.
1597 * doc/internals.texi: Rewrite, and add a lot of documentation.
1598 * doc/Makefile.in (internals.info): New target.
1603 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
1604 against weak symbols.
1608 * config/tc-v850.c (md_assemble): Don't lose for relaxable
1611 * config/tc-v850.c (md_convert_frag): Make sure we insert the
1612 fixup at the right address within the frag.
1614 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
1615 to an absolute value, instead increment it as needed.
1617 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
1618 * config/tc-v850.c: Fix some indention problems.
1619 (md_relax_table): Define for D9->D99 branch displacement
1621 (md_convert_frag): Do something useful instead of aborting.
1622 (md_estimate_size_before_relax): Likewise.
1623 (md_assemble): Note if the matching instruction has a relaxable
1624 operand. If it does, allocate frag with frag_var and don't
1630 * config/tc-d10v.h (md_cleanup): New function. This is needed to
1631 write out any buffered instructions when a ".end" is found.
1635 * read.c (read_a_source_file): New hook md_cleanup().
1639 * write.c (fix_new_exp): Use make_expr_symbol to build an
1640 expression symbol for a complex fixup.
1645 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
1646 hacks to improve parsing of complex hi, lo, zda, etc
1648 (md_assemble): Don't demand and eat a trailing ')' after finding
1649 a v850 relocation prefix. Sign extend the constant in a
1650 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
1652 (parse_cons_expression_v850): Don't eat a trailing ')' after
1653 finding a v850 relocation prefix.
1655 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
1656 (TC_CONS_FIX_NEW): Likewise.
1657 * config/tc-v850.c (parse_cons_expression_v850): New function.
1658 (cons_fix_new_v850): Likewise.
1660 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
1665 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
1666 for an unknown relocation type.
1671 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
1676 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
1677 differently for movea & sst/sld insns.
1682 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
1687 * obj.h (struct format_ops): Add frob_file_after_relocs field.
1688 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
1689 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
1690 frob_file_after_relocs field.
1691 * config/obj-elf.c (elf_format_ops): Likewise.
1692 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
1693 including obj-elf.h.
1697 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
1698 specially if not ELF.
1699 (md_apply_fix): Handle BFD_RELOC_64.
1700 (tc_gen_reloc): Handle BFD_RELOC_64.
1702 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
1703 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
1706 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
1707 Don't check S_GET_OTHER.
1711 * config/tc-mips.c (mips_ip): Accept an odd floating point
1712 register with l.s or s.s.
1714 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
1716 (obj_aout_type): New static function.
1720 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
1725 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
1726 and sdaoff expressions.
1728 * write.c (fixup_segment): Don't add symbol value to addend if
1729 TC_V850 and OBJ_ELF.
1730 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
1733 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
1734 (md_pcrel_from_section): Likewise.
1735 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
1739 * config/tc-v850.c (md_pcrel_from): Delete unused function.
1740 (md_pcrel_from_section): New function.
1741 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
1746 * config/tc-mips.c (load_register): Add cast to offsetT when using
1751 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
1752 (elf_frob_file_after_relocs): ... here. New function.
1753 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
1754 * write.c (write_object_file): Call *frob_after_relocs after the
1755 call to write_relocs.
1757 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
1759 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
1760 symbol, as this messes up linking. Instead, expand the recursive call
1761 inline and change up the appropriate bits to get the 0x8000 offset
1762 in the reloc addend.
1766 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
1767 two symbols in the same segment to be adjusted.
1769 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
1770 * configure: Rebuild.
1774 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
1776 (ppc_elf_validate_fix): Ditto.
1777 (md_assemble): Ditto.
1779 (ppc_pe_section): Ditto.
1780 (ppc_frob_symbol): Ditto.
1784 * config/mn10300.c (md_assemble): Pass an extra shift count
1785 to mn10300_insert_operand based on the opcode format.
1786 (mn10300_insert_operand): Accept and use extra shift count
1789 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
1790 formats rather than hard-coded constants.
1792 * config/tc-mn10300.c (md_assemble): Format D5 instructions
1793 are 7 bytes long. Write out instructions in big-endian format.
1797 * config/tc-mn10300.c (md_assemble): Tweak further so
1798 that all instructions are parsed correctly.
1802 * as.h: Include libiberty.h.
1803 (xmalloc, xrealloc): Don't declare.
1804 * as.c: Don't include libiberty.h.
1805 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
1806 * config/tc-mips.c: Likewise.
1807 * messages.c: Likewise.
1808 (xstrerror): Don't declare.
1809 * xmalloc.c: Remove.
1813 * config/tc-mn10300.h (pre_defined_registers) Remove.
1814 (system_registers, cc_names): Likewise.
1815 (address_registers, data_registers, other_registers): New register
1817 (register_name, system_register_name, cc_name): Remove.
1818 (mn10300_reloc_prefix): Likewise.
1819 (data_register_name): New function.
1820 (address_register_name, other_register_name): Likewise.
1821 (md_assemble): Rough cut at parsing operands. Remove lots of
1823 (md_apply_fix3): Disable for now.
1827 * config/tc-m68k.c (select_control_regs): New function, extracted
1828 out of m68k_init_after_args.
1829 (m68k_init_after_args): Use it.
1830 (mri_chip): Use it here as well to update set of allowed control
1835 * config/obj-elf.c (elf_begin): New function.
1836 (obj_elf_section): Add the section symbol to the symbol table.
1837 * config/obj-elf.h (obj_begin): Define.
1838 (elf_begin): Declare.
1839 * as.c (perform_an_assembly_pass): Call obj_begin if it is
1844 * config/obj-coff.c (fixup_segment): Subtract the section address
1845 from a PC relative reloc if TC_M68K.
1849 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
1850 .uaxword available even if not OBJ_ELF.
1851 (md_atof): Remove unused local variable wordP.
1855 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
1856 for Matsushita MN10x00 support.
1857 * configure.in: Recognize mn10x00-*-*
1858 * configure: Rebuilt.
1862 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
1863 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
1864 (load_expression): track clobbering of base reg before jmp/jsr.
1865 (s_alpha_file): pass case_hack flags and source filename via
1866 symbol table to bfd.
1867 * tc-alpha.h (TC_CONS_FIX_NEW): define
1871 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
1872 * configure: Rebuild.
1876 * read.c (s_macro): Warn if a macro has the same name as a
1878 (s_space): In m68k MRI mode, align to a word boundary.
1879 * macro.c (define_macro): Add namep parameter. Change all
1881 * macro.h (define_macro): Update declaration.
1883 * as.c (show_usage): Print bug report address.
1884 (parse_args): Change version printing to match current GNU
1886 * gasp.c (show_usage): Print bug report address.
1887 (main): Change version printing to match current GNU standards.
1889 * config/tc-m68k.c (init_table): Correct access control unit
1892 * config/tc-alpha.c: Add some static function prototypes.
1893 (alpha_macros): Move to top of file. Make static.
1894 (alpha_num_macros): Move to top of file.
1899 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
1905 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
1906 and just call sprintf_vma.
1910 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
1914 * read.c (float_cons): Call md_flush_pending_output if it is
1919 * config/tc-d10v.c (md_operand): Created. Allows operands to
1921 * config/tc-d10v.h (md_operand): Undefined.
1925 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
1926 (md_assemble): A fixup width of '3' means a 1 byte reloc.
1930 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
1931 reloc for the i960 for a reloc in the same section. This undoes
1932 one of the two changes made Aug 19.
1936 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
1937 symbols to the position of the debugging information.
1941 * expr.c (expr): Always use unsigned right shifts for >>.
1945 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
1946 slots when processing BL fixups.
1948 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
1949 on the first half of the instruction.
1953 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
1954 stabs expression, rather than giving an error.
1956 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
1961 * config/tc-mips.c (append_insn): Fill in the value for a constant
1962 jump, rather than creating a reloc.
1966 * config/tc-mips.c (append_insn): Don't swap an instruction which
1967 sets a condition code with an instruction which uses a condition
1969 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
1972 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
1973 (mips_ip): Give a better error message if the ISA level is wrong.
1974 (md_parse_option): Recognize -mcpu=[v][r]5000.
1978 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
1980 (load_register): Ensure hi32 bits are not lost during lo32bit
1981 processing. Fix shift offset that was overflowing into the next
1982 instruction field. Add code to generate shorter sequences for
1983 constants with a single contiguous seqeuence of ones.
1987 * config/tc-d10v.c (d10v_dot_word): New function to support
1988 "@word" with the word pseudo-op.
1989 (md_apply_fix3): Cleanup and changes to support correct sizes
1990 for 16 and 18-bit relocs.
1994 * configure.in (sparc-*-aout): Set `em'.
1995 * configure: Regenerated.
1996 * config/te-sparcaout.h: New file.
1997 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
1998 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
1999 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
2000 (SPECIAL_CASE_{SETSW,SETX}): Define.
2001 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
2002 (md_begin): Delete setting of `target_big_endian'.
2003 (output_insn): New function.
2004 (md_assemble): Rewrite. Add `setx' support.
2005 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
2007 (md_atof): Add little endian support.
2008 (md_number_to_chars): Likewise.
2009 (md_apply_fix): Likewise.
2010 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
2011 (md_parse_option): Likewise.
2012 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
2016 * ecoff.c (ecoff_new_file): New function.
2017 * ecoff.h (ecoff_new_file): Declare.
2018 * config/obj-ecoff.h (obj_app_file): Define.
2022 * config/tc-alpha.c (load_expression): Bias the .lit8 section
2023 symbol by 32k so that our 16-bit signed offset can address the
2028 * config/tc-mips.c (load_register): Remove unused variable tmp.
2032 * config/tc-mips.c (load_register): Remove unnecessary code that
2033 was causing the high 32bits of 64bit constants to be lost.
2037 * config/tc-d10v.c: Added changes to support function
2038 pointers and "@word" syntax.
2043 * config/tc-v850.c: Remove commented out and #if 0'd code.
2044 (v850_reloc_prefix): Provide prototype.
2045 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
2047 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
2048 (md_assemble): Add default to case statement.
2052 * config/tc-v850.c (md_assemble): Compute size of the instrction
2055 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
2060 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
2061 routines to fetch/store the updated instruction from/to memory.
2062 (v850_insert_operand): If the operand has a specialized insert
2067 * config/tc-v850.c (reg_name_search): Align calling convention to
2068 be like identical function found in tc-ppc.c.
2069 (get_reloc): Removed.
2070 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
2071 (md_assemble): emit fixups.
2072 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
2074 (md_apply_fix3): handle fixups/relocs.
2075 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
2081 * configure.in (sh-*-elf*): New target.
2082 * config/tc-sh.h (TARGET_ARCH): Define.
2083 (WORKING_DOT_WORD): Define.
2084 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
2085 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
2086 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
2087 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
2088 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
2089 (SUB_SEGMENT_ALIGN): Likewise.
2090 (RELOC_32): Don't define.
2091 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
2092 (target_big_endian): Declare if OBJ_ELF.
2093 (TARGET_FORMAT): Define if OBJ_ELF.
2094 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
2096 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
2097 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
2098 (struct sh_count_relocs): Define.
2099 (sh_count_relocs): New static function, broken out of
2100 sh_frob_file. Add BFD_ASSEMBLER code.
2101 (sh_frob_section): Likewise.
2102 (sh_frob_file): Call sh_frob_section.
2103 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
2104 call section_symbol rather than seg_info (seg)->dot.
2105 (md_section_align): Add OBJ_ELF version.
2106 (SWITCH_TABLE_CONS): Define.
2107 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
2108 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
2109 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
2111 (struct reloc_map): Define if not BFD_ASSEMBLER.
2112 (coff_reloc_map): Likewise.
2113 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
2114 (tc_gen_reloc): New function if BFD_ASSEMBLER.
2115 * write.c (write_relocs): Ifdef out fx_where test which triggers
2116 inappropriately for SH ELF.
2117 (write_object_file): Call tc_frob_file_before_adjust and
2118 obj_frob_file_before_adjust if they are defined.
2120 * write.c (write_object_file): Use BFD_RELOC_16, not
2121 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
2123 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
2128 * config/tc-d10v.c (find_opcode): Fix problem with calculating
2129 branch sizes in across sections.
2134 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
2139 * config/tc-v850.c (md_begin): Deal with end of opcode
2145 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
2146 the wrong opcode for cases like st2w where there are many forms
2147 of the same instruction.
2151 * expr.c (operand): If md_parse_name is defined, call it before
2152 calling symbol_find_or_make.
2153 * config/tc-ppc.h (md_parse_name): Define.
2154 (ppc_parse_name): Declare.
2155 * config/tc-ppc.c (reg_name_search): Add regs and regcount
2157 (register_name): Update call to reg_name_search.
2158 (cr_operand): New static variable.
2159 (cr_names): New static const array.
2160 (ppc_parse_name): New function.
2161 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
2162 cr_operand before calling expression.
2166 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
2167 hppa_gen_reloc_type call.
2171 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
2172 on addresses. Improved error messages.
2173 * doc/c-d10v.texi: Added docs for register pairs.
2177 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
2182 * ecoff.c (init_file): Initialize fMerge to 1.
2183 (add_file): Restore old file merging code, but only merge files if
2185 (ecoff_directive_loc): Clear fMerge field of current file.
2186 (ecoff_generate_asm_lineno): Likewise.
2190 * doc/c-d10v.texi: Fix typo.
2195 * config/tc-v850.c (md_assemble): Correct bit masking for
2196 hi and lo expressions.
2198 * config/tc-v850.c (md_assemble): Rough cut at demanding
2199 "ep" or "r30" in sst and sld instructions.
2200 (md_apply_fix3): Don't abort. Just warn that we don't
2203 * config/tc-v850.c (CC_NAME_CNT): Define.
2204 (cc_name): New function.
2205 (md_assemble): Handle V850_OPERAND_CC correctly.
2207 * config/tc-v850.c (md_assemble): Don't forget to initialize
2210 * config/tc-v850.c (reg_name_search): Generalize to search
2211 any given register table.
2212 (register_name): Pass appropriate table and size to reg_name_search.
2213 (system_register_name): New function.
2214 (SYSREG_NAME_CNT): Define.
2215 (md_assemble): Handle operands which are system registers.
2217 * config/tc-v850.c (md_assemble): If we find a register, but the
2218 opcode doesn't want a register, then we don't have a match.
2219 (md_assemble): Get size of the instruction from the opcode table.
2224 * configure.in: Set and substitute HLDENV.
2225 * configure: Rebuild.
2226 * Makefile.in (HLDENV): New variable.
2227 (as.new): Use $(HLDENV).
2229 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
2230 if an array dimension is not known.
2234 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
2235 * doc/c-d10v.texi: Cleanup.
2240 * config/tc-v850.c: New file.
2241 * config/tc-v850.h: New file.
2242 * configure (v850-*-elf): New target.
2243 * configure.in (v850-*-elf): New target.
2248 * doc/c-d10v.texi: New file.
2249 * doc/all.texi: Added D10V stuff.
2250 * doc/as.texinfo: Added D10V stuff.
2254 * config/tc-d10v.c: All references to defined symbols should
2255 now use the optimal instruction. .float and .double now work.
2259 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
2260 section address for the i960 as is done for the i386.
2264 * mpw-config.in: Add wildcards for config matching, add mips-*-*
2265 case, forward-include bfd/elf-bfd.h.
2269 * config/tc-d10v.c: Add additional information to the opcode
2270 table to help determinine which instructions can be done
2275 * config/tc-arm.c: Major changes to add Thumb support, with lots
2277 Reverted to INSN_SIZE macro, rather than insn_size variable.
2278 (insns): Added ARM "bx" instruction support.
2279 (tinsns): Added Thumb instruction definition structure.
2280 (arm_tops_hsh): Added hash structure for Thumb opcodes.
2281 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
2282 (opcode_select,s_arm,s_thumb,s_code): Added.
2283 (decode_shift): Allow upper-case RRX.
2284 (do_ldst): Simpler halfword support.
2285 (do_ldmstm): Improved.
2286 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
2287 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
2288 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
2289 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
2290 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
2292 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
2293 (md_parse_option): Add support for -mthumb.
2294 (md_show_usage): Updated to reflect new command line option.
2295 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
2296 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
2297 fragment information to be held.
2301 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
2302 non-null, but is a constant.
2303 (fix_new_arm): Call make_expr_symbol to make the expression symbol
2304 so that error reporting will work correctly.
2308 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
2309 against weak symbols.
2313 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
2314 (ppc_force_relocation): Declare if OBJ_XCOFF.
2315 * config/tc-ppc.c (ppc_force_relocation): New function if
2320 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
2324 * config/tc-d10v.c: Fix problem with relocs.
2328 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
2329 with nops if not in data_section or bss_section.
2333 Add support for openVMS/Alpha.
2334 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
2335 * config/obj-evax.c: New file.
2336 * config/obj-evax.h: New file.
2337 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
2339 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
2340 defined. Add case for bfd_target_evax_flavour.
2341 * config/vms-a-conf.h: New file.
2342 * conf-a-gas.com: New file.
2343 * configure.in: Add target alpha-*-*vms*.
2344 * configure: Rebuild.
2345 * makefile.vms: New file.
2346 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
2347 Don't call ffs on openVMS/Alpha.
2351 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
2352 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
2354 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
2355 (NEED_DECLARATION_SBRK): New macro.
2356 * configure, conf.in: Rebuild.
2357 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
2358 (strstr): Declare if NEED_DECLARATION_STRSTR.
2359 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
2363 * symbols.c (resolve_symbol_value): Handle addition or subtraction
2364 by a constant before entering the main switch. Reject attempts to
2365 apply an arithmetic function to non-absolute symbols, except for
2366 the special case of subtraction of two symbols in the same
2369 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
2370 not to more than a 16 byte boundary.
2372 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
2374 (tc_gen_reloc): If out of memory call as_fatal rather than
2375 assert. If no howto found, call as_bad_where rather than
2376 as_fatal. Change the error message slightly. Set howto to a
2377 non-NULL value in order to keep going.
2381 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
2385 * config/tc-h8300.c (get_specific): New operand "size" derived
2386 from ".b", ".w" and ".l" extensions. All callers changed. If
2387 the base instruction has no operands, then use the size to
2388 determine which specific instruction to use.
2392 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
2396 * config/tc-mips.c (md_section_align): Don't change addr if
2401 * config/tc-hppa.c: Revert yesterday's changes.
2405 * config/tc-d10v.c: Disable range checking on 16-bit values.
2409 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
2410 not just the ones that don't match.
2414 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
2418 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
2419 pre-cursor to adding Thumb support. Also added cpu_variant flag
2420 information to each of the asm_flg structures.
2421 (md_parse_option): Updated ARM7 parsing to allow 't' for
2422 thumb/halfword support, aswell as 'm' for long multiply.
2423 (md_show_usage): Updated help message.
2424 (md_assemble): Check that instruction flags are applicated to the
2425 current cpu variant.
2426 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
2427 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
2428 signextension instructions.
2429 (do_ldst): Generate halfword and signextension variants if
2430 mnemonic flags match.
2431 (ldst_extend): Do not allow shifts in the offset field of halfword
2432 or signextension instructions.
2433 (validate_offset_imm): Provide check on halfword and signextension
2435 (add_to_lit_pool): Merge identical literal pool values.
2439 * config/tc-hppa.c (selector_table): Add 'E' selector.
2440 (cons_fix_new_hppa): Don't coke on e_esel.
2441 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
2442 to help generate exception handling tables.
2443 (md_apply_fix): Don't try to apply fixups with an e_esel
2445 (hppa_fix_adjustable): Fixups with e_esel selectors
2450 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
2455 * config/tc-d10v.c: Added lots of error checking. Added hacks
2456 to support accumulator shifts.
2460 * symbols.c (S_SET_EXTERNAL): Let .weak override.
2461 (S_CLEAR_EXTERNAL): Likewise.
2462 (S_SET_WEAK): Remove error; just let .weak override.
2466 * config/tc-d10v.c (md_assemble): Now handles multiline
2471 * config/tc-d10v.c: Fix packaging bug. Added range checking.
2472 Added kludge for divs instruction. Fixed minor problem with
2473 multiple text sections.
2474 * config/tc-d10v.h (d10v_cleanup): Change prototype.
2478 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
2479 addresses to be right-shifted by 2.
2483 * config/tc-d10v.c: Many changes to get relocs working.
2484 (register_name): No longer creates a symbol for register names.
2485 (pre_defined_registers): moved to opcodes/d10v-opc.c.
2486 (d10v_insert_operand): Now works correctly for either container.
2487 * config/tc-d10v.h (d10v_cleanup): Declare.
2491 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
2492 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
2496 * tc-alpha.c: Patches to track current minimum alignment to reduce
2497 the number of fragments created with frag_align.
2498 (alpha_current_align): New static variable.
2499 (s_alpha_text): Reset alignment to 0.
2500 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
2501 (s_alpha_stringer, s_alpha_space): New functions.
2502 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
2503 (alpha_cons_align): New function to replace both of them.
2504 (emit_insn): Only align if alpha_current_align is less than 2;
2505 reset alpha_current_align to 2.
2506 (s_alpha_gprel32): Likewise.
2507 (s_alpha_section): New function. Basically duplicate the other
2508 alpha section change hooks. Only define for ELF.
2509 (s_alpha_float_cons): Simplify alignment handling.
2510 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
2511 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
2512 Don't define the s_alpha_cons pseudo-ops. Do define
2513 s_alpha_stringer and s_alpha_space pseudo-ops.
2514 (alpha_align): Skip if less than current default alignment. Set
2516 * tc-alpha.h (md_flush_pending_output): Remove.
2517 (md_cons_align): Add.
2519 * tc-alpha.c: Add oodles of function description comments.
2520 (md_bignum_to_chars): Remove; there are no callers.
2521 (md_show_usage): Mention some more variants.
2526 * configure.in (sparc-*-sysv4*): New target.
2527 * configure: Rebuild.
2529 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
2530 uaxword to use s_uacons.
2531 (sparc_no_align_cons): New static variable.
2532 (s_uacons): New static function.
2533 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
2536 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
2538 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
2539 5 with regard to shared libraries.
2540 * configure: Rebuild.
2542 * config/tc-m68k.c (m68k_ip): Use the correct length when
2543 allocating space for the unsupported architecture error message.
2547 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
2551 * config/tc-d10v.c: New file.
2552 * config/tc-d10v.h: New file.
2553 * configure (d10v-*-elf): New target.
2554 * configure.in (d10v-*-elf): New target.
2558 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
2562 * config/tc-alpha.c (alpha_align): Change fill parameter
2563 to a pointer. Take NULL as 0 or nop depending on section. Change
2565 (s_alpha_align): Rename local variables.
2567 * doc/as.texinfo (.align): Document action of omitted
2572 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
2573 when an unsupported PC relative reloc is seen, rather than calling
2576 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
2577 Instead, check state when '#' comment is seen.
2581 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
2583 (tc_gen_reloc): If fixup was changed to be PC relative, change
2584 reloc type accordingly. Use name of reloc in error message.
2586 * as.h: Don't define const or volatile.
2587 * flonum.h: Don't define const.
2589 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
2590 if fx_pcrel is set. Correct setting the addend case in the
2591 OBJ_ELF case (from Andreas Schwab
2593 (md_show_usage): Correct -mfc5200 to -m5200.
2597 * doc/c-m68k.texi: Document -m5200 flag.
2598 * doc/as.texinfo: Likewise.
2600 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
2605 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
2606 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
2610 * Makefile.in (VERSION): Set to cygnus-2.7.1.
2612 * Released binutils 2.7.
2616 * config/tc-mips.c (mips_ip): Only perform range check when
2617 dealing with O_constant expressions.
2621 * m68k-parse.h (m68k_register): Add new coldfile control
2624 * config/tc-m68k.c (mcf5200_control_regs): New variable,
2625 array of control registers for the coldfire.
2626 (cpu_of_arch): Added mcf5200.
2627 (archs): Added mcf5200.
2628 (init_table): Add new control registers.
2629 (m68k_ip): Added support for new control registers.
2630 (m68k_init_after_args): Likewise.
2632 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
2636 * read.h (is_it_end_of_statement): Declare.
2637 * read.c (is_it_end_of_statement): Remove declaration.
2639 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
2641 (md_assemble): Fix handling of @l with an unsigned constant. Add
2642 default case to reloc switch.
2644 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
2647 * ecoff.c (last_lineno): New static variable.
2648 (add_procedure): Set last_lineno.
2649 (ecoff_directive_loc): Likewise.
2650 (ecoff_generate_asm_lineno): Likewise.
2651 (ecoff_fix_loc): New function.
2652 * ecoff.h (ecoff_fix_loc): Declare.
2653 * config/tc-mips.c (append_insn): When inserting nops, and using
2654 ECOFF debugging, call ecoff_fix_loc.
2658 * config/tc-h8300.c (build_bytes): If an operand type is
2659 marked as SRC_IN_DST retrieve it from the "destination" op.
2663 * configure.in (arm-*-riscix*): Set emulation to riscix.
2664 * configure: Rebuild.
2665 * config/te-riscix.h: New file to define TE_RISCIX.
2667 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
2671 * Makefile.in (config.status): Just run config.status as other
2676 * mpw-config.in (TARGET_OS): Add definition to conf.
2680 * config/tc-mips.c (append_insn): Parenthesize
2681 cop_interlocks expressions.
2685 * listing.c (listing_print): Close the listing file if it is not
2686 stdout. Close the other files opened for the listing.
2688 * config/tc-sparc.h (md_cons_align): Define.
2689 (sparc_cons_align): Declare.
2690 (HANDLE_ALIGN): Define.
2691 (sparc_handle_align): Declare.
2692 * config/tc-sparc.c (sparc_cons_align): New function.
2693 (sparc_handle_align): New function.
2694 * read.c (cons_worker): Call md_cons_align if it is defined.
2696 * as.h (struct frag): Add fr_file and fr_line fields.
2697 * frags.c (frag_new): Set fr_file and fr_line.
2698 (frag_var): Likewise.
2699 (frag_variant): Likewise.
2701 * as.h (struct frag): Remove unused align_mask and align_offset
2704 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
2709 * configure.in (mips-*-osf*): New target.
2710 * configure: Rebuild.
2712 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
2716 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
2717 between co-processor comparisons and branches for the VR4300.
2721 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
2722 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
2724 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
2725 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
2726 includedir): Use autoconf set values.
2731 * listing.c (listing_eject): Don't do anything if listing is 0.
2732 (listing_list): Likewise.
2733 (listing_source_line): Likewise.
2734 (listing_title): Don't save title if listing is 0.
2735 (listing_source_file): Check listing rather than listing_tail.
2737 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
2738 using shared libraries.
2739 * configure: Rebuild.
2743 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
2744 absolute expression if a relocation type was specified.
2748 * configure.in: Add support for *-*-rtems* configurations.
2749 * configure: Rebuild.
2753 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
2754 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
2755 opcodes library for alpha.
2756 * configure: Rebuild with autoconf 2.10.
2757 * config/tc-alpha.c: Substantial rewrite to add ELF support and
2758 use new opcode table.
2759 * config/tc-alpha.h (md_undefined_symbol): Don't define.
2760 (LOCAL_LABEL): Define differently if OBJ_ELF.
2761 (FAKE_LABEL_NAME): Define if OBJ_ELF.
2762 * config/alpha-opcode.h: Remove.
2763 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
2764 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
2765 include/opcode/alpha.h rather than config/alpha-opcode.h.
2769 * config/obj-aout.c (obj_emit_relocations): Give an error if the
2770 relocation symbol was not resolved.
2771 * config/obj-coff.c (do_relocs_for): Likewise.
2773 * write.c (adjust_reloc_syms): Refetch the symbol section after
2774 calling S_GET_VALUE, since it may have changed.
2776 * expr.c (struct expr_symbol_line): Define.
2777 (expr_symbol_lines): New static variable.
2778 (make_expr_symbol): Add entry to expr_symbol_lines.
2779 (expr_symbol_where): New function.
2780 * expr.h: Use extern on function declarations.
2781 (expr_symbol_where): Declare.
2782 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
2783 rather than printing the meaningless name of an expression
2788 * config/tc-i386.c (md_number_to_chars): Deleted.
2789 * config/tc-i386.h (md_number_to_chars): New macro.
2791 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
2793 (load_symbol_address, load_expression): Use build_mem.
2794 (build_operate): New function.
2795 (emit_addq_r): Use it.
2799 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
2803 * config/tc-vax.h (NOP_OPCODE): Define.
2807 * config/obj-vms.h (S_IS_COMMON): Define.
2808 (S_IS_LOCAL): Check for \002 as well as \001.
2809 (LONGWORD_ALIGNMENT): New macro.
2810 (SUB_SEGMENT_ALIGN): Use it.
2814 * config/vms-conf.h: Reconcile with conf.in.
2818 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
2821 * doc/as.texinfo: Document that any number of hex digits can
2824 * as.c (struct defsym_list): Define.
2825 (defsyms): New static variable.
2826 (parse_args): Just put --defsym arguments on defsyms list, rather
2828 (main): Define defsyms after output file is created.
2830 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
2833 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
2834 insn_error rather than calling check_absolute_expr.
2836 * as.c (emulation_name): Remove unused static variable.
2837 (default_emul_bfd_name): Add return NULL to avoid warning.
2838 * ecoff.c (ecoff_stab): Remove unused variables name and
2840 * frags.c (frag_new): Remove unused variable tmp.
2841 * hash.c (hash_grow): Parenthesize + within <<.
2842 (hash_print_statistics): Use %lu, not %d, to print unsigned
2844 * messages.c: Include "libiberty.h".
2845 (fprint_value): Add cast to avoid printf warning.
2846 (sprint_value): Likewise.
2847 * read.c: Include "ecoff.h".
2848 (emit_expr): Add casts to avoid printf warnings.
2849 * read.h: Use extern for function declarations.
2850 (pop_insert): Declare.
2851 * stabs.c: Include "ecoff.h".
2852 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
2853 former_last_fragP, and new_fragP.
2854 * subsegs.h (subsegs_print_statistics): Declare.
2855 * symbols.c (debug_verify_symchain): Change macro to discard
2857 * write.c (dump_section_relocs): Likewise.
2858 * write.h: Use extern for function declarations.
2859 (write_print_statistics): Declare.
2860 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
2862 * config/e-mipself.c (mipself_bfd_name): Likewise.
2863 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
2865 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
2866 emit relocations against external symbols.
2868 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
2869 message if bfd_reloc_type_lookup fails, rather than calling
2872 * config/tc-alpha.c (alpha_force_relocation): Add
2873 BFD_RELOC_12_PCREL to switch.
2877 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
2878 TE_PE (Lfoo, not .Lfoo).
2882 * read.c (s_fill): Don't warn about a zero repeat count.
2884 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
2885 coprocessor register.
2887 * config/tc-i386.c (md_assemble): When checking the size of a
2888 register to set the size of an instruction, do a bitwise and with
2889 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
2894 * config/tc-h8300.c (parse_reg): Tweak error messages.
2895 (build_bytes): Likewise.
2896 (skip_colonthing): Handle :32 suffix.
2897 (get_specific): Promote L_24 to L_32 if it makes a match.
2898 Don't always promote L_8 to L_16.
2899 (do_a_fix_imm): Clean up L_32 and L_24 handling.
2901 * config/tc-h8300.c (Smode): New variable.
2902 (h8300hmode): Turn off Hmode.
2903 (h8300smode): New function. Turn on Smode and Hmode.
2904 (md_pseudo_table): New ".h8300s" pseudo-op.
2905 (parse_reg): Handle "exr" register.
2906 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
2907 Handle "mach" and "machl" operands for ldmac.
2908 (get_specific): Handle "stm.l" and "ldm.l".
2909 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
2910 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
2915 * doc/as.texinfo: Reorder chapter of machine dependent options so
2916 that it is sorted by chip name.
2918 * doc/as.texinfo: Use consistant spelling of Vax.
2919 * doc/c-vax.texi: Likewise.
2923 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
2925 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
2926 (md_apply_fix): Likewise.
2927 (pa_try): New function.
2928 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
2932 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
2933 (pa_level): New function.
2937 * listing.c (listing_newline): Don't do anything if listing is 0.
2941 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
2942 symbol_create, not symbol_new.
2946 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
2947 absolute symbol unless TC_FORCE_RELOCATION returns true.
2949 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
2951 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
2952 the symbol to a pointer. Don't set previous_file_symbol.
2953 Simplify symbol list rearrangement.
2954 (coff_frob_symbol): Don't do anything with C_FILE symbols.
2955 (coff_adjust_symtab): Don't check previous_file_symbol.
2959 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
2961 (md_pseudo_table): Add ppc_elf_lcomm.
2965 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
2966 d0{24:1},d0' works without an immediate prefix on the bit numbers.
2967 (md_begin): Add digits to alt_notend_table.
2968 (md_parse_option): Make s a const pointer.
2970 * config/tc-sparc.c (md_pseudo_table): Add "empty".
2971 (s_empty): New static function.
2973 * config/obj-coff.c (struct filename_list): Only define if not
2975 (filename_list_head, filename_list_tail): Likewise.
2976 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
2977 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
2978 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
2979 have a section of N_ABS rather than N_DEBUG. If we do a merge,
2980 remove the new symbol from the list.
2981 (obj_coff_endef, both versions): Call tag_insert even if there is
2982 an old symbol with the same name, if the old symbol does not
2984 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
2985 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
2986 symbol that will be moved to the end of the symbol list.
2987 (coff_adjust_section_syms): Always call section_symbol for .text,
2989 (coff_frob_section): Likewise. Also, remove unused variable
2992 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
2993 manipulating frags directly.
2994 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
2997 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
2998 whether the next character is '*' before checking whether we are
2999 at the start of a line. Permit LINE_COMMENT_START to start a
3000 comment in state 1 (seen some whitespace) as well, to match the
3003 * gasp.c (do_align): Permit a fill value for .align.
3007 * read.c (next_char_of_string): Warn if a newline is seen in the
3008 middle of a string. Call bump_line_counters when appropriate.
3012 * symbols.c (colon): Use LOCAL_LABEL.
3016 * Makefile.in (install): Don't check to see if tooldir exists.
3017 Make $(tooldir) and $(tooldir)/bin.
3021 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
3026 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
3027 insn_error. Don't put the string "ERROR" in insn_error. Set
3028 insn_error rather than calling as_warn for an unsupported opcode.
3032 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
3033 before permitting -64.
3034 * output-file.c (output_file_create): Remove duplicate
3039 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
3040 section rather than a .reginfo section.
3041 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
3044 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
3045 bit value with the high bit set.
3049 * read.c (s_lcomm): Set section flags for .sbss section.
3051 * config/tc-mips.c (mips_64): New static variable.
3052 (mips_target_format): If mips_64, return elf64 targets rather than
3054 (md_longopts): Add "32" and "64".
3055 (md_parse_option): Handle -32 and -64.
3056 (md_show_usage): Mention -32 and -64.
3057 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
3062 * config/tc-ppc.c (comment_chars): Make '!' a comment character
3063 for Solaris compatibility.
3065 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
3066 .stabd with 4 arguments into a .stabn.
3070 * config/tc-mips.c (macro): When passing X_add_number to
3071 macro_build, cast it to int first.
3075 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
3076 created by emit_expr.
3078 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
3079 as_perror rather than assert.
3083 * config/tc-mips.c (mips_ip): Mark sections created to hold
3084 floating point information as read only.
3088 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
3092 * read.c (potable): Add .skip as a synonym for .space.
3094 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
3095 4 arguments, providing the 4th argument is 0, to allow
3096 compatibility with the Solaris assembler.
3100 * config/tc-sh.h (struct sh_segment_info_type): Define.
3101 (TC_SEGMENT_INFO_TYPE): Define.
3102 (sh_frob_label): Declare.
3103 (tc_frob_label): Define.
3104 (sh_flush_pending_output): Declare.
3105 (md_flush_pending_output): Define.
3106 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
3107 reloc before the instruction if necessary.
3108 (sh_frob_label): New function.
3109 (sh_flush_pending_output): New function.
3110 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
3111 when looking for the reloc for the target of .uses.
3112 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
3113 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
3115 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
3116 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
3117 the absolute symbol.
3119 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
3120 TC_SEGMENT_INFO_TYPE is defined.
3124 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
3125 matches the register size.
3129 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
3131 (do_relocs_for): Likewise.
3132 (fixup_segment): Don't just quit if linkrelax is set. Try to
3133 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
3138 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
3139 relative relocations with -mrelocatable. Also allow unfixed
3140 relocs in .ex_shared.
3144 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
3145 non-zero before assuming this is a long file name.
3146 (w_strings): Likewise.
3147 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
3149 * config/obj-coff.c (w_strings): Move declaration of i inside
3150 #ifdef block which uses it.
3154 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
3155 "relaxmode". Output relocs which identify various relaxing
3156 possibilities for mov.[bwl] instructions.
3157 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
3161 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
3162 (MD_APPLY_FIX3): Define.
3163 * config/tc-arm.c (my_get_expression): Only watch for bad segments
3165 (md_apply_fix3): Renamed from md_apply_fix.
3166 If pcrel reloc and symbol is in different section, undo effects
3171 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
3172 any reloc with an LR% or RR% field selector for SOM.
3176 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
3181 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
3183 (yank_symbols): Likewise.
3187 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
3188 BFD_ASSEMBLER, add name field.
3189 * config/obj-coff.c: Include "libiberty.h".
3190 (coff_header_append): Handle long section names.
3191 (crawl_symbols): Just use the name field for the symbol name,
3192 without worrying about null byte termination.
3193 (w_strings): Handle long section names.
3194 (write_object_file): Likewise. Also, use the name field, rather
3196 (obj_coff_add_segment): Permit long section names.
3197 (obj_coff_init_stab_section): Use the name field, rather than
3199 (adjust_stab_section): Likewise.
3200 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
3202 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
3203 (mem_fmt): Likewise.
3204 (md_apply_fix): Don't check fx_im_disp field.
3208 * configure.in: Add * after sparc*-*-vxworks.
3209 * configure: Rebuild.
3211 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
3212 define it to comment_chars. Use tc_comment_chars rather than
3214 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
3215 * config/tc-m68k.h (tc_comment_chars): Define.
3216 (m68k_comment_chars): Declare.
3217 * config/tc-m68k.c (m68k_comment_chars): Rename from
3218 comment_chars. Change into a pointer rather than an array.
3219 (md_longopts): Add "bitwise-or".
3220 (md_parse_option): Handle OPTION_BITWISE_OR.
3221 (md_show_usage): Mention --bitwise-or.
3222 * doc/c-m68k.texi: Document --bitwise-or.
3226 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
3229 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
3230 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
3231 correct handling of input line pointer, and call
3232 demand_empty_rest_of_line.
3236 * config/tc-sparc.c (in_bitfield_range): New static function.
3237 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
3238 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
3239 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
3240 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
3241 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
3245 * config/obj-coff.c: BFD_ASSEMBLER:
3246 (coff_last_bf): New static variable.
3247 (coff_frob_symbol): Set endndx of a .bf symbol.
3249 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
3250 (last_bfP): New static variable.
3251 (yank_symbols): Set endndx of a .bf symbol.
3255 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
3256 and -s options. Add -be for good measure.
3260 * read.c (s_space): Support non-constant fill value. Handle fill
3261 value correctly for a size other than 1.
3265 * config/tc-arm.c (my_get_float_expression): Update call to
3266 gen_to_words, X_PRECISION changed from 6 to 5.
3270 * config/tc-ppc.c (register_name,reg_name_search): Move register
3271 name lookup from PE specific code to all targets. Add support for
3272 -mregnames/-mno-regnames to control whether register names are
3274 (md_assemble): Call register_name for all platforms.
3275 (md_parse_option): Add support for -mregnames/-mno-regnames.
3277 * configure.in (powerpcle*-*-solaris): Add support.
3278 (powerpc*-*-linux): Ditto.
3279 * configure: Regenerate.
3281 * config/ppc-sol.mt: New config file for PowerPC Solaris.
3285 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
3286 be associated with a single %lo reloc.
3288 * config/tc-mips.c (load_address): Cast X_add_number to valueT
3289 before comparing against MAX_GPREL_OFFSET, so that negative
3290 numbers are handled correctly.
3295 * config/tc-sparc.c (last_insn): New static variable.
3296 (md_assemble): Warn about putting floating point branches in a
3297 delay slot. If architecture is less than v9, insert NOP
3298 instructions between floating point instructions and floating
3299 point branches. (The SunOS assembler does both these operations.)
3300 Save the last instruction opcode.
3301 (sparc_ip): Add pinsn parameter. Change caller.
3303 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
3304 for byte jump to next instruction to skip empty frags.
3308 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
3309 for addq with a 16 bit signed value, just emit a lda instruction
3314 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
3319 * configure.in: Permit --enable-shared to specify a list of
3321 * configure: Rebuild.
3325 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
3326 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
3330 * config/atof-ieee.c: Fix handling of denormalized extended
3331 precision numbers and overflow/underflow detection.
3332 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
3333 not include the 16 bit gap in the m68k extended precision format.
3337 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
3339 * configure: Rebuild.
3343 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
3344 be accessed using the $gp register.
3345 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
3349 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
3350 bfd_mach_sparc_sparclet if current_architecture is sparclet.
3354 * read.c (get_line_sb): Bump line counters based on
3355 input_line_pointer[-1], not *input_line_pointer. Don't bother to
3356 call LISTING_NEWLINE.
3357 (s_macro): Don't call demand_empty_rest_of_line.
3358 * app.c (do_scrub_chars): When handling C style comments, unget
3363 * read.h (enum linkonce_type): Define.
3364 (s_linkonce): Declare.
3365 * read.c (potable): Add "linkonce".
3366 (s_linkonce): New function.
3367 * subsegs.h (segment_info_type): Add linkonce field to
3368 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
3369 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
3370 (obj_coff_pe_handle_link_once): Declare if TE_PE.
3371 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
3373 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
3374 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
3375 entry based on the linkonce field in segment_info.
3376 * doc/as.texinfo: Document .linkonce.
3380 * doc/as.1: Changed to be recognized by catman -w on Solaris.
3384 * stabs.c (s_stab_generic): Call the listing functions before
3385 doing the rest of the processing, which may involve freeing the
3386 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
3387 SEPARATE_STAB_SECTIONS case.
3389 * config/tc-hppa.c: Remove nested comment.
3390 (tc_gen_reloc): Move label done inside the ifdef in which it is
3392 (md_apply_fix): Pass pointers to correct types to libhppa.h
3393 functions. Always return a value.
3395 * config/tc-mips.h (tc_frob_file): Define.
3396 (mips_frob_file): Declare.
3397 * config/tc-mips.c (struct mips_hi_fixup): Define.
3398 (mips_hi_fixup_list): New static variable.
3399 (imm_unmatched_hi): New static variable.
3400 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
3401 offset_reloc. Pass imm_unmatched_hi to append_insn.
3402 (append_insn): Add unmatched_hi parameter. If it is set, add the
3403 new fixup to mips_hi_fixup_list. Change all callers.
3404 (mips_ip): Set imm_unmatched_hi when appropriate.
3405 (mips_frob_file): New function.
3409 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
3410 * configure: Regenerated.
3414 * as.c (main): Call bfd_set_error_program_name.
3418 * as.h (strdup): Don't declare.
3419 * stabs.c: Include libiberty.h
3420 (get_stab_string_offset): Use xstrdup rather than strdup.
3421 (s_stab_generic): Likewise.
3422 * as.c (parse_args): Likewise.
3423 * read.c (s_mri_sect): Likewise.
3425 * gasp.c (change_base): Recognize \(...) construct documented to
3426 pass through enclosed characters literally through to the output.
3427 (process_assigns): Likewise. Also, be more careful to avoid
3428 looking past the end of the buffer.
3432 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
3437 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
3441 * config/alpha-opcode.h: Added cvtst instruction.
3445 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
3446 string. Extract the addend from the result of expression.
3450 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
3451 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
3452 mode, change the state to 3 rather than 1.
3456 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
3457 correspond to 11 March change.
3461 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
3462 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
3466 * configure: Rebuild with autoconf 2.8.
3470 * config/atof-ieee.c (gen_to_words): Improve handling of
3471 X_PRECISION numbers. Based on patches from Andreas Schwab
3476 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
3478 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
3479 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
3480 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
3481 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
3486 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
3490 * config/tc-hppa.c (SUBSEG_MILLI): Define.
3491 (pa_def_subspaces): Add $MILLICODE$.
3492 (pa_spaces_begin): Set section flags for $MILLICODE$.
3496 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
3497 not SHT_NOBITS. Don't tamper with flags based on type if a
3498 special section was found (revert Feb 29 change).
3500 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
3501 if aligning to a longword boundary or greater.
3505 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
3509 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
3510 * configure: Regenerated.
3514 * config/tc-hppa.c: Add default definitions for R_N0SEL and
3515 R_N1SEL since they're not defined for old versions of hpux.
3517 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
3518 Set "sym_ptr_ptr" and "addend" fields to dummy values for
3519 R_N0SEL and R_N1SEL.
3523 * flonum-konst.c: Add two more constants for 1e+-2048 and
3524 1e+-4096, and correct the other constants.
3526 * symbols.c (resolve_symbol_value): Handle O_logical_not.
3530 * config/obj-elf.c (obj_elf_section): Allow predefined section
3531 types to set the nobits type. Avoid a shadowed declaration.
3535 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
3536 reduce relocs using e_nlrsel field selectors.
3538 * write.c (fix_new_exp): Don't use #elif. Some compilers
3541 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
3543 (pa_chk_field_selector): Handle new field selectors for SOM.
3547 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
3548 * configure: Rebuild.
3549 * config/te-linux.h (LOCAL_LABELS_FB): Define.
3550 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
3552 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
3557 * mpw-make.sed: Update to handle shared library support.
3561 * config/tc-sparc.c (sparc_ip): Print all architectures that support
3562 the insn on mismatch.
3566 * configure.in: Add support for a29-coff.
3567 * configure: Rebuild.
3571 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
3572 of the section when looking for the R_SH_USES fixup, because the
3573 frag addresses have not yet been adjusted.
3575 * gdbinit.in: Set a breakpoint on as_warn_where.
3577 * config/tc-mips.c (macro): Add missing arguments to macro_build
3582 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
3583 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
3587 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
3591 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
3592 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
3593 Recognize [uU] format args as sparclet cpregs.
3597 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
3602 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
3603 well as S_IS_EXTERNAL.
3604 (tc_fix_adjustable): Likewise.
3605 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
3606 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
3608 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
3609 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
3613 * config/tc-sparc.c (max_architecture): Change to sparclite for
3615 (default_compatible): Delete.
3616 (sparc_ffs): New function.
3617 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
3618 (sparc_ip): Rewrite architecture match and bump logic.
3622 * configure.in: Check for 'do not mix' from native linker before
3623 trying to use -rpath.
3624 * configure: Rebuild.
3628 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3629 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3630 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3631 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
3633 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
3634 to the symbol lnnoptr field, to get the correct file offset.
3638 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
3639 symbols that are not function, file, or section symbols to be
3644 * configure: Set and substitute RPATH_ENVVAR.
3645 * configure: Rebuild.
3646 * Makefile.in (RPATH_ENVVAR): New variable.
3647 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
3650 doesn't need opcodes. If configuring shared, opcodes needs bfd.
3651 * configure: Rebuild.
3655 * read.c (s_mri_sect): Don't return '\0' in type. Set all
3656 appropriate flags in BFD section.
3658 * configure.in (m68k-*-psos*): New target.
3659 * configure: Rebuild.
3660 * config/te-psos.h: New file.
3661 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
3667 * configure.in: Remove duplicate setting of cpu_type. Check
3668 whether opcodes library is required for on all targets, not just
3670 * configure: Rebuild.
3672 * config/tc-mips.c (mips_big_got): New static variable.
3673 (s_extern): Don't declare.
3674 (reg_needs_delay): New static function.
3675 (macro_build): Permit GOT/CALL_HI/LO relocs.
3676 (macro_build_lui): If place is not NULL, use the number in the
3678 (load_address): Handle mips_big_got case.
3679 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
3681 (OPTION_XGOT): Define.
3682 (md_longopts): Add "xgot" if OBJ_ELF.
3683 (md_parse_option): Handle -xgot.
3684 (md_show_usage): Mention -xgot.
3685 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
3686 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
3690 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
3691 the byte relocation, point it to the low byte of the word.
3695 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
3696 * configure: Rebuild.
3700 * configure.in: Recognize any sparc* cpu.
3701 * configure: Regenerated.
3705 * read.c (potable): Add "mri" and ".mri".
3706 (s_mri): New function.
3707 * read.h (s_mri): Declare.
3708 * app.c (scrub_m68k_mri): New static variable.
3709 (mri_pseudo): New static variable.
3710 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
3711 flag_m68k_mri. Initialize scrub_m68k_mri.
3712 (mri_state, mri_last_ch): New static variables.
3713 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
3715 (app_push): Save new fields.
3716 (app_pop): Restore new fields.
3717 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
3718 If TC_M68K, use a trivial state machine to look for occurrences of
3719 the .mri pseudo-op, and change the mode appropriately.
3720 * as.h (do_scrub_begin): Update prototype.
3721 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
3723 * config/tc-m68k.c (reg_prefix_optional_seen): New static
3725 (m68k_mri_mode_change): New function.
3726 (md_parse_option): Set reg_prefix_optional_seen.
3727 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
3728 (MRI_MODE_CHANGE): Define.
3729 * doc/as.texinfo: Document .mri pseudo-op.
3731 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
3736 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
3737 * config/m68k-parse.h (enum pic_relocation): Define.
3738 (struct m68k_exp): Add pic_reloc field.
3739 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
3740 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
3741 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
3743 * config/tc-m68k.c: Delete definition of NO_RELOC.
3744 (struct m68k_it): Add pic_reloc field.
3745 (add_fix): Copy over pic_reloc field.
3746 (md_pseudo_table): Interpret .align parameter as byte count.
3747 (mote_pseudo_table): Likewise.
3748 (tc_m68k_fix_adjustable): New function.
3749 (get_reloc_code): New function.
3750 (md_assemble): Use it as last argument to fix_new_exp.
3751 (md_apply_fix_2): For a relocation against a symbol don't put the
3752 addend into the data.
3753 (tc_gen_reloc): Different addend computation for OBJ_ELF.
3754 (m68k_ip): Don't relax an operand that requires pic relocation.
3755 (md_begin): Align .text, .data and .bss on 4 byte boundary by
3757 * write.c (fixup_segment): Don't add symbol value to addend if
3758 TC_M68K and OBJ_ELF.
3759 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
3760 (motorola_operand): Add rule for `(zapc, EXPR)'.
3764 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
3765 *really* is permissible before the comma.
3769 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
3770 inserting nop instructions.
3774 * config/te-aux.h: Change include of aux.h to aux-coff.h.
3778 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
3779 class setting for a CALLNAME symbol in COFF.
3781 * read.c (potable): Pass negative numbers for new .balign[wl] and
3782 .p2align[wl] pseudo-ops.
3783 (s_align_bytes): Treat a negative argument as specifying the fill
3785 (s_align_ptwo): Likewise.
3789 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
3790 (do_align): Take new len parameter. Change all callers. Pass it
3792 (s_align_bytes): Arg now indicates the length of the fill pattern.
3793 (s_align_ptwo): Likewise.
3794 * config/obj-coff.c (write_object_file): Pass length to
3796 * config/tc-i386.h (md_do_align): Take new len parameter.
3797 * config/tc-m88k.h (md_do_align): Likewise.
3798 * config/tc-m88k.c (m88k_do_align): Likewise.
3799 * config/tc-sh.h (md_do_align): Likewise.
3800 * config/tc-sh.c (sh_do_align): Likewise.
3801 * doc/as.texinfo: Document new pseudo-ops.
3803 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
3804 done in cvt_frag_to_fill.
3806 * config/tc-sh.h (sh_do_align): Declare.
3807 (md_do_align): Define.
3808 * config/tc-sh.c (sh_do_align): New function.
3810 * ecoff.c (ecoff_build_lineno): Don't try to store the address
3811 difference if the next address is before the current one.
3813 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
3814 (archs): Initialize new field.
3815 (m68k_ip): Don't list alias names when listing CPUs which support
3818 * as.c (main): Call parse_args before read_begin.
3819 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
3820 front of generated pseudo-ops.
3821 * read.c (potable): Ignore "name".
3822 (s_app_file): Permit a single quote after the string, since one
3823 may appear in m68k MRI mode.
3825 * configure.in: Check for --enable-shared. If linking against
3826 shared BFD and opcodes, fix library name on SunOS, and try to set
3828 * configure: Rebuild.
3832 * as.h (flag_m68k_mri): Declare.
3833 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
3834 * Many files: For MRI syntax that is specific to the m68k MRI
3835 assembler, check flag_m68k_mri rather than flag_mri or
3836 MRI_MODE_NEEDS_PSEUDO_DOT.
3840 * config/tc-i960.c (ARCH_HX): Define.
3842 (targ_has_sfr): Handle ARCH_HX.
3843 (targ_has_iclass): Handle ARCH_HX.
3844 (tc_coff_fix2rtype): Add return 0 to avoid warning.
3845 (tc_headers_hook): If the architecture was specified explicitly,
3846 use it when setting the flags. Set the extern variable coff_flags
3847 rather than headers->filehdr.f_flags, since the latter is set
3848 unconditionally in obj-coff.c.
3849 (i960_handle_align): Remove unused variable fixp.
3851 Support for building bfd and opcodes as shared libraries, based on
3853 * configure.in: Set OPCODES and BFD to search directories.
3854 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
3855 * configure: Rebuild.
3856 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
3857 (LIBDEPS): New variable.
3858 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
3859 $(HLDFLAGS) in link.
3860 (check): Set LD_LIBRARY_PATH in the environment.
3864 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
3865 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
3866 read-only, not read/write.
3870 * Makefile.in (INSTALL_XFORM): Remove -e.
3874 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
3875 than comparing S_GET_SEGMENT to undefined_section.
3876 (write_object_file): Skip symbols which were equated to an
3877 undefined or common symbol.
3878 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
3879 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3881 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
3882 or S_IS_COMMON call S_GET_VALUE.
3883 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3884 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
3885 to an undefined or common symbol, convert the reloc to be against
3887 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3888 undefined or common symbol.
3889 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3890 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
3891 to an undefined or common symbol, convert the reloc to be against
3893 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3894 undefined or common symbol.
3895 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
3896 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3898 (yank_symbols): Skip symbols which were equated to an undefined or
3903 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
3904 * config/obj-bout.h (S_IS_LOCAL): Likewise.
3906 * configure.in: Make sure we only add m68k-parse.o to
3907 ${extra_objects} once, no matter how many m68k targets have been
3909 * configure: Rebuild.
3913 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
3914 * configure: Rebuild.
3918 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
3919 * doc/c-m68k.texi: Document .extend and .ldouble.
3921 * configure.in (m68*-apple-aux*): New target.
3922 * config/te-aux.h: New file.
3923 * config/obj-coff.c (compare_external_relocs): New static function
3925 (do_relocs_for): Sort relocs if TE_AUX.
3926 (fixup_segment): If TE_AUX, store common symbol value in segment.
3927 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
3931 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
3933 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
3935 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
3936 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
3938 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
3939 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
3940 an undefined symbol and the size of a function symbol.
3941 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
3942 for all common symbols.
3946 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
3949 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
3954 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
3957 * ecoff.c (localsym_t): Add addend field.
3958 (add_ecoff_symbol): Add addend argument. Change all callers.
3959 (coff_sym_value): Make static.
3960 (coff_sym_addend): New static variable.
3961 (ecoff_directive_def): Initialize coff_sym_addend.
3962 (ecoff_directive_val): Accept symbol + constant.
3963 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
3964 (ecoff_build_symbols): Include the addend in the symbol value.
3968 * config/tc-ppc.c (md_assemble): Ignore overflow on
3969 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
3973 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
3974 bytes in size, so offset appropriately in big endian mode when
3975 writing the bottom 2 bytes.
3979 * config/tc-sparc.c (default_compatible): New static local.
3980 (md_begin): Initialize it. Rewrite warn_on_bump handling.
3981 (sparc_ip): If no architecture or -bump specified, don't mark as
3982 mismatched those in default_compatible.
3987 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
3988 * configure: Rebuild.
3989 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
3990 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
3991 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
3992 (sco_id): Declare if SCO_ELF.
3996 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
3997 ({max,warn_after}_architecture): New static locals.
3998 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
3999 If both architecture and -bump requested, set max_architecture to max.
4000 (sparc_md_end): Simplify.
4001 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
4002 Rewrite code to bump architecture and check for conflicts.
4003 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
4004 with Solaris assembler.
4005 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
4006 (md_show_usage): Update.
4010 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
4014 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
4015 #ifdef sparcv9 when choosing value.
4018 (sparc_md_end): Declare.
4019 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
4020 (initial_architecture): New static local.
4021 (can_bump_v9_p): Likewise.
4022 (NO_V9): Delete all occurrences.
4023 (sparc_md_end): New function.
4024 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
4025 Don't bump architecture to v9 unless can_bump_v9_p set.
4026 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
4027 * configure.in (sparc64 target cpu): Don't set obj_format here.
4028 (SPARC_V9): Renamed from sparcv9.
4029 (sparc64-*-elf*): Define SPARC_ARCH64.
4030 * configure: Regenerated.
4031 * acconfig.h (SPARC_V9): Renamed from sparcv9.
4032 (SPARC_ARCH64): Add.
4033 * conf.in: Regenerated.
4034 * config/vmsconf.h: Update.
4038 * config/tc-mips.c (load_register): Optimise "dli" loads.
4039 (md_show_usage): add "-mcpu=vr4100" to help text.
4043 * symbols.c (resolve_symbol_value): If a symbol is equated to an
4044 undefined symbol, preserve the X_op of O_symbol.
4045 (S_GET_VALUE): Fix check to permit this case.
4046 * write.c (write_relocs): If a reloc is against an undefined
4047 symbol equated to another symbol, change the reloc to be against
4049 * config/obj-coff.c (do_relocs_for): Likewise.
4051 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
4056 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
4057 computer is PowerPC.
4059 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
4061 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
4062 mentioned in the eabi.
4066 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
4067 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
4071 * config/tc-ppc.c (mapping): Add more relocation suffixes.
4075 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
4076 as a section it is ok to have unadorned -mrelocatable pointers in.
4080 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
4085 * config/obj-elf.c (obj_elf_section): Add hooks so machine
4086 dependent section attributes can be handled.
4088 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
4089 macros to add support for exclude section flag and ordered section
4092 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
4093 functions to add support for exclude section flag and ordered
4098 * subsegs.c (section_symbol): Don't try to look up the section
4099 symbol in the hash table. It should be possible to have a symbol
4100 with the same name as a section, but no connection to it.
4102 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
4105 * expr.c (operand): Skip whitespace after a close parenthesis.
4110 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
4111 as_warn_where for relocation overflow.
4112 (parse_reg): Accept register name only if next character is
4117 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
4118 addresses for static variables.
4122 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
4123 and xgot give the old toc16 relocation.
4124 (md_apply_fix3): Support all GOT relocations.
4128 * config/tc-mips.c (load_address): Correctly handle a constant in
4133 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
4134 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
4135 (build_Mbytes): Handle REG_B.
4139 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
4143 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
4147 * config/obj-coff.c (write_object_file): Set the s_align field to
4148 the number of bytes, rather than to the power of 2.
4152 * Makefile.in (DISTCLEAN_HERE): New variable.
4153 (distclean): Use it.
4154 (maintainer-clean): Depend upon clean-here rather than clean,
4155 distclean, and clean-info. Run make maintainer-clean in doc.
4156 Remove files listed in DISTCLEAN_HERE.
4157 * doc/Makefile.in (maintainer-clean realclean): Split out from
4158 distclean. Depend upon clean-info and distclean.
4162 * mac-as.r: Fix copyright and version strings.
4163 (cfrg): Use PROG_NAME instead of literal name.
4167 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
4169 * config/tc-a29k.h (tc_unrecognized_line): Define.
4170 * config/tc-a29k.c (a29k_unrecognized_line): New function.
4171 (md_operand): Handle a29k style local dollar labels.
4175 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
4179 * read.c (s_fill): If md_flush_pending_output is defined, call
4184 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
4185 Treat rs_align_code like rs_align.
4189 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
4190 converting an absolute address into a PC-relative one.
4194 * mpw-config.in: Don't always use te-generic.h for emulation.
4195 (powerpc-apple-macos): Use emulation te-macos.h.
4197 (mips-gm-magic): New configuration.
4199 * mpw-make.sed (install, install-only): Edit in Mac-specific
4204 * configure.in: Improve message about unsupported ELF targets.
4205 * configure: Rebuild.
4207 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
4208 pattern is zero. From Manfred Hollstein.
4212 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
4213 the various idata sections, we check the name on the .section pseudo.
4217 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
4218 add section's paddr to add_number; compatibility to native as and
4223 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
4225 * hash.c (struct hash_entry): Moved here...
4226 * hash.h (struct hash_entry): ...from here.
4228 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
4229 if it's already known to be null.
4233 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
4234 bit relocs converted to PC relative relocs. This turned out to
4235 add the offset from the beginning of .text twice.
4239 * stabs.c (s_stab_generic): In 's' case, free string from
4242 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
4244 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
4247 * subsegs.h (segment_info_type): Make bitfields unsigned.
4249 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
4250 and move them together. On most systems this will reduce the
4251 structure size by one word.
4252 (operatorT): Define O_max.
4253 * expr.c (expr_begin): Verify that X_op is wide enough to hold
4256 * read.c (pop_insert): Print error returned by hash table
4259 * as.c (dump_statistics): Split out from main; dump some hash
4260 table stats and target-specific stats.
4261 (start_time): No longer automatic to main.
4262 (main): Set file-level start_time and call dump_statistics at
4263 exit. Exit by calling xexit.
4264 (show_usage): Make --statistics description less specific.
4265 * subsegs.c (subsegs_print_statistics): New function.
4266 * write.c (write_print_statistics): New function.
4267 (n_fixups): New static variable.
4268 (fix_new_internal): Increment it.
4269 * read.c (read_print_statistics): New function.
4270 * read.h (read_print_statistics): Declare.
4271 * symbol.c (symbol_print_statistics): New function.
4272 * symbol.h (symbol_print_statistics): Declare.
4273 * hash.c (hash_print_statistics): New function.
4274 * hash.h (hash_print_statistics): Declare.
4275 * config/tc-i386.c (i386_print_statistics): New function.
4276 * config/tc-i386.h (i386_print_statistics): Declare.
4277 (tc_print_statistics): New macro.
4278 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
4281 * hash.c (DELETED): Rewrite to use a valid but unique address.
4282 (START_POWER): Reduce to 10.
4283 (enum stat_enum): New enumerator, replacing STAT_* index macros.
4284 Add new values for counting strcmp calls.
4285 (GROW_FACTOR): New macro.
4286 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
4288 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
4289 (hash_new): Use FULL_VALUE.
4290 (struct hash_control): Definition moved here.
4291 (hash_code): Don't mask to low bits.
4292 (hash_ask): Mask returned hash code here. Check hash value before
4293 calling strcmp; count strcmp calls.
4294 * hash.h (struct hash_control): Declare, don't define, here.
4295 (HASH_STATLENGTH): Deleted.
4296 (struct hash_entry): Add field for hash code.
4297 (hash_say, hash_apply): Don't declare.
4299 * hash.c (destroy): Return void.
4300 (applicatee): Ditto.
4301 (main): Fix declarations.
4302 (hash_apply): Return void. Argument `function' returns void. Put
4303 inside "#ifdef TEST".
4304 (hash_say): Define only if TEST is defined.
4305 * hash.h (hash_apply, hash_say): Declarations deleted.
4309 * configure: Rebuild with autoconf 2.7.
4313 * aclocal.m4 (AC_PROG_CC): Remove local definition.
4314 * configure: Rebuild with autoconf 2.6.
4318 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
4319 (ppc_stabx): Don't increment ppc_debug_name_section_size.
4321 (ppc_frob_file): Remove.
4322 * config/tc-ppc.h (tc_frob_file): Don't define.
4323 (ppc_frob_file): Don't declare.
4327 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
4328 * config/alpha-opcode.h: Include one-operand variants of jmp and
4331 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
4332 with te-delta.h in 8.3 file systems.
4333 * configure.in: Adjusted.
4337 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
4339 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
4341 * expr.c (operand): Handle MRI suffixes after unadorned 0.
4345 Version 2.6 released.
4346 * Makefile.in (VERSION): Updated to 2.6.
4348 * config/obj-coff.c (write_object_file): Change use of md_do_align
4349 to pass a pointer rather than a fill value, to match other uses.
4353 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
4355 * as.c (main): Move md_end call to just after call to
4356 perform_an_assembly_pass. Delete cpu-specific code here.
4357 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
4358 * config/tc-arm.c (md_end): Unused function deleted.
4359 * config/tc-ns32k.c (md_end): Ditto.
4361 * config/tc-i386.c (i386_align_code): New function, moved here
4362 from HANDLE_ALIGN macro.
4363 * config/tc-i386.h (HANDLE_ALIGN): Call it.
4367 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
4372 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
4373 multi-byte nops to avoid changing gcc. The necessary gcc change
4374 might break old assemblers.
4378 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
4379 generate optimal multi-byte nop instructions for ".align n"
4380 ".align n,0x90", and aligns requiring more than 15 bytes of
4381 padding still generate multiple 0x90's as before.
4385 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
4389 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
4390 * configure: Update.
4391 * mpw-make.sed: Reorder commands to make sed happier.
4392 * config/te-macos.h: New file.
4393 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
4397 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
4398 conditional branches.
4399 (md_apply_fix): Likewise.
4403 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
4404 generate a reloc which the linker will not handle correctly. Fix
4405 overflow checking--R_IREL is 18 bits, not 17.
4409 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
4410 from a PC relative reloc if TC_A29K.
4412 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
4413 code depends upon it.
4415 * config/tc-a29k.c (md_operand): Handle $float, $double, and
4416 $extend. Based on code from Eric Freudenthal
4418 * config/tc-a29k.h (LEX_DOLLAR): Define.
4419 * read.c (LEX_DOLLAR): Define if not defined.
4420 (lex_type): Use LEX_DOLLAR.
4424 * configure.in (a29k-nyu-sym1): New target, just like other a29k
4429 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
4433 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
4434 used only by "dlltool.c".
4438 * config/tc-alpha.c: Undefine inline if not __GNUC__.
4439 (md_pseudo_table): Don't define "extern".
4443 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
4445 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
4449 * macro.c (macro_expand_body): Don't warn about == with a
4450 nonexistent parameter, in case it is in a comment field.
4452 * as.c (main): On TC_A29K, call macro_init with macro_alternate
4454 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
4455 set, even if macro_alternate is set.
4456 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
4457 (sub_actual): If macro_strip_at, and kind is '@', don't look up
4458 the token unless it ended in '@'.
4459 * config/tc-a29k.c (line_separator_chars): Remove '@'.
4460 * doc/c-a29k.texi: Document macro usage on A29K.
4464 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
4466 (install_operand): Likewise.
4468 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
4472 * configure.in (m88k-motorola-sysv*): New target.
4473 * configure: Rebuild.
4474 * config/te-delta88.h: New file.
4475 * config/obj-coff.c (write_object_file): Use md_do_align if it is
4477 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
4478 (md_do_align): Define.
4479 * config/tc-m88k.c: Include "subsegs.h".
4480 (m88k_do_align): New function.
4482 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
4483 (COFF_NOLOAD_PROBLEM): Define.
4484 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
4488 * config/tc-i386.c (md_assemble): For a jump instruction with
4489 non-constant target, require 7 available bytes in the current
4494 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
4499 * configure.in (alpha-*-linux*): Use ecoff.
4500 * configure: Rebuild.
4501 * ecoff.c (ecoff_directive_extern): New function.
4502 (ecoff_directive_weakext): New function.
4503 (ecoff_build_symbols): Handle weak symbols.
4504 (ecoff_setup_ext): Likewise.
4505 (ecoff_frob_symbol): Warn about weak common symbols.
4506 * ecoff.h (ecoff_directive_extern): Declare.
4507 (ecoff_directive_weakext): Declare.
4508 * symbols.c (S_IS_WEAK): New function.
4509 * symbols.h (S_IS_WEAK): Declare.
4510 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
4512 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
4517 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
4519 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
4520 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
4521 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
4522 * config/obj-coff.c (coff_frob_symbol): Don't call
4523 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
4524 one. If OBJ_XCOFF, try to figure out whether the symbol is going
4527 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
4528 (ppc_stab_symbol): New static variable.
4529 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
4530 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
4531 sy_tc.real_name to the stab string.
4532 (ppc_bc, ppc_ec): New static functions.
4533 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
4535 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
4537 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
4540 * input-scrub.c (input_scrub_push): Reset sb_index.
4544 * listing.c (listing_newline): Don't create a frag in the absolute
4549 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
4550 (ppc_csect): Move most of the code to ppc_change_csect, and call
4552 (ppc_change_csect): New static function, taken from ppc_csect.
4553 (ppc_section): New static function.
4554 (ppc_saw_abs): New static varable.
4555 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
4556 a symbol has no csect.
4557 (ppc_adjust_symtab): New function.
4558 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
4559 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
4561 * write.c (write_object_file): If tc_adjust_symtab is defined,
4562 call it just before the call to obj_adjust_symtab.
4564 * symbols.c (symbol_find_or_make): Change name to be const.
4565 * symbols.h (symbol_find_or_make): Update declaration.
4569 * doc/as.texinfo (Align): Mention SH.
4570 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
4572 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
4573 (Z8000 Directives, even): Don't give numeric align value, instead
4578 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
4583 * Makefile.in (diststuff): Also make info.
4584 (maintainer-clean realclean): Also make clean-info.
4588 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
4589 (pa_subspace): For ".nsubspa", always create a new subspace
4590 with the given attributes, even if one already exists with the
4595 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
4596 TC_FORCE_RELOCATION, taking an additional section argument. If
4597 the section of the target symbol is not the same as the current
4598 section, always force the relocation to be used.
4599 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
4601 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
4602 md_pcrel_from function, taking an additional section argument.
4603 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
4605 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
4606 older TC_FORCE_RELOCATION if not defined.
4607 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
4608 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
4609 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
4610 TC_FORCE_RELOCATION.
4614 * input-scrub.c (as_where): Set name to null pointer if we don't
4616 * messages.c (identify): Only print filename if non-null.
4617 (as_show_where): Ditto, for line number too.
4618 (as_warn_internal, as_bad_internal): Ditto.
4620 * input-file.c (input_file_open): If the input file can't be
4621 opened, consider it an error.
4625 * config/tc-mips.c: Added mips_4100 control, and support for
4626 accepting the 4100 as a MIPS architecture variant (md_begin,
4627 macro_build, mips_ip, md_parse_option). Adding suitable
4628 command-line OPTIONs, and updating the help text (md_show_usage).
4632 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
4633 BFD_ASSEMBLER is defined.
4635 * Use one active frag and one obstack per frag chain:
4636 * frags.c (frags): Variable deleted.
4637 (frag_alloc): New function.
4638 (frag_grow, frag_more, frag_variant, frag_now_fix,
4639 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
4641 (frag_new): Ditto. Verify that frch_last and frag_now match on
4642 entry and exit, and that old frag_now has non-zero type. Replace
4643 "know" uses with "assert". Use frag_alloc instead of mucking with
4645 * frags.h (frags): Declaration deleted.
4646 * subsegs.h (struct frchain): Add new field frch_frag_now.
4647 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
4649 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
4650 give it any stricter alignment than frchainS structures need. Do
4651 not initialize frags obstack. Set frag_now to point to
4652 dummy_obstack. Initialize absolute_frchain.
4653 (subseg_set_rest): Save and restore frag_now in frch_frag_now
4654 field of frchainS. Don't create new frags on section switch, and
4655 use frag_alloc when creating a new frag chain. For absolute
4656 section, set frchain_now to absolute_frchain. Verify that
4657 frch_last and frag_now match on entry and exit. Initialize
4658 per-chain obstack, and under gcc, set required alignment to that
4659 needed by fragS structure.
4661 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
4663 * stabs.c (get_stab_string_offset): Only copy input string if a
4664 fresh copy is needed, not if the section already exists.
4665 (s_stab_generic): Cache stab section name to bypass lookups, since
4666 usually it will match. Could be made faster still by changing the
4667 memory allocation rules.
4668 (s_xstab): Cache section name to bypass repeated string
4671 * frags.c (frag_new): Deleted register declarations.
4673 * listing.c (frag_now): Don't declare.
4675 * as.c (chunksize): New variable.
4676 (debug_memory): New variable.
4677 (main): If debug_memory is set, reduce chunksize and
4679 * as.h (chunksize): Declare it.
4680 * read.c (read_begin): Use it.
4682 * config/tc-alpha.c (md_shortopts): Include 'g'.
4683 (md_parse_option): Ignore it.
4685 * Makefile.in (distclean): Remove Makefile and config.status from
4686 testsuite directory.
4687 (clean-here): Don't delete testsuite. Instead, delete only the
4688 files within it that would be generated by running tests.
4690 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
4691 frag_now_fix instead of accessing obstack info directly.
4692 * config/tc-arm.c (s_ltorg): Ditto.
4693 (md_assemble): Ditto.
4695 * config/tc-i386.c (md_assemble): Call frag_grow instead of
4700 * stabs.c (aout_process_stab): Insert debug symbol into symbol
4701 chain after parsing value expression, if any, to avoid separating
4702 continued .stabs lines.
4706 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
4707 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
4711 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
4712 -mcom support and PPC_OPCODE_ANY for -many.
4713 (md_parse_option): Ditto.
4715 (md_begin): For duplicate instructions, print all duplicates
4720 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
4721 common mode operation.
4722 (md_show_usage): Add -mcom to usage message.
4726 * expr.c (op_rank): Add O_symbol_rva.
4727 * expr.h (operatorT): Add O_symbol_rva.
4728 * read.c (cons_worker): Set O_symbol_rva when necessary.
4729 * write.c (fix_new_exp): Understand O_symbol_rva.
4733 * config/tc-mips.c: Correct s_cons arguments. From Michael
4738 * config/tc-ppc.c (ppc_macro): Make count unsigned.
4739 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
4740 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
4742 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
4743 make non-static. Change all users.
4744 * config/obj-coff.h (coff_n_line_nos): Declare.
4750 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
4752 * config/obj-elf.c (obj_elf_common): Convert specified byte
4753 alignment to power of two. Set size of local bss symbol.
4755 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
4759 * sb.c, macro.c: Decide whether to include <string.h> or
4760 <strings.h> just as as.h does.
4764 * Makefile.in (site.exp): Fix setting of $srcdir.
4766 * config/tc-arm.c (md_atof): Fix little-endian output.
4767 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
4772 * doc/as.texinfo: Split out the NS32k family documentation,
4773 despite its being commented out for now.
4774 * doc/c-ns32k.texi: New file.
4776 * sb.c, macro.c: Include string.h.
4778 * Makefile.in (comparison): Only check *.o; we don't care if
4779 timestamps inserted by the native linker differ.
4781 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
4782 if alignment stricter than 4 bytes is requested; in that case,
4783 align to a 4-byte boundary first.
4787 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
4788 uninitialized" warning for `Max_Source_Offset'.
4792 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
4793 qualifiers and detect errors if present.
4794 (md_assemble): Add call to parse_toc_entry. Also added some support
4795 for the [tocv] qualifier.
4796 (ppc_pe_tocd): New function to support data in the toc section.
4800 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
4805 * config/tc-m68k.c (isword): Accept all values from -65536 to
4806 +65535, so ~VAL will not be rejected.
4808 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
4809 skip characters after the pseudo-op.
4811 * read.c (get_line_sb): Don't look past buffer_limit.
4812 (s_include): In MRI mode, skip characters after the file name.
4816 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
4817 periods in register names.
4819 For older changes see ChangeLog.1.