3 * config/tc-arm.c (md_apply_fix3): Correct handling of ADRL when
8 * configure.in (AC_ISC_POSIX): Put after AC_CANONICAL_SYSTEM.
13 * atof-generic.c: Fix formatting.
14 * config/tc-mips.c: Likewise.
15 * config/tc-vax.c: Likewise.
16 * input-scrub.c: Likewise.
20 * config/tc-sh.h (TARGET_FORMAT): Use sh-linux targets.
21 * configure.in (sh-*-linux*): Added.
26 * config/tc-hppa.c: Fix formatting.
28 * ecoff.c: Fix formatting.
32 * configure: Rebuilt with new libtool.m4.
36 * cgen.c: Fix formatting.
37 * config/tc-ia64.c: Likewise.
41 * aclocal.m4: Regenerate.
42 * config.in: Regenerate.
43 * configure: Regenerate.
44 * po/gas.pot: Regenerate.
48 * config/tc-cris.c: Correct comment typos.
52 * config/tc-mn10300.c: Cleanup.
53 (md_pcrel_from): Enable.
57 * expr.c (operand): Fix a comment typo.
58 * write.c (write_relocs): Fix a signed/unsigned warning.
60 * config/tc-hppa.c (fudge_reg_expressions): New
61 (hppa_force_reg_syms_absolute): New.
62 (pa_equ): Allow reg_section expressions.
63 * config/tc-hppa.c (md_optimize_expr): Define.
64 (hppa_force_reg_syms_absolute): Prototype.
66 * config/tc-hppa.c (pa_11_fp_reg_struct): Delete.
67 (pa_parse_number): Pass in arg to select fp reg parsing.
68 Return 1 to indicate format checks pass. If strict, then only
69 accept a register or register symbol. Return value in...
70 (pa_number): New static for pa_parse_number.
71 (FP_REG_BASE): Define.
72 (FP_REG_RSEL): Define.
73 (pre_defined_registers): Apply FP_REG_BASE and FP_REG_RSEL as
74 appropriate. White space changes.
75 (need_pa11_opcode): Don't bother passing any params, get them from
77 (pa_ip): Modify all calls to pa_parse_number and need_pa11_opcode.
78 Remove extraneous check in case 'Q'.
79 (pa_equ): Modify call to pa_parse_number to do strict parsing. If
80 reg, set section of resulting symbol to reg_section.
81 (pa_parse_space_stmt): Modify call to pa_parse_number.
84 * config/tc-hppa.c: (md_apply_fix): Handle vtable relocs.
85 (hppa_force_relocation): Handle vtable relocs.
86 (pa_vtable_entry): New.
87 (pa_vtable_inherit): New.
88 (md_pseudo_table): Add entries for vtable pseudos.
89 (hppa_fix_adjustable): Reject reduction of R_PARISC_GNU_VTINHERIT
90 and R_PARISC_GNU_VTENTRY relocs. Reject reduction of relocs
92 (tc_gen_reloc): Remove ELF_ARG_RELOC_INSN code.
93 (pa_type_args): Don't call symbol_get_bfdsym multiple times.
94 Set STT_PARISC_MILLICODE for OBJ_ELF when encountering a
96 * config/obj-elf.c (obj_elf_type): Allow md_elf_symbol_type to
97 specify a symbol type.
99 * config/tc-hppa.h: Reorganize file a little, grouping OBJ_ELF
100 dependent things together.
101 (md_elf_symbol_type): Define.
103 * config/tc-hppa.c (fix_new_hppa): Elide "$PIC_pcrel$0" pseudo
105 * config/tc-hppa.h (tc_frob_symbol): Elide "$PIC_pcrel$0" here too.
107 * config/obj-elf.h (obj_elf_vtable_inherit): Declare.
108 (obj_elf_vtable_entry): Declare.
110 * config/obj-elf.c (obj_elf_vtable_inherit): Return struct fix *
112 (obj_elf_vtable_entry): Similarly.
113 (elf_pseudo_table): Fix the damage with a cast.
117 * config/tc-ia64.c (emit_one_bundle): Stop collecting insns
118 for template selection when a label is needed.
122 * config/tc-ia64.c: Fix formatting.
126 * configure.in: Increase version number to 2.10.91.
127 * configure: Regenerate.
128 * aclocal.m4: Regenerate.
129 * config.in: Regenerate.
130 * po/gas.pot: Regenerate.
131 * Makefile.in: Regenerate.
135 * config/tc-sh.h [OBJ_ELF] (TC_FIX_ADJUSTABLE): Define.
136 * config/tc-sh.c (md_apply_fix): Map 32-bit relocations that
137 become PC-relative to BFD_RELOC_32_PCREL. Reject 16- or 8-bit
139 (sh_obj_adjustable): Return 1 for PC-relative offsets used in
144 * config/tc-sh.h (DIFF_EXPR_OK, GLOBAL_OFFSET_TABLE_NAME,
145 TC_RELOC_GLOBAL_OFFSET_TABLE, TC_RELOC_RTSYM_LOC_FIXUP): Define.
146 * config/tc-sh.c (sh_elf_cons, sh_elf_suffix): New functions.
147 [OBJ_ELF] (md_pseudo_table) <long, int, word, short>: Use them.
148 (GOT_symbol): New variable.
149 (md_undefined_symbol): Set it.
153 * config/tc-ia64.c (match): Don't inline.
154 (extra_goodness): New.
155 (md_begin): Prefer nop.f and nop.b for best_template.
159 * as.c: Fix formatting.
166 * config/tc-mn10300.c: Cleanup and fix warnings.
167 (md_pseudo_table): Add initializers.
168 (md_show_usage): Cleanup.
169 (md_parse_option): Fix warnings.
170 (md_undefined_symbol): Fix warnings.
171 (md_conver_frag): Fix warnings.
172 (tc_gen_reloc): Fix warnings.
173 (md_apply_fix3): Fix warnings.
174 (check_operand): Fix warnings.
178 * acinclude.m4: Include libtool and gettext macros from the
180 * aclocal.m4, configure: Rebuilt.
184 * config/tc-ppc.c (md_parse_option): Recognize -m405.
188 * listing.c: Fix formatting.
192 * app.c: Fix a comment typo. Fix formatting.
196 * tc-vax.c (md_convert_frag): Correctly calculate the pc relative
197 offset of the target destination for jmp instructions.
198 (md_assemble): Change mode to VAX_ABSOLUTE_MODE as per comments.
202 * NEWS: Mention support for CRIS.
206 * config/tc-avr.h (TC_IMPLICIT_LCOMM_ALIGNMENT): New macros.
207 Sets `.lcomm' alignment to zero.
211 * config/tc-i386.h <OBJ_MAYBE_ELF, OBJ_MAYBE_COFF>
212 (TC_FIX_ADJUSTABLE): Define.
216 * config/tc-ia64.c (output_unw_records): Set U & E flags only if
217 unwind.personality_routine is set.
221 * write.c (TC_FIX_ADJUSTABLE): Remove the duplicate.
225 * config/tc-i386.h (TC_FIX_ADJUSTABLE): Do *NOT* define if target
230 * config.in (STRICTCOFF): New for strict COFF.
232 * configure.in: Define STRICTCOFF for i386-*-msdosdjgpp*,
233 i386-*-go32* and i386-go32-rtems*.
234 * configure: Rebuilt.
236 * config/obj-coff.c (obj_coff_endef): Follow the historical
237 behavior if STRICTCOFF is not defined.
239 * doc/internals.texi: Document STRICTCOFF.
243 * write.c (TC_FIX_ADJUSTABLE): Define to 1, if not defined.
244 (fixup_segment) Use it instead of TC_DONT_FIX_NON_ADJUSTABLE.
245 * config/tc-i386.h (TC_DONT_FIX_NON_ADJUSTABLE): Remove.
246 <OBJ_ELF, OBJ_COFF, TE_PE> (TC_FIX_ADJUSTABLE): Define.
247 * config/tc-arm.h (TC_DONT_FIX_NON_ADJUSTABLE): Remove.
248 <OBJ_ELF> (TC_FIX_ADJUSTABLE): Define.
249 * config/tc-i960.h, config/tc-m68k.h, config/tc-v850.h:
254 * config/tc-mn10300.c: (md_apply_fix): New function.
255 (mn10300_force_relocation): New function.
256 (mn10300_fix_adjustable): New function.
258 * config/tc-mn10300.h: (TC_FORCE_RELOCATION): Define.
259 (TC_HANDLES_FX_DONE): Define.
260 (obj_fix_adjustable): Define.
261 (MD_APPLY_FIX3): Define.
262 (TC_LINKRELAX_FIXUP): Define.
264 * write.c: (TC_LINKRELAX_FIXUP): Define if not
266 (fixup_segment): Use TC_LINKRELAX_FIXUP.
268 * doc/internals.texi: Document TC_LINKRELAX_FIXUP.
272 * config/tc-i860.c (md_apply_fix3): Do not insert the immediate
273 if the fixup resulted in a relocation.
277 * config/tc-arm.c (decode_shift): Replace as_tsktsk with as_warn.
278 Make reference to first element of shift_names explicit.
282 * write.c (fixup_segment) [TC_DONT_FIX_NON_ADJUSTABLE]: Use
283 obj_fix_adjustable() and tc_fix_adjustable() to tell whether to
284 add a symbol's address. Removed all target-specific #ifdefs that
285 used to accomplished the same.
286 * config/tc-v850.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
287 * config/tc-m68k.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
288 * config/tc-arm.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
289 * config/tc-i960.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
290 * config/tc-i386.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
294 * dwarf2dbg.c: Fix formatting.
298 * config/tc-arm.c (decode_shift): Allow illegal shifts by zero
299 to be recoded as logical shift lefts by zero.
303 * config/tc-ia64.c (specify_resource, case IA64_RS_GR): Handle
304 postincrement modified registers. Handle IA64_OPND_R3_2 addl
306 (note_register_values): Handle IA64_OPND_R3_2 operands.
310 * config/tc-i860.c (md_operand): Silly typo fixed.
314 * config/tc-arm.c (struct asm_shift): Delete.
316 (enum asm_shift_index): New.
317 (struct asm_shift_properties): New.
318 (struct asm_shift_name): New.
319 (shift_properties[]); New.
320 (shift_names[]); New.
322 (decode_shift): Use new structures.
323 Issue a warning is "ROR #0" is used.
324 Issue a warning if "ASR #0" or "LSR #0" is used.
326 (md_begin): Initialise arm_shift_hsh table from new
327 asm_shift_name array.
331 * config/tc-sparc.c: Kill all warnings.
332 (md_parse_option): Set -32/-64 for -xarch=, allow all -A archs
334 (md_show_usage): Update usage text.
338 * config/tc-arm.c (do_bx): Warn about "bx px" not being very
343 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Defined.
345 * config/tc-sh.c (md_assemble): Changed so debug_type
346 test performed for ppi_assemble
347 * config/tc-sh.c: Included dwarf2dbg.h.
348 (debug_line): Defined.
349 (md_assemble): Generates dwarf2 line info.
350 (sh_finalize): New function. Finalize dwarf2 info.
351 (assemble_ppi): Returns size of code generated.
352 (build_Mytes): Returns size of code generated.
353 (md_pseudo_table): Added "file" and "loc" psuedo ops.
354 * config/tc-sh.h (md_end): Defined.
355 (sh_finalize): Declared.
359 * config/tc-sh.c (md_apply_fix) [BFD_RELOC_32, BFD_RELOC_16]: Use
364 * config/tc-arm.c (do_bx): Allow "bx pc".
368 * config/tc-ia64.c (md_longopts): Add -mconstant-gp and -mauto-pic.
369 (md_parse_option): Add OPTION_MCONSTANT_GP and OPTION_MAUTO_PIC.
370 (md_begin): Change assignment to md.flag to OR in the new bit.
374 * config/obj-coff.c (obj_coff_endef) [BFD_ASSEMBLER]: Set the debug
375 flag for storage types C_ARG, C_REGPARM, C_FIELD, C_MOS, C_MOE,
380 * NEWS: Mention i860 support.
384 * config/tc-mn10300.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
388 * doc/c-arm.texi (ARM Directives): Fix warnings from makeinfo.
392 * doc/c-i860.texi (Opcodes for i860): Remove braces from @item
397 * expr.c: Fix formatting.
398 * config/obj-bout.c: Likewise.
402 * doc/c-i860.texi: Flesh out the i860 section more.
406 * symbols.c: Fix formatting.
411 * tc-i386.c (md_assemble): Skip suffix check if the opcode modifier
412 has the IgnoreSize bit set.
417 * configure.in: Use elf on Unixware 7 (i586-sco-sysv5uw7.1.0)
418 * configure: Regenerate.
422 * hash.c: Fix formatting.
427 * config/tc-i860.h: Rework completely for BFD_ASSEMBLER.
428 (i860_fix_info): New enum.
429 (MD_APPLY_FIX3): Define.
430 (WORKING_DOT_WORD): Define.
431 (TC_HANDLES_FX_DONE): Define.
432 (DIFF_EXPR_OK): Define.
433 (LISTING_HEADER): Define.
434 (TARGET_FORMAT): Select target format based on endian flag.
435 (TARGET_BYTES_BIG_ENDIAN): Default to little endian.
436 (target_big_endian): Add external declaration.
438 * config/tc-i860.c: All existing code reworked completely. Other
439 new code shown below.
440 (SYNTAX_SVR4): Define.
441 (target_warn_expand): New variable.
442 (md_shortopts): Declare and define (-Qy, -Qn, and -V options).
443 (md_longopts): Declare and define with new options (-EL, -EB,
445 (md_show_usage): New function.
446 (md_operand): New function.
447 (obtain_reloc_for_imm16): New function.
448 (md_apply_fix3): New function.
449 (tc_gen_reloc): New function.
453 * config/tc-m68hc11.c (build_jump_insn): Make sure the
454 2 bytes of the jump address are in the same frag.
455 (find): Accept 68hc12 register indirect modes.
457 * NEWS: Mention 68HC11 & 68HC12 support.
461 * config/tc-ia64.c (unwind): Add prologue_mask member.
462 (dot_vframe): Elide psp_gr record if it overlaps prologue_gr.
463 (dot_save): Likewise for pfs_gr, rp_gr, and preds_gr.
464 (dot_body): Clear unwind.prologue_mask.
465 (dot_prologue): Set it. Accept a register second argument.
469 * config/atof-ieee.c: Fix formatting.
470 * config/atof-tahoe.c: Likewise.
474 * config/tc-sparc.c (md_begin): Fix typo in recent formatting
477 * doc/as.texinfo (Pseudo Ops): Update to include descriptions
478 of .popsection, .previous, .pushsection, .subsection,
479 .version, .vtable_entry, .vtable_inherit and .weak.
483 * config/tc-cris.c: Fix formatting.
484 * config/tc-i386.c: Likewise.
485 * config/tc-sparc.c (sparc_ip): Simplify the code.
489 * config/tc-cris.c: Rearrange code for readability.
490 * config/tc-d10v.c: Fix formatting.
491 * config/tc-m32r.c: Likewise.
492 * config/tc-sparc.c: Likewise.
496 * config/tc-ia64.c (emit_one_bundle): Call ia64_free_opcode
497 before ia64_find_opcode.
498 (md_assemble): Likewise.
502 * config/tc-arm.c (do_mrs): Fix skip of 'cpsr_all' flag.
503 Undo some formatting fixes.
507 * config/obj-som.c: Fix formatting.
508 * config/obj-ieee.c: Likewise.
509 * config/tc-arm.c: Likewise.
510 * config/tc-v850.c: Likewise.
514 * doc/c-m68k.texi (section M680x0 Options): Turn into a table
515 index by command line option.
519 * doc/c-m68k.texi (@cindex @samp{--pcrel}): Rewrite option description.
520 (@node M68K-Branch): Rewrite to match the reality.
524 * doc/c-i860.texi: New file.
525 * doc/Makefile.am (CPU_DOCS): Add c-i860.texi.
526 * doc/Makefile.in: Regenerate.
527 * doc/all.texi: Add I860 as relevant architecture.
528 * doc/as.texinfo: Include i860 dependent file c-i860.texi.
532 * config/tc-d30v.c: Fix formatting.
536 * config/tc-m68k.c (flag_keep_pcrel, OPTION_PCREL): Add --pcrel option.
537 (md_convert_frag_1, md_estimate_size_before_relax): When making DBcc
538 long emit a long branch if available instead of an absolute jump, never
539 emit absolute jumps for anything with --pcrel.
541 * doc/c-m68k.texi: Document new command line option.
545 * config/tc-avr.c: Use PARAMS macro in function declarations.
546 Don't declare md_pcrel_from_section (already in tc-avr.h).
547 (avr_operands): Use AVR_UNDEF_P and AVR_SKIP_P macros.
548 (avr_operand): Don't set (unsigned) op_mask to -1.
552 * configure.in: Add bits for i860-stardent-{sysv4, elf}*.
553 * configure: Regenerated.
554 * config/obj-elf.c (obj_elf_type): Recognize a fifth type
555 of operand to the .type directive (.e.g, "type").
559 * as.h (warn_comment, found_comment, found_comment_file): Declare.
560 * app.c (do_scrub_chars): Record where first comment found.
561 * read.c (read_a_source_file): Init found_comment on entry, and
562 notify whether comments found on exit.
563 * config/tc-hppa.c (md_shortopts): Add "c".
564 (md_longopts): Add warn-comment.
565 (md_parse_option): Handle it.
566 (md_show_usage): Show available options.
567 * config/tc-hppa.h (WARN_COMMENTS): Define if TE_LINUX
571 * config/tc-mn10300.c (md_convert_frag): Fix printfs.
572 (tc_gen_reloc): Add cast when assigning bfd_abs_symbol to
574 (md_estimate_size_before_relax): Don't fall off end of function.
578 * config/tc-avr.c: Fix formatting.
579 * config/tc-ns32k.c: Likewise.
583 * config/tc-d10v.c (find_opcode): Remove extraneous `='.
587 * config/tc-d10v.c: Fix formatting.
588 * config/tc-z8k.c: Likewise.
589 * config/tc-sparc.c: Likewise.
593 * cgen.c (queue_fixup): Declare opinfo.
594 (gas_cgen_parse_operand): Mark unused parameters with ATTRIBUTE_UNUSED.
595 (gas_cgen_md_operand): Ditto.
596 (gas_cgen_md_apply_fix3): Ditto.
600 * config/obj-coff.c (obj_frob_symbol): Don't merge
601 labels. Don't merge if the symbol isn't constant. Return
602 immediately if a symbol is merged.
606 * frags.c (frag_align): Correct absolute section alignment.
610 * config/obj-coff.c (obj_frob_symbol): revert previous change,
611 it breaks linking against DLLs.
615 * configure.in: Add CRIS support.
616 * configure: Regenerate.
617 * Makefile.am: (CPU_TYPES): Add cris.
618 (CPU_OBJ_VALID) [aout]: Add cris.
619 (MULTI_CPU_TYPES): Add cris.
620 (MULTI_CPU_OBJ_VALID) [aout]: Add cris.
621 [coff]: Only i386 and mips are valid.
622 (TARGET_CPU_CFILES): Add config/tc-cris.c.
623 (TARGET_CPU_HFILES): Add config/tc-cris.h.
624 (MULTI_CFILES): Add config/e-crisaout.c and config/e-criself.c.
625 Regenerate dependencies.
626 * Makefile.in: Regenerate.
627 * aclocal.m4: Regenerate.
628 * as.c: Declare crisaout, criself.
629 * config/tc-cris.h, config/tc-cris.c: New.
630 * config/e-criself.c, config/e-crisaout.c: New.
631 * po/POTFILES.in, po/gas.pot: Regenerate.
635 * read.c: Fix formatting.
636 * write.c: Fix formatting.
640 * sb.c: Include <stdlib.h> if exists for abort ().
644 * config/tc-hppa.c (hppa_fix_adjustable): Correct LR%/RR% comment.
645 (arg_reloc_stub_needed): #ifdef OBJ_SOM, not #ifdef SOM.
646 (pa_type_args): Same here.
650 * config/obj-coff.c (obj_frob_symbol): Don't merge labels. Don't
651 merge if the symbol isn't constant. Don't call S_SET_EXTERNAL if
652 the storage class is already set.
656 * config/tc-m68hc11.c: Fix formatting.
657 * config/tc-mn10200.c: Likewise.
658 * config/tc-mn10300.c: Likewise.
659 * config/tc-pj.c: Likewise.
660 * config/tc-tic80.c: Likewise.
661 * config/tc-w65.c: Likewise.
665 * expr.c (operand): Permit %bin literals if LITERAL_PREFIXPERCENT_BIN
670 * doc/c-mips.texi (MIPS Opts): Remove erroneous space after
675 * config/tc-hppa.c (hppa_fix_adjustable): Use the same checks for
676 ELF as are used for SOM (except the 32-bit reloc one) to decide
677 whether a symbol can be reduced to a section symbol. Expand on
678 the comment for symbols involved in LR% and RR% expressions.
682 * config/tc-mips.c (mips_disable_float_construction): New
683 static variable. Set to true if doubles should not be
684 constructed by loading two single width fp registers with
686 (mips_ip): Test mips_disable_float_construction.
687 (md_longopts): Add command line switches --construct-floats
688 and --no-construct-floats.
689 (md_parse_option): Parse new command line options.
690 (md_show_usage): Describe new command line options.
692 * doc/c-mips.texi: Document new command line options.
696 * configure.in: Remove the test /usr/dde for mips-*-sysv4*MP*
697 * configure: Regenerate.
701 * configure.in (DEFAULT_EMULATION setting): Revert part of
702 2000-07-01 change that set te_multi=multi unless set to tmips.
703 * configure: Regenerate.
707 * config/obj-coff.c (coff_frob_section): Add padding to the last
708 section when aligning it increases its size.
712 * config/tc-tic54x.c: Fix formatting.
716 * config/tc-h8500.c: Fix formatting.
717 * config/tc-tic54x.c: Fix formatting.
721 * config/tc-hppa.c (md_apply_fix): Check fmt 12 and 22 pc-rel
722 displacements correctly.
724 * read.h (s_abort): Add ATTRIBUTE_NORETURN.
728 * hash.c (hash_insert): Add cast to obstruct_alloc to fix
735 * config/te-hppalinux64.h: Add a new emulation.
736 * configure.in (emulations): Add configure bits to support the
737 64-bit Linux/parisc target.
738 * configure: Regenerate.
740 * config/tc-hppa.c (md_parse_option): Support `-V' for ELF.
741 (md_shortopts): Add `V' for ELF.
743 * config/tc-hppa.c (arg_reloc_stub_needed): Define as zero except
744 when SOM or ELF_ARG_RELOC are defined.
745 (pa_type_args): Only set symbol_arg_reloc_info when SOM or
746 ELF_ARG_RELOC are defined.
747 (pa_stringer_aux): Don't pa_check_current_space_and_subspace here..
748 (pa_stringer): ..Do it here instead. Fix comment typos.
749 (hppa_force_relocation): Cast enums to int before comparing with
753 * config/tc-hppa.c: Removed unneeded libbfd.h to fix macro
754 redifinition warning.
755 (md_apply_fix): Added cast from enum to int for fixP->fx_r_type.
756 (hppa_force_relocation): ditto
757 (md_apply_fix): Added cast to buf to fix warnings.
759 * config/tc-hppa.h (pa_define_label, parse_cons_expression_hppa,
760 cons_fix_new_hppa, hppa_force_relocation): Prototype.
762 * config/tc-hppa.c (reloc_type): It's an enum for OBJ_ELF.
763 (R_N0SEL, R_N1SEL): Define only for OBJ_SOM.
764 (tc_gen_reloc): Make `code' a reloc_type and `codes' a
765 reloc_type** to avoid warnings in switch.
766 (md_apply_fix): Make insn, val signed. Zap buf_wd and read insn a
767 little earlier instead.
769 * config/tc-hppa.c (symbol_arg_reloc_info): Define for both som
771 (pa_type_args): Use symbol_arg_reloc_info.
772 (struct pa_it): Make arg_reloc unsigned int.
773 (struct hppa_fix_struct): Likewise for fx_arg_reloc.
774 (pa_text, pa_data, pa_comm): Don't compile for TE_LINUX.
775 (pa_code): Delete. pa_text duplicates this function.
776 (md_pseudo_table): Call obj_elf_text for ".code" if TE_LINUX.
777 (fix_new_hppa): Argument offset is offsetT, arg_reloc is unsigned
779 (cons_fix_new_hppa): Actually change selector to e_fsel when
780 warning about assuming so.
781 (tc_gen_reloc): More example elf arg reloc code.
782 (md_apply_fix): Use arg_reloc_stub_needed for elf too.
783 (hppa_force_relocation): Likewise.
785 * config/tc-hppa.h: Use TARGET_ARCH_SIZE to select target include
787 (pa_end_of_source): Prototype.
788 (hppa_fix_adjustable): Prototype.
789 (LABELS_WITHOUT_COLONS): Move it..
790 * config/te-hppa.h: To here.
792 * config/te-hppa64.h: New file.
794 * config/tc-hppa.c: Use TARGET_ARCH_SIZE to select target reloc
796 (md_apply_fix): Pass stdoutput to bfd_hppa_insn2fmt. Handle
797 format -10, -16, 16 relocs.
798 (hppa_elf_mark_end_of_function): Test for null
799 last_call_info->start_symbol
801 * config/tc-hppa.c (pa_ip): In case 'V', pass `strict' to
802 CHECK_FIELD, not INSERT_FIELD_AND_CONTINUE. Don't pass opcode to
803 re_assesmble_* functions. Delete extraneous statements. Fix
805 (md_apply_fix): Compare against 1048575 in case 21. Don't pass
806 insn to re_assemble_*.
810 * doc/internals.texi (Expressions): Fix typo.
814 * config/tc-sh.c: Fix formatting.
815 * config/tc-tic54x.c: Fix formatting.
816 * depend.c: Fix formatting.
817 * flonum-konst.c: Likewise.
818 * flonum-mult.c: Likewise.
822 * config/tc-sh.c: Fix comments.
823 * config/obj-vms.c: Fix comments.
824 * config/tc-a29k.c: Likewise.
825 * config/tc-alpha.c: Likewise.
826 * config/tc-h8300.c: Likewise.
827 * config/tc-h8500.c: Likewise.
828 * config/tc-i370.c: Likewise.
829 * config/tc-ia64.c: Likewise.
830 * config/tc-m68hc11.c: Likewise.
831 * config/tc-m68k.c: Likewise.
832 * config/tc-mips.c: Likewise.
833 * config/tc-ns32k.c: Likewise.
834 * config/tc-ppc.c: Likewise.
835 * config/tc-z8k.c: Likewise.
839 * config/obj-coff.c (TC_COFF_SECTION_DEFAULT_ATTRIBUTES): New.
840 Default to '(SEC_LOAD | SEC_DATA)'.
841 (obj_coff_section) [BFD_ASSEMBLER]: Use it.
843 * doc/internals.texi (CPU Backend): Describe
844 TC_COFF_SECTION_DEFAULT_ATTRIBUTES.
848 * as.c (parse_args): NULL terminate the long option list.
852 * config/tc-d30v.h: Include "write.h" for fixS.
853 (d30v_start_line, md_pcrel_from_section): Add function prototypes.
857 * config/tc-d30v.c (write_2_short): Further changes to warning
858 messages produced when combining EITHER_BUT_PREFER_MU attributed
867 * config/tc-arm.c (psrs): Accept combinations of flags.
871 * config/tc-avr.c: Change _ () to _() around all strings marked
872 for translation (exception from the usual coding style).
873 (avr_opt): New struct variable, how the new switches are set.
874 (OPTION_MMCU): Define as 'm' and actually use.
875 (md_longopts): Add -mall-opcodes, -mno-skip-bug, -mno-wrap.
876 (show_mcu_list): New function, display the list of known MCUs.
877 (md_show_usage): Document the new switches. Call show_mcu_list.
878 (avr_set_arch): Change 'm' to OPTION_MMCU.
879 (md_parse_option): Call show_mcu_list if unknown MCU specified.
880 Handle the new switches.
881 (avr_operands): Disable warnings for undefined combinations of
882 operands if -mall-opcodes. Disable warnings for skipping two-word
883 instructions if enhanced core or -mno-skip-bug.
884 (avr_operand): Accept all addressing modes on avr1 if -mall-opcodes.
885 (md_apply_fix3): Reject 8K wrap if >8K or -mno-wrap.
886 (md_assemble): Accept opcodes not supported by MCU if -mall-opcodes.
887 (avr_ldi_expression): Warn about implicit lo8().
888 * config/tc-avr.h (md_pcrel_from_section): Add prototype.
892 * configure.in: Add entry for mips-*-sysv4*MP*
894 * config/tc-mips.c (mips_target_format): Return elf32-tradbigmips or
895 elf32-tradlittlemips for traditional mips targets.
896 * config/tc-mips.c (md_estimate_size_before_relax): Duplicate the
897 test for Link Once sections as in adjust_reloc_syms.
898 * config/te-tmips.h: New file for traditional mips targets. Define
903 * config/obj-coff.c (obj_coff_setcion) [BFD_ASSEMBLER]: If the
904 flags argument is not present, don't change an existing section's
905 section's attributes. If the flags argument is present, warn if the
906 attributes don't match the section's current attributes. When
907 long section names are supported, set SEC_LINK_ONCE and
908 SEC_LINK_DUPLICATES_DISCARD for a new .gnu.linkonce section.
912 * config/obj-aout.c (obj_aout_type): Do not ignore for undefined
913 symbols; create them.
917 * write.c (set_segment_vma): New: Set vma and lma for a segment.
918 (write_object_file) [BFD_ASSEMBLER && OBJ_COFF && TE_GO32]: Use it.
922 * config/tc-mips.c (mips_ip): handle "(foo-.-4)" type of
923 expressions. Ignore the problem when handling 16 bit signed
924 immediates, because the assembler will take care of the relocation
929 * config/tc-d30v.c (write_2_short): Do not allow opcodes with
930 the EITHER_BUT_PREFER_MU attribute to be combined into a reverse
931 sequential order, and emit warning messages if the input source
932 code contains constructs like that, or parallel constructs
933 containing such opcodes.
937 * config/tc-avr.c (mcu_types): Rename avr4 to avr5, add avr4.
938 Add more MCU types for avr4 and avr5. Replace at94k{10,20,40}
939 with just at94k. Change AVR_ISA_85xx back to AVR_ISA_2xxx.
940 (md_show_usage): Update usage message.
941 (md_parse_option): Allow redefinition of MCU type within the
942 same avr[1-5] bfd machine type. Show both old and new MCU type
943 in the error message.
944 (md_apply_fix3): Support 8K wrap if AVR_ISA_MEGA is not set.
945 Simplify 8K wrap code.
949 * config/obj-aout.c: Remove all uses of DEFUN.
950 * config/obj-ieee.c: Likewise.
951 * config/tc-sh.c: Fix comment typos.
952 * config/tc-tahoe.c: Likewise.
953 * config/tc-vax.c: Likewise.
954 * config/tc-w65.c: Likewise.
955 * config/tc-z8k.c: Likewise.
956 * config/tc-h8300.c (build_bytes): Assemble ldmac correctly.
960 * config/tc-i386.c (md_estimate_size_before_relax): Revert
961 more changes from Sept 1999
962 (tc_i386_fix_adjustable): ditto
963 (md_apply_fix3): ditto
967 * cgen.c (expr_jmp_buf_p): New validity flag for expr_jmp_buf.
968 (gas_cgen_parse_operand): Set it around expression() call.
969 (gas_cgen_md_operand): Test for it before longjmp().
973 * config/tc-h8500.c: Remove all uses of DEFUN.
974 * config/tc-sh.c: Likewise.
975 * config/tc-w65.c: Likewise.
976 * config/tc-z8k.c: Likewise.
978 * config/tc-h8500.c: Fix typos in comments.
982 * expr.c (operand): Permit $hex literals if LITERAL_PREFIXDOLLAR_HEX
987 * expr.c (operand): Do not as_bad() if RELAX_PAREN_GROUPING is
988 defined. Fix error message for `[' grouping.
992 * config/tc-h8300.c: Fix formatting and comment typos.
996 * config/tc-ia64.c (note_register_values): Move premature QP
997 notation clearing into the appropriate place.
1001 * dep-in.sed: Escape literal `.'s on patterns. Trim off `../'
1002 first before anything else. Add bin-bugs.h, emul.h and progress.h
1003 Sort list of files as for $(OBJS) in Makefile.am.
1005 * Makefile.am (DEP): grep for leading `/' in DEPA, and fail if we
1006 find one. Remake dependencies.
1007 ($(OBJS)): Add bin-bugs.h, emul.h, and progress.h Sort the list.
1008 * Makefile.in: Regenerate.
1009 * doc/Makefile.in: Regenerate.
1011 * config/tc-i386.c (i386_displacement): Don't assume a constant
1012 displacement is necessarily 16 bits when in 16 bit code mode.
1013 (md_assemble): Instead size the displacement here after we know
1014 for sure that a .code16gcc operand hasn't automatically added
1015 operand size prefixes.
1019 * Makefile.am: Rebuild dependency.
1020 * Makefile.in: Rebuild.
1024 * config/tc-h8300.c (parse_reg): Make the function static.
1025 (parse_exp): Likewise.
1029 * config/tc-i386.c (tc_i386_fix_adjustable): Revert change from
1030 Sept 1999; RVA relocs need to be treated more like DIR32 relocs
1031 for cygwin import libraries to work properly.
1035 * Makefile.am: Rebuild dependency.
1036 * Makefile.in: Rebuild.
1037 * configure: Likewise.
1038 * doc/Makefile.in: Likewise.
1042 * doc/internals.texi (CPU backend): Add @itemx for
1043 TC_START_LABEL_WITHOUT_COLON.
1044 * doc/c-tic54x.texi: New.
1045 * doc/as.texinfo: Add tic54x features and include primary tic54x
1047 * doc/all.texi: Add C54X.
1048 * doc/Makefile.am (CPU_DOCS): Add c-tic54x.texi.
1049 * doc/Makefile.in: Regenerate.
1050 * configure.in: Add tic54x and define LIBM for tic54x.
1051 * configure: Regenrate.
1052 * config/tc-tic54x.[ch]: New.
1053 * config/obj-coff.h: Add tic54x.
1054 * Makefile.am: (CPU_TYPES): Add tic54x.
1055 (TARGET_CPU_CFILES): Add 'tc-tic54x.c'.
1056 (TARGET_CPU_HFILES): Add 'tc-tic54x.h'.
1057 (as_new_LDADD): Add $(LIBM).
1058 * Makefile.in: Regenerate.
1062 * doc/Makefile.am (CPU_DOCS): Added 68hc11 file.
1063 * doc/c-m68hc11.texi: Document 68HC11 and 68HC12 port.
1064 * doc/as.texinfo: Likewise.
1066 * configure, Makefile.in: Regenerate.
1067 * configure.in (emulations): Recognize m6811 and m6812.
1068 * Makefile.am (CPU_TYPES, TARGET_CPU_CFILES, TARGET_CPU_HFILES):
1069 Added files for 68hc11 and 68hc12 assembler.
1070 * config/tc-m68hc11.c: Assembler for 68hc11 and 68hc12.
1071 * config/tc-m68hc11.h: Header definition for that assembler.
1075 * symbols.c (resolve_symbol_value): Use bfd_octets_per_byte
1076 instead of OCTETS_PER_BYTE.
1078 * config/tc-v850.c: Fix compile time warnings.
1079 * config/tc-ppc.c: Fix compile time warnings.
1083 * configure.in: Don't emulate i386-pc-pe-coff with i386coff.
1084 * configure: Rebuild.
1088 * config/obj-coff.c (obj_coff_weak): Typo fix: Change BFD_ASSEMLER
1093 * config/tc-mips.c (md_parse_option): Accept RM5200,RM5230,
1094 RM5231, RM5261, RM5721 and RM7000 as r5000 cpu variants.
1096 * doc/c-mips.texi: Document newly accepted cpu variants.
1100 * config/tc-mips.h: Remove definition of ONLY_STANDARD_ESCAPES.
1104 * macro.c (getstring): Make it possible to escape the quote
1109 * config/tc-hppa.c (pa_export): Weak symbols can be global.
1113 * configure: Regenerate.
1117 * app.c (do_scrub_begin): Don't default lex[';'] as a line
1119 * doc/internals.texi (line_separator_chars): Semicolon is no
1120 longer a default. Mention null and newline as defaults.
1122 * read.c (is_end_of_line): Remove ifdef TC_HPPA.
1124 * config/tc-i386.h (line_separator_chars): Explicitly mention `;'
1125 * config/tc-i860.h (line_separator_chars): Likewise.
1126 * config/tc-h8300.c (line_separator_chars): Likewise.
1127 * config/tc-i960.c (line_separator_chars): Likewise.
1128 * config/tc-m68k.c (line_separator_chars): Likewise.
1129 * config/tc-mips.c (line_separator_chars): Likewise.
1130 * config/tc-ns32k.c (line_separator_chars): Likewise.
1131 * config/tc-sparc.c (line_separator_chars): Likewise.
1132 * config/tc-vax.c (line_separator_chars): Likewise.
1134 * config/tc-h8300.c (comment_chars): Use string initialiser.
1135 * config/tc-i960.c (line_comment_chars): Likewise.
1136 * config/tc-z8k.c (comment_chars, line_comment_chars,
1137 line_separator_chars): Likewise.
1139 * config/tc-arm.c (line_separator_chars): Always use `;', not just
1144 * config/tc-arm.c (cons_fix_new_arm): Assign correct reloc value
1149 * configure.in (VERSION): Update to show this is the CVS mainline.
1153 * config/tc-alpha.c (md_undefined_symbol): Properly understand that
1154 $at is the integer register $r28, vs. both $r28 and the floating
1155 point register $f28.
1159 * config/tc-ia64.c (generate_unwind_image): Call ia64_flush_insns.
1160 (dot_endp): Don't call ia64_flush_insns.
1161 (emit_one_bundle): Don't delete prologue/body records from
1162 unwind_record list in first loop. Rewrite second loop to account for
1167 * config/tc-ia64.c: Add missing prototypes.
1168 (generate_unwind_image): Cast argument to output_unw_records call.
1172 * config/tc-avr.c (avr_operand): fix the formatting of the comment.
1176 * config/tc-avr.c (AVR_ISA_???): moved to include/opcode/avr.h
1177 (REGISTER_P): likewise.
1178 (avr_opcodes): uses include/opcode/avr.h
1179 (avr_operand): enable ld r,Z or st r,Z for at90s1200.
1183 * read.c (is_end_of_line): No ';' for TC_HPPA. Add missing
1188 * read.c (is_end_of_line): Put back `;'.
1192 * config/tc-i386.c (md_shortopts): Remove 'm', add 'q' to non-elf.
1196 * expr.c (operand): Test is_end_of_line outside switch to catch
1197 line separator chars that are also operators.
1198 (operator): Return O_illegal for line separator chars.
1200 * read.c (is_end_of_line): Use 1 instead of 99. Don't set `;'
1201 entry (or `!' entry for TC_HPPA).
1203 * config/tc-arm.c (my_get_float_expression): Cast to unsigned char
1204 before indexing is_end_of_line. Remove redundant check for '\0'.
1206 * config/tc-h8500.c (md_assemble): Likewise.
1207 * config/tc-mcore.c (md_assemble): Likewise.
1208 * config/tc-tic30.c (tic30_find_parallel_insn): Likewise.
1211 * config/tc-m88k.c (s_bss): Cast to unsigned char before indexing
1213 * config/tc-mcore.c (mcore_cons): Likewise.
1214 (mcore_float_cons): Likewise.
1215 (mcore_stringer): Likewise.
1216 * config/tc-tic30.c (tic30_find_parallel_insn): Likewise.
1220 * config/tc-arm.c (do_mrs): Allow SPSR_BIT to be set correctly.
1224 * config/tc-sh.c: Fix compile time warning messages.
1226 * config/tc-mips.c: Fix compile time warning messages.
1230 * doc/as.texinfo: Update copyright dates.
1231 (Local Labels): Delete misplaced mention of ARM.
1232 * NEWS: Mention ARM ELF support.
1236 * config/tc-mn10300.c (md_assemble): Copy size to real_size before
1237 it is modified, and use the real_size to compute the frag address
1238 for dwarf2 line info.
1242 * Makefile.am (DEP, DEP1, dep, dep-in, dep-am): Use a better sed
1243 line-matching scheme to cope with automake moving variables around.
1244 ($(TARG_CPU_O)): Remove dependency on TARG_CPU_DEP_@target_cpu_type@
1245 * Makefile.in: Regenerate.
1249 * config/tc-sparc.c (sparc_relax): New.
1250 (md_longopts): Add -relax and -no-relax options.
1251 (md_parse_options, md_show_usage): Likewise.
1252 (md_apply_fix3): Optimize tail call into branch always if possible.
1256 * config/tc-d10v.c (write_2_short, parallel_ok, md_assemble,
1257 d10v_cleanup) implement Mitsubishi's newly explained branch-packing
1258 rules, with warning when a GAS statement specifies a packing that
1259 will result in an instruction being squashed.
1260 Added typdef packing_type and enumerals, changed various integer
1261 literals to use the enumerals.
1265 * config/tc-ia64.c (dot_restorereg_p): New function.
1266 (md_pseudo_table): Add restorereg.p.
1267 (output_X3_format): Fix typo: record type should be UNW_X3, not UNW_X1.
1268 (output_X4_format): Fix typo: record type should be UNW_X4, not UNW_X2.
1270 * config/tc-ia64.h (unw_record_type): Add unwabi.
1271 (unw_r_record): Rename member MASK to GRMASK. Add sub-structure
1272 called MASK with members for imask, and the masks produced by
1273 fr_mem, gr_mem, br_mem, and frgr_mem.
1274 (unw_p_record): Add members ABI and CONTEXT.
1275 (unw_x_record): Add member AB.
1276 * config/tc-ia64.c (enum reg_symbol): Add REG_PSP and REG_PRIUNAT
1277 as pseudo-register for use during unwind info generation.
1278 (AR_PFS, AR_LC): New macros.
1279 (enum pseudo_type): Add PSEUDO_FUNC_REG to permit declaring registers
1280 whose names start with an at sign (as in "@priunat").
1281 (pseudo_func): Add "svr4", "hpux", "nt" constants and "priunat"
1283 (unwind_list, unwind_tail, current_unwind_entry, proc_start,
1284 proc_end, unwind_info, personality_routine): Consolidate into
1285 "unwind" structure to reduce offset-table use. Add member
1286 NEXT_SLOT_NUMBER to track the slot number for the next instruction
1288 (output_R1_format, output_R3_format, output_P3_format,
1289 output_P6_format): Initialize R with zero to reduce compiler warnings.
1290 (output_P7_format): Ditto. Add `default' branch to switch
1291 statement to reduce compiler warnings.
1292 (output_P8_format, output_B1_format, output_B4_format): Ditto.
1293 (output_P4_format): Rename 2nd & 3rd arg to IMASK and IMASK_SIZE.
1294 (format_ab_reg): Rename from format_a_b_reg. Merge A and B args
1295 into single argument.
1296 (output_X1_format, output_X3_format): Initialize R with zero to reduce
1297 compiler warnings. Merge A and B args into single argument.
1298 (output_X2_format, output_X4_format): Remove unused variable R. Merge
1299 A and B args into single argument.
1300 (free_record): Removed (wasn't used).
1301 (free_list_records): Also free imasks in prologue records.
1302 (output_prologue, output_prologue_gr): Initialize mask bits to zero.
1303 (output_spill_mask): Remove.
1304 (output_unwabi): New function.
1305 (output_epilogue, output_label_state, output_copy_state): Call
1307 (output_spill_psprel, output_spill_sprel, output_spill_psprel_p,
1308 output_spill_sprel_p, output_spill_reg, output_spill_reg_p): Add AB
1310 (process_one_record): New locals FR_MASK and GR_MASK. Ignore
1311 gr_mem, fr_mem, br_mem, and frgr_mem records and instead emit them
1312 as part of handling the prologue records. Emit region's imask if
1313 we have one. Handle unwabi, epilogue, label_state, copy_state,
1314 spill_psprel, spill_sprel, spill_reg, spill_psprel_p,
1315 spill_sprel_p, and spill_reg_p records.
1316 (set_imask, count_bits, slot_index): New function.
1317 (fixup_unw_records): Fix region size computation. Handle
1318 epilogue, spill_reg, spill_sprel, spill_psprel, spill_reg_p,
1319 spill_sprel_p, and spill_psprel_p records. Merge mask bits of
1320 frgr_mem, fr_mem, gr_mem, br_mem on a per-region basis and
1321 set_imask accordingly. Update imask for gr_gr, and br_gr records.
1322 (convert_expr_to_ab_reg, convert_expr_to_xy_reg): New function.
1323 (dot_save): Use manifest constants for applicaton registers.
1324 Handle REG_PR and REG_PRIUNAT.
1325 (dot_restore): Don't just ignore it.
1326 (dot_restorereg): New function..
1327 (generate_unwind_image): Ensure unwind info is a multiple of eight
1328 bytes, not just four bytes.
1329 (dot_handlerdata, dot_unwentry): Demand empty rest of line.
1330 (dot_altrp): Don't just ignore it.
1331 (dot_savemem): New function. Replaces dot_savesp() and
1332 dot_savepsp(). Use manifest constants for applicaton registers.
1333 Handle REG_PR and REG_PRIUNAT.
1334 (dot_savef): Simplify.
1335 (dot_saveb): Support generation of br_gr.
1336 (dot_spillreg, dot_spillmem, dot_spillreg_p, dot_spillmem_p,
1337 dot_label_state, dot_copy_state): New function.
1338 (dot_unwabi): Don't just ignore it.
1339 (md_pseudo_table): Add restorereg, spillreg, spillsp, spillpsp,
1340 spillreg.p, spillsp.p, spillpsp, label_state, copy_state,
1341 unwabi, vframesp, and vframepsp. Fix typo alprp->altrp.
1342 (emit_one_bundle): Set slot number for prologue/body records
1343 *before* emitting the first insn.
1344 (emit_one_bundle): Set UNWIND.NEXT_SLOT_NUMBER.
1345 (md_begin): Declare "psp" pseudo-register.
1346 (md_operand): Handle PSEUDO_FUNC_REG. Fix printing of error message
1347 so we don't get segfault.
1348 (output_psp_sprel): Output sp/psp relative offsets as 4-byte word
1349 counts as required per SW Conventions manual
1350 (output_rp_psprel, output_rp_sprel, output_pfs_psprel,output_pfs_sprel,
1351 output_preds_psprel, output_preds_sprel, output_spill_base,
1352 output_unat_psprel, output_unat_sprel, output_lc_psprel,
1353 output_lc_sprel, output_fpsr_psprel, output_fpsr_sprel,
1354 output_priunat_psprel, output_priunat_sprel, output_bsp_psprel,
1355 output_bsp_sprel, output_bspstore_psprel, output_bspstore_sprel,
1356 output_rnat_psprel, output_rnat_sprel, output_spill_psprel,
1357 output_spill_sprel, output_spill_psprel_p, output_spill_sprel_p):Ditto.
1358 (dot_vframe): Implement.
1359 (dot_vframesp, dot_vframepsp): New function.
1363 * configure.in (i386-*-freebsd a.out entry): Quote properly.
1364 * configure: Regenerate.
1368 * config/tc-i386.c (md_assemble): Pass jump reloc in fr_var...
1369 (md_estimate_size_before_relax): so we can use it here instead of
1370 old kludges. Localise vars to blocks. Comment.
1372 * frags.c (frag_new): Update fr_var comments.
1373 * frags.h (struct frag): Ditto.
1377 * config/tc-ia64.c (FUNC_PC_RELATIVE): New.
1378 (pseudo_func): Add pcrel.
1379 (operand_match): Handle IA64_OPND_TGT64.
1380 (build_insn): Likewise.
1381 (md_begin): Initialize pseudo_func[FUNC_PC_RELATIVE].
1382 (ia64_gen_real_reloc_type): Handle FUNC_PC_RELATIVE.
1383 (fix_insn): Handle all three 64-bit relocation types.
1387 * obj.h (struct format_ops): New members begin, app_file,
1388 s_set_other, s_set_desc, s_get_type, s_set_type,
1389 separate_stab_sections, init_stab_section.
1391 * config/obj-multi.h: Update GPL notice to v2.
1393 (obj_app_file): New.
1394 (S_SET_SIZE): Test s_set_size for NULL before calling.
1395 (S_SET_ALIGN): Similar for s_set_align.
1400 (SEPARATE_STAB_SECTIONS): New.
1401 (INIT_STAB_SECTION): New.
1402 (EMIT_SECTION_SYMBOLS): New.
1403 (AOUT_STABS) [OBJ_MAYBE_AOUT]: Define.
1405 * config/obj-elf.h: Update GPL notice to v2.
1406 Mention that this file is included from obj-multi.h.
1407 (obj_begin): Wrap definition in ifndef.
1408 (elf_file_symbol): Constify declaration.
1409 (obj_app_file): Ditto.
1410 (SEPARATE_STAB_SECTIONS, INIT_STAB_SECTION, OBJ_PROCESS_STAB):
1411 Wrap in ifndef SEPARATE_STAB_SECTIONS.
1413 * config/obj-elf.c (elf_s_set_other): New.
1414 (elf_file_symbol): Constify argument.
1415 (elf_separate_stab_sections): New.
1416 (elf_init_stab_section): New.
1417 (elf_format_ops): Add new members. Remove comma at end.
1419 * config/obj-ecoff.c (ecoff_separate_stab_sections): New.
1420 (ecoff_format_ops): Add new fields. Remove comma at end.
1421 Mention inconsistency for emit_section_symbols.
1423 * config/obj-coff.h (c_dot_file_symbol): Constify declaration.
1425 * config/obj-coff.c (c_dot_file_symbol): Constify argument.
1426 (coff_separate_stab_sections): New.
1427 (coff_format_ops): Add new members.
1429 * config/obj-aout.c (obj_aout_sec_sym_ok_for_reloc): New.
1430 (obj_aout_s_set_other): New.
1431 (obj_aout_s_set_desc): New.
1432 (obj_aout_s_get_type): New.
1433 (obj_aout_s_set_type): New.
1434 (obj_aout_separate_stab_sections): New.
1435 (aout_format_ops): New members added. Use obj_aout_process_stab,
1436 not 0. Use obj_aout_sec_sym_ok_for_reloc, not 0.
1437 (obj_aout_frob_symbol): Add ATTRIBUTE_UNUSED to args as
1439 (obj_aout_line, obj_aout_weak, obj_aout_type): Ditto.
1443 * config/tc-i386.c (tc_i386_fix_adjustable): Prevent adjustment
1444 for OBJ_MAYBE_ELF too. Use S_IS_EXTERNAL instead of S_IS_EXTERN.
1445 (md_estimate_size_before_relax): Ensure jumps to weak and
1446 externally visible symbols are relocatable.
1450 * stabs.c (aout_process_stab): Make global.
1451 (s_desc): Add ATTRIBUTE_UNUSED to args as appropriate.
1452 * read.h (aout_process_stab): Declare.
1454 * configure.in (EMULATIONS) [i386aout, i386coff, i386elf]:
1455 Generalize to *aout, *coff *elf.
1456 * configure: Regenerated.
1458 * doc/internals.texi (Object format backend): Say
1459 SEPARATE_STAB_SECTIONS needs to be nonzero, not just defined.
1461 * Makefile.am (TARG_ENV_HFILES): Delete te-multi.h.
1462 * Makefile.in: Regenerated.
1466 * cgen.h (GAS_CGEN_MAX_FIXUPS): Check if already defined.
1470 * config/tc-hppa.c (md_apply_fix): Mask out immediate bits of
1471 instruction to reflect change in re_assemble_*.
1475 * configure.in (hppa-*-hpux11*): If the cpu is hppa*64*, then
1476 build PA64 ELF tools.
1477 * configure: Rebuilt.
1481 * Makefile.am: Regenerate dependencies.
1482 * Makefile.in: Regenerate.
1486 * config/tc-arm.c (struct asm_psr): Add boolean field
1487 distinguishing between CSPR and SPSR. Rename 'number' field
1489 (psrs): Rearrange contents to match new asm_psr structure.
1490 (arm_psr_parse): Move next to psr_required_here. Make it
1491 return an asm_psr structure.
1492 (psr_required_here): Use asm_psr structure returned by
1494 (do_msr): Reorganise to allow psr_required_here to be called
1496 (md_undefined_name): Mark 'name' parameter as unused, since
1497 the COFF target does not use it.
1501 * config/te-386bsd.h: Clean up comments to adhere to the GNU coding
1503 * config/te-aux.h: Likewise.
1504 * config/te-dpx2.h: Likewise.
1505 * config/te-go32.h: Likewise.
1506 * config/te-hp300.h: Likewise.
1507 * config/te-hppa.h: Likewise.
1508 * config/te-i386aix.h: Likewise.
1509 * config/te-ic960.h: Likewise.
1510 * config/te-interix.h: Likewise.
1511 * config/te-nbsd532.h: Likewise.
1512 * config/te-pc532mach.h: Likewise.
1513 * config/te-ppcnw.h: Likewise.
1514 * config/te-psos.h: Likewise.
1515 * config/te-sparcaout.h: Likewise.
1516 * config/te-sun3.h: Likewise.
1517 * config/te-sysv32.h: Likewise.
1521 * config/tc-h8300.c (do_a_fix_imm): Don't rely on `short' being 16
1522 bits. Instead explicitly mask and sign extend. Do the 8 bit mask
1523 and sign extend without an if statement.
1524 (build_bytes): Likewise.
1528 * config/tc-h8300.c (do_a_fix_imm): Output a reloc for no
1529 X_add_symbol L_32 case.
1533 * config/te-freebsd.h: New file.
1537 * asintl.h (gettext, dgettext, dcgettext, textdomain,
1538 bindtextdomain): Replace defines with those from intl/libgettext.h
1539 to quieten gcc warnings.
1541 * NEWS: Mention x86 .arch and -q.
1543 * config/tc-i386.c (quiet_warnings): New.
1544 (md_assemble): Use quiet_warnings.
1545 (md_parse_option): Set quiet_warnings from -q.
1546 (md_show_usage): Mention -q, delete -m.
1547 (flag_do_long_jump): Delete.
1548 (md_parse_option): Remove -m.
1549 (md_show_usage): Remove -m.
1550 (md_create_long_jump): Remove useless flag_do_long_jump code.
1552 * as.c (parse_args): In case OPTION_DEFSYM, use a valueT to hold
1553 the symbol value, and use bfd_scan_vma if BFD_ASSEMBLER.
1558 * doc/c-i386.texi (i386-Arch): New section.
1559 (i386-Syntax): Mention .intel_syntax and .att_syntax.
1561 * config/tc-i386.c (cpu_arch_name, cpu_arch_flags): New.
1562 (smallest_imm_type): Use smallest opcode for shift by one if cpu
1563 architecture has been given and is not 486.
1564 (set_cpu_arch): New.
1565 (md_pseudo_table): Add .arch.
1566 (md_assemble): Warn if cpu architecture has been given and an
1567 unsupported instruction.
1569 * config/tc-i386.h (SMALLEST_DISP_TYPE): Delete.
1570 Move operand_types bit defines after relevant template field.
1571 (template): Add cpu_flags.
1577 * config/tc-mn10300.h (md_end): Define.
1578 (mn10300_finalize): Declare.
1579 * config/tc-mn10300.c: Include dwarf2dbg.h.
1580 (debug_line): Define.
1581 (md_assemble): Generate dwarf2 line info.
1582 (mn10300_finalize): New function. Finalize dwarf2 info.
1586 * config/tc-mips.c (md_estimate_size_before_relax): Use the
1587 external version of the relocation for weak symbols.
1591 * config/tc-ia64.c (output_P7_format, case mem_stack_f): Output fixed
1592 frame size in units of 16 bytes, as required per SW Conventions manual.
1593 (output_unw_records): Output info-block header as a dword to get
1598 * as.h: #include "file", not <file> on files from ../include.
1599 (as_abort, as_fatal): Add ATTRIBUTE_NORETURN.
1600 * config/tc-m68k.c (m68k_ip): Fix signed/unsigned warnings.
1601 (md_convert_frag): Add ATTRIBUTE_UNUSED.
1602 (tc_coff_symbol_emit_hook): Ditto.
1603 (OPTCOUNT): Cast to int to avoid compiler warning.
1604 (md_begin): Fix signed/unsigned warnings.
1608 * config/tc-m68k.c (md_convert_frag_1): Abort if we end up in the
1609 ABRANCH LONG case for a conditional branch on a 68000.
1610 (md_estimate_size_before_relax): Likewise. Also handle
1611 flag_short_refs correctly for ABRANCH, BCC68000, and DBCC.
1612 (m68k-ip: case ABSL): Relax absolute references to 16-bit
1613 PC-relative on all CPUs.
1614 (md_estimate_size_before_relax): Likewise.
1618 * as.c (parse_args): Just mention current year in printed
1623 * config/tc-ppc.c (pre_defined_registers): Add entries for vector
1625 (md_parse_option): Recognize -m7400.
1629 * config/atof-ieee.c (gen_to_words): When adding carry back in,
1630 don't permit lp to become less than the words array.
1634 config/tc-mcore.c (md_apply_fix3): BFD_RELOC_MCORE_PCREL_IMM11BY2
1635 Fix little-endian case.
1639 * as.c (parse_args): Update copyright.
1643 * gas/config/tc-i386.h (SUB_SEGMENT_ALIGN): If TE_GO32, return 4
1644 for the .bss section too.
1648 * configure.in: Set em=linux for hppa-*-linux.
1649 * configure: Regenerate.
1650 * doc/Makefile.in: Regenerate with correct automake.
1652 * frags.c (frag_grow): Sanity check chunk_size.
1654 * config/obj-elf.h: #include "bfd.h" not <bfd.h>
1655 * config/obj-som.h: Likewise.
1656 * config/obj-ieee.h: Likewise.
1658 * config/tc-hppa.h: Test BFD_ARCH_SIZE, not BFD64.
1660 * config/tc-hppa.c (log2): Only compile when OBJ_SOM.
1661 (md_pseudo_table): Fully initialise OBJ_ELF cases.
1662 (fix_new_hppa): Add ATTRIBUTE_UNUSED to args as appropriate.
1663 (pa_ip): low_sign_unext now returns via function value. Use
1664 re_assemble_* instead of dis_assemble_* and
1665 INSERT_FIELD_AND_CONTINUE combination. Don't call sign_unext
1667 (md_convert_frag): Add ATTRIBUTE_UNUSED to args as appropriate.
1668 (md_section_align, md_parse_option, md_show_usage,
1669 md_undefined_symbol, pa_align, pa_block, pa_brtab, pa_try,
1670 pa_callinfo, pa_code, pa_comm, pa_end, pa_enter, pa_entry,
1671 pa_exit, pa_export, pa_import, pa_label, pa_leave, pa_level,
1672 pa_origin, pa_param, pa_proc, pa_procend, pa_space, pa_spnum,
1673 pa_version, pa_compiler, pa_copyright, pa_data, pa_fill, pa_lsym,
1675 (md_apply_fix): Change type of new_val to offsetT. Delete w1, w2,
1676 w, resulti. Add insn, val. Move bfd_get_32 and bfd_put_32
1677 outside of switch. Correct mask and shifting errors in case 10
1678 and case -11. In case 21, compare against signed range to suit
1679 hppa_field_adjust changes. In case 12, use re_assemble_12. In
1680 case 17 and case 22, use offsetT variable to properly check range.
1681 Use re_assemble_* here too.
1682 (evaluate_absolute): Change type of value to offsetT. Call
1683 hppa_field_adjust to do the work for us.
1684 (pa_parse_cmpb_64_cmpltr): Delete save_s.
1685 (pa_parse_cmpib_64_cmpltr): Ditto.
1686 (pa_build_unwind_subspace): Delete unused var subseg. Change type
1687 of i to unsigned int.
1688 (pa_type_args): Conditionally declare symbol if OBJ_SOM.
1689 (pa_end_of_source): Return type is void.
1693 * macro.c (macro_expand_body): Don't prepend macro number with zeroes.
1697 * config/tc-avr.c: ATTRIBUTE_UNUSED added to the necessary places.
1698 More comments added.
1699 (md_begin): Removed "construct symbols for each register name".
1700 Because register names conflicts with GCC generated function
1702 (avr_operand): Now constant numbers can be used as a register
1703 identifiers (0 as r0, 31 as r31).
1704 (md_assemble): use skip_space () before parsing instruction
1709 * configure.in: Set bfd_gas=yes on i386-*-pe and i386-*-nt* to
1710 ensure all pe targets use bfd. Remove unnecessary bfd_gas=yes on
1711 arm-*-netbsd* and arm-*-wince as this is set for all arm*.
1712 * configure: Regenerate.
1716 * as.h: Correctly check GCC version.
1720 * doc/as.1: Fix unbalanced brackets.
1722 * config/tc-i386.c (comment_chars): Don't use '/' as comment start if
1724 (line_comment_chars): Set to '/' if TE_FreeBSD.
1728 * configure.in: Configury support for PA64 (currently disabled).
1729 * configure: Rebuilt.
1733 * config/tc-mips.c (s_change_sec): Use record_alignment, not
1734 bfd_set_section_alignment.
1738 * config/tc-i386.c (offset_in_range): Ensure shift counts are less
1743 * doc/c-arm.texi (ARM Directives): Document behaviour of .align 0.
1744 * doc/as.texinfo (Align): Include arm and strongarm in list of
1745 targets that have the second form of the behaviour of the .align
1750 * obj-som.c: Terminate obj_pseudo_table.
1754 * as.c (parse_args): Allow md_parse_option to override -a listing
1756 * config/obj-coff.c (add_lineno): Change type of offset parameter
1757 from "int" to "bfd_vma."
1758 * config/tc-ppc.c (md_pseudo_table): Add "llong" and "machine."
1759 (ppc_mach, ppc_subseg_align, ppc_target_format): New.
1760 (ppc_change_csect): Align correctly for XCOFF64.
1761 (ppc_machine): New function, which discards "ppc_machine" line.
1762 (ppc_tc): Cons for 8 when code is 64 bit.
1763 (md_apply_fix3): Don't check operand->insert. Handle 64 bit
1765 (md_parse_option): Handle -a64 and -a32.
1767 * config/tc-ppc.h (TARGET_MACH): Define.
1768 (TARGET_FORMAT): Move to function.
1769 (SUB_SEGMENT_ALIGN): Use ppc_subseg_align.
1773 * config/tc-avr.c: New AVR_ISA_ defined.
1774 (md_assemble): Handle opcodes with optional operands (lpm,elpm).
1775 (avr_operand): Handle 'a', 'v' and 'z' constraint letters needed
1776 for `fmul', `movw' and `lpm R,Z' instructions.
1777 (avr_operands): Warn if current opcode is a two-word instruction
1778 and previous opcode was cpse/sbic/sbis/sbrc/sbrs.
1779 (avr_opcodes): New commands added.
1780 (REGISTER_P): Check 'a' and 'v' constraint letters.
1781 (mcu_types): New MCU added.
1785 * config/tc-ia64.c (pseudo_func[]): Add new "nat" entry equivalent
1787 (operand_match): Conditionally insert default bit values for IMMU9.
1791 * configure.in: Add NetBSD/sparc ELF and NetBSD/sparc64 support.
1792 * configure: Rebuilt.
1797 * config/tc-hppa.c (md_apply_fix): Handle new PA2.0 formats.
1799 * config/tc-hppa.c (CHECK_ALIGN): New macro.
1800 Added handling of new operand types l,y,&,fe,fE,fx.
1808 * Makefile.am (CPU_TYPES): Add ia64.
1809 (TARGET_CPU_CFILES): Add config/tc-ia64.c.
1810 (TARGET_CPU_HFILES): Add config/tc-ia64.h.
1811 * Makefile.in: Rebuild.
1812 * app.c (do_scrub_chars): Handle DOUBLESLASH_COMMENTS.
1813 * configure: Rebuild.
1814 * configure.in: Recognize ia64 as cpu type. Set bfd_gas.
1815 (ia64-*-elf*, ia64-*-linux-gnu*): New targets.
1816 * expr.c (expr): Handle md_optimize_expr.
1817 * read.c (LEX_HASH): Add comment.
1818 * config/tc-ia64.c, config/tc-ia64.h: New files.
1822 * config/tc-d30v.c (write_2_short): Disregard opcode1->ecc when
1823 bundling a non-delayed branch type instruction.
1827 * config/tc-mn10300.c (HAVE_AM30): Define.
1828 (md_assemble): Use it.
1832 * config/obj-elf.c (obj_elf_change_section): Check for changed
1835 * Makefile.am: (CPU_MULTI_VALID): Remove.
1836 (MULTI_CPU_TYPES): Define.
1837 (MULTI_CPU_OBJ_VALID): Define.
1838 (DEPTC): Use the above.
1839 (DEPOBJ): Same here.
1841 Regenerate dependencies.
1842 * Makefile.in: Regenerate.
1846 * Makefile.am (YACC, LEX): Get them from configure.
1850 * config/tc-i386.c (offset_in_range): Use addressT instead of
1851 bfd_vma for non-bfd assemblers.
1855 * config/tc-i386.c (offset_in_range): Sign extend val so BFD64
1856 doesn't give spurious errors.
1860 * as.h (SEEK_SET): Define if undefined.
1864 * config/tc-arm.c (md_apply_fix3): Don't use UL suffix on
1865 constants, and don't assume offsetT is 32 bits.
1869 * config/tc-d10v.h: Include "write.h" to get definition of fixS.
1870 (md_pcrel_from_section): Add prototype.
1871 (d10v_fix_adjustable): Add prototype.
1872 (d10v_force_relocation): Replace 'struct fix' with 'fixS'.
1874 * config/tc-d10v.c (md_apply_fix3): Add paren around &&.
1878 * config/tc-d10v.c (find_opcode): Correctly calculate position of
1879 symbol in frag chain.
1883 * config/tc-i386.c (fits_in_signed_byte): Change arg to offsetT.
1884 (fits_in_unsigned_byte, fits_in_unsigned_word): Ditto.
1885 (fits_in_signed_word, smallest_imm_type): Ditto.
1886 (md_assemble): Use an offsetT var to hold offsetT values, not a
1888 (offset_in_range): New.
1889 (md_assemble): Use it.
1890 (md_convert_frag): Change type of target_address, opcode_address,
1891 and displacement_from_opcode_start to offsetT.
1892 (md_create_short_jump): Change type of offset to offsetT.
1893 (md_create_long_jump): Ditto.
1894 (md_apply_fix3): Use -4, not 0xfffffffc for BFD_RELOC_386_PLT32.
1895 (md_chars_to_number): Remove.
1896 (output_invalid): Remove duplicate prototype.
1900 * Makefile.am (CPU_TYPES): Add 'avr'.
1901 (TARGET_CPU_CFILES): Add 'tc-avr.c'.
1902 (TARGET_CPU_HFILES): Add 'tc-avr.h'.
1904 * Makefile.in: Regenerate.
1906 * doc/as.texinfo: Add M32R documentation.
1910 * configure.in (WARN_CFLAGS): Set to -W -Wall by default. Add
1911 --enable-build-warnings option.
1912 * Makefile.am (AM_CFLAGS, WARN_CFLAGS): Add definitions.
1913 * Makefile.in, configure: Re-generate.
1917 * config/tc-arm.c (md_apply_fix3): Treat BFD_RELOC_ARM_PCREL_BLX
1918 in the same way as BFD_RELOC_ARM_PCREL_BRANCH, and
1919 BFD_RELOC_THUMB_PCREL_BLX lie BFD_RELOC_THUMB_PCREL_BRANCH.
1920 (tc_gen_reloc): Accept BFD_RELOC_{ARM|THUMB}_PCREL_BLX.
1921 (arm_force_relocation): Force relocations for
1922 BFD_RELOC_{ARM|THUMB}_PCREL_BLX as well.
1926 * config/tc-sh.c (get_operands): There's no third operand if the
1927 first operand is an immediate.
1931 * config/tc-sh.c (immediate): Delete.
1932 (sh_operand_info): Add immediate member.
1933 (parse_reg): Use A_PC for pc.
1934 (parse_exp): Add second argument 'op'. All callers changed.
1935 (parse_at): Expect pc to be coded as A_PC.
1936 Use immediate field in *op.
1937 (insert): Add fourth argument 'op'. All callers changed.
1938 (build_relax): Add second argument 'op'. All callers changed.
1939 (insert_loop_bounds): New function.
1940 (build_Mytes): Remove DISP_4.
1941 Split IMM_[48]{,BY[24]} into IMM[01]_[48]{,BY[24]}. Add REPEAT.
1942 (assemble_ppi): Use immediate field in *operand.
1943 (sh_force_relocation): Handle BFD_RELOC_SH_LOOP_{START,END}.
1944 (md_apply_fix): Likewise.
1945 (tc_gen_reloc): Likewise. Check for a pcrel BFD_RELOC_SH_LABEL.
1949 * config/tc-sparc.c (sparc_ip): Avoid string pasting.
1953 * internals.texi (CPU backend): Document
1954 TC_CHECK_ADJUSTED_BROKEN_DOT_WORD.
1958 * po/gas.pot: Regenerate.
1960 * as.c (show_usage): Restore translated part of bug string.
1961 * gasp.c (show_usage): Likewise.
1963 * Makefile.am (MKDEP): Use gcc -MM rather than ../mkdep.
1964 (DEP): Quote when passing vars to sub-make. Use "mv -f" rather
1965 than move-if-change.
1966 (DEP1): Modify for "gcc -MM".
1970 (CLEANFILES): Add DEPTCA, DEPOBJA, DEP2a, DEPA.
1971 Update dependencies.
1972 * Makefile.in: Regenerate.
1976 * config/tc-mn10300.c (md_pseudo_table): Use constant names.
1977 (md_begin): Likewise.
1978 (HAVE_AM33): New macro.
1979 (md_assemble): Use it. Match r_regs and xr_regs only if
1984 * as.h: #include "bin-bugs.h"
1985 * as.c (show_usage): Use REPORT_BUGS_TO.
1986 * gasp.c: #include "bin-bugs.h"
1987 (show_usage): Use REPORT_BUGS_TO.
1989 * config/tc-sparc.c (md_show_usage): Add a trailing newline.
1993 * write.c (write_object_file) [! WORKING_DOT_WORD]: If defined,
1994 call TC_CHECK_ADJUSTED_BROKEN_DOT_WORD for each word after
1999 * config/tc-i386.c (i386_immediate): Don't assume a constant
2000 immediate is necessarily 16 bits when in 16 bit code mode.
2001 (md_assemble): Instead set guess_suffix here after we have checked
2006 * config/tc-d30v.c (check_range): Allow signed or unsigned 32-bit
2007 quantities. Correct right shift sign extension.
2008 (build_insn): Make `number' unsigned long. Mask top 6 bits of
2009 32-bit value when shifting into place.
2013 * app.c: Add ATTRIBUTE_UNUSED as needed.
2014 * config/tc-ppc.c: Likewise.
2015 (ppc_size): Make unsigned long.
2016 (ppc_insert_operand): Add casts to avoid warnings.
2020 * config/tc-d10v.h (md_flush_pending_output): Define.
2024 * config/tc-sh.h (SEG_NAME): New macro: return the name of a
2025 segment. Works for both BFD_ASSEMBLER and others.
2026 (SUB_SEGMENT_ALIGN): Use SEG_NAME.
2030 * config/tc-arm.c (tinsns): Add "bal" instruction pattern.
2034 * listing.c (LISTING_LHS_WIDTH): Default depends on
2036 (LISTING_LHS_WIDTH_SECOND): Default to LISTING_LHS_WIDTH.
2040 * config/tc-sh.c (md_show_usage): Use backslash before newline in
2045 * config/tc-avr.h (TC_HANDLES_FX_DONE): Define.
2047 * config/tc-avr.c (mcu_types): Add missing initialiser.
2048 (md_pcrel_from_section): Add prototype.
2049 (avr_operand): Remove redundant test of unsigned < 0.
2050 (avr_cons_fix_new): Ensure exp_mod_pm zero on function exit.
2054 * config/tc-avr.c: New file for AVR support.
2055 * config/tc-avr.h: Likewise.
2056 * configure.in: Add AVR support.
2057 * configure: Regenerate.
2061 * gasp.c (macro_op): Add new argument to check_macro call.
2062 Macro structure definitions moved to macro.h
2063 * sb.h: Add argument to prototype for input_scrub_include_sb.
2064 * input-scrub.c (input_scrub_include_sb): Allow disabling of sb
2065 nesting checks with an additional flag.
2066 (struct input_save): Add flag to indicate whether current sb
2067 should be checked for proper macro/conditional nesting.
2068 (input_scrub_push/pop): Save/restore nest check flag.
2069 (input_scrub_next_buffer): Ditto. Also call end of macro hook if
2071 * macro.c (check_macro): Allow caller to retrieve parsed macro
2072 information if a pointer is provided. This information may be
2073 used by the new macro hooks.
2074 * macro.h: Update prototype for check_macro. Macro struct
2075 definitions moved here from macro.c/gasp.c.
2076 * read.c (read_a_source_file): Add parameter to check_macro call,
2077 and pass macro info to the macro hook, if defined.
2078 (input_scrub_insert_line): New. Allow insertion of a line of
2079 characters into the input stream.
2080 (input_scrub_insert_file): New. Allow insertion of an arbitrary
2081 file into the input stream.
2082 (s_include): Use input_scrub_insert_file.
2083 * internals.texi: Document new macro hooks.
2084 * as.h: New prototypes added.
2088 * config/tc-i386.c: Don't start any as_bad or as_warn message with
2089 an initial capital letter.
2090 (i386_index_check): Reindent.
2094 * config/tc-arm.c (md_apply_fix3): Fix bug detecting overflow of pc
2099 * config/tc-arm.c (do_t_adr): Flag "adr Rd,label"
2100 instruction operand bad if Rd > 7 when generating
2101 thumb instructions. Prevents for example,
2102 "adr r12,label" from silently failing and generating
2103 the wrong instruction.
2107 * config/tc-arm.c (md_apply_fix3): Handle same-section relocations
2108 that have a destingation >= 0x400000.
2109 Fix compile time warning messages.
2113 * config/tc-sh.c (md_begin): When encountering insn that are
2114 not supported by the current arch, only change the name if
2115 its contents are the same as prev_name.
2116 (get_specific): If the the architecture doesn't match, fail.
2120 * config/tc-sh.c (IDENT_CHAR): Define.
2121 (parse_reg): Use it instead of isalnum. Put r[0..7]_bank operand
2122 matching back where it came from.
2126 * config/tc-sh.c (md_show_usage): Add description of -dsp.
2130 * config/tc-sh.c (parse_reg): Match r[0..7]_bank operands before
2135 * config/tc-h8300.c: Add ATTRIBUTE_UNUSED as appropriate.
2139 * expr.c (operand) [case 'f']: When testing if '0f' can start a
2140 floating-point-number, make sure 'f' is in FLT_CHARS.
2144 * read.c (TC_IMPLICIT_LCOMM_ALIGNMENT): New default-definition.
2145 (s_lcomm_internal): Use it.
2146 * doc/internals.texi (CPU backend): Document it.
2147 * config/obj-evax.h (TC_IMPLICIT_LCOMM_ALIGNMENT): Set to 2**3
2152 * config/tc-mips.c (mips_ip): Don't put stuff in .rodata
2155 * config/tc-mips.c (SWITCH_TABLE): The ELF embedded-pic
2156 implementation doesn't have special handling for switch
2158 (macro_build): Allow for code in sections other than .text.
2160 (mips_ip): Likewise.
2161 (md_apply_fix): Do pc-relative relocation madness for MIPS ELF.
2162 Don't perform relocs if we will be outputting them.
2163 (tc_gen_reloc): For ELF, just use fx_addnumber for pc-relative
2164 relocations. Allow BFD_RELOC_16_PCREL_S2 relocs when
2169 * config/tc-m32r.c (m32r_fix_adjustable): Look up the
2170 relocation type based on the entry in the fixup structure.
2171 Put S_IS_EXTERN processing back in.
2175 * Makefile.am (install-exec-tooldir): Depend on
2176 install-exec-bindir for parallel make.
2177 * Makefile.in: Regenerated.
2181 * config/tc-m32r.c (struct md_longopts): Add -m32r command line
2183 (md_parse_option): Parse -m32r command line switch - disable m32rx
2185 (md_show_usage): Document new option.
2187 * doc/Makefile.am (CPU_DOCS): Add c-m32r.texi.
2188 * doc/Makefile.in: Regenerate.
2189 * doc/c-m32r.texi (M32R-Opts): Document new command line switch.
2193 * config/tc-d30v.c (check_range): Remove code that incorrectly
2194 sign extended values where bits < 32.
2198 * configure.in: Support --enable-targets=all on ia32.
2199 * configure: Regenerated.
2203 * gasp.c (do_align): Remove bogus check of alignment value.
2207 * config/obj-elf.c (elf_frob_symbol): Remove code which when
2208 TC_PPC was defined forced the type of a symbol with no other type
2213 * doc/internals.texi (CPU backend): Mention that
2214 line_separator_chars do not break up comments. Fix typos for
2215 LEX_AT and LEX_NAME descriptions. Document operands for
2216 TC_EQUAL_IN_INSN, md_operand and md_section_align. Correct
2217 description of md_create_short_jump usage. Document argument for
2218 md_undefined_symbol.
2222 * config/tc-sparc.c (OPTION_UNDECLARED_REGS): New option.
2223 (md_parse_option): Handle it.
2224 (md_show_usage): Document it.
2228 * config/tc-alpha.c (md_assemble): Accept `1' and `9' in an
2229 opcode, for the instruction `pal19'. From Andrea Arcangeli
2234 * config/tc-i386.c (i386_immediate): Move constant operand sizing
2236 (md_assemble): To here, before template operands are matched.
2237 Also ensure a constant immediate is sign extended when we know the
2238 size is at most 16 bits. This is to catch cases like "add
2239 $0xffc0,%ax" where we don't know the size, and thus that the
2240 immediate can be represented as Imm8S until after parsing the
2242 (i386_displacement): Similarly sign extend 16 bit constant
2244 (md_assemble): Relax 16-bit jump constant range check to suit sign
2245 extended displacements.
2249 * doc/c-mips.texi (MIPS Opts): Fix typo in last patch.
2253 * config/tc-i386.c (md_assemble): Don't swap intersegment jmp and
2254 call operands when intel_syntax.
2255 (intel_float_operand): Return 2 for "fi...".
2256 (i386_operand_modifier): Change "DWORD PTR" test to suit above.
2257 Return SHORT_MNEM_SUFFIX for "WORD PTR" when "fi...". Revert
2258 earlier "SHORT" change.
2259 (md_assemble): When determining suffix from Regs, exclude
2264 * configure: Add arm-wince, mips-pe and sh-pe targets.
2265 * configure: Regenerate.
2267 * config/obj-coff.h (COFF_WITH_PE): Define for mips-pe and
2269 (TARGET_FORMAT): Set to "pe-shl" for the sh-pe target and to
2270 "pe-mips" for the mips-pe target.
2272 * config/tc-arm.c (insns): Change displacement encoded in BL
2273 and B instructions if the target port is arm-wince.
2274 (do_ldst): Do not bias the relocation offset if the target
2276 (md_pcrel_from): Add in missing relocation offset bias if the
2277 target os arm-wince.
2279 * config/tc-mips.c (mips_target_format): Support COFF flavour.
2280 (md_begin): Disable -G support for mips-pe target.
2281 (md_apply_fix): Treat BFD_RELOC_RVA reloc as BFD_RELOC_32.
2282 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Add support for
2285 * config/tc-sh.c (md_begin): sh-pe target is little endian.
2286 * config/tc-sh.h (SUB_SEGMENT_ALIGN): If using a BFD
2287 assembler, just set the alignment to 4.
2289 * config/te-wince-pe.h: New file for WinCE targets. Define
2294 * config/tc-i386.c (md_assemble): Swap segments too for intel mode
2295 string instructions.
2296 (i386_operand_modifier): Set i.suffix = WORD_MNEM_SUFFIX for SHORT.
2297 (i386_intel_memory_operand): After finding a segment override,
2298 check again for no `[' before looking for a displacement. Bomb if
2299 more than one displacement rather than silently discarding the
2300 second and subsequent ones. Free strings malloc'd by
2301 build_displacement_string.
2305 * config/obj-som.c (obj_pseudo_table): Add "weak".
2306 (obj_som_weak): New routine.
2310 * config/tc-i386.c (union i386_op): New.
2311 (struct _i386_insn): Delete disps[], imms[], regs[]. Add op[].
2312 Throughout file replace occurences of disps[n], imms[n], regs[n]
2313 with equivalent op[n].disps, op[n].imms, op[n].regs. Simplify
2314 intel mode operand swapping. Add assert in regKludge and
2315 fake_zero_displacement code. Test i.types[n] when outputting
2316 displacements and immediates. Combine output of Disp16 with
2318 (md_assemble): Don't try to fix broken UNIXWARE_COMPAT opcodes
2319 when in intel mode by (not) reversing fsub and fdiv operands
2320 before the template search. This fails for single operand
2321 shorthand forms of the instruction, and if UNIXWARE_COMPAT is
2322 undefined. Instead fix the base_opcode after we've found the
2323 template. Move base_opcode xor with found_reverse_match from
2324 opcode output code to before this fix so we test for the correct
2326 (md_assemble): Don't use strcmp when deciding to ignore the suffix
2327 check in intel mode. Instead compare opcodes.
2329 * config/tc-i386.h (TC_RELOC): Delete.
2330 * config/tc-i386.c (TC_RELOC): Delete. Replace usage of TC_RELOC
2331 with equivalent call to reloc.
2333 * as.h (flag_m68k_mri): Move declaration after target include, and
2334 only declare when TC_M68K defined. Define as zero otherwise.
2335 (LABELS_WITHOUT_COLONS, NO_PSEUDO_DOT): If undefined, define as 0.
2336 * app.c (scrub_m68k_mri): Declare only when TC_M68K defined.
2337 Define as zero otherwise.
2338 (do_scrub_begin): Use m68k_mri parameter only when TC_M68K defined.
2339 (struct app_save): Declare scrub_m68k_mri only when TC_M68K.
2340 (app_push, app_pop): Save scrub_m68k_mri only when TC_M68K.
2341 (do_scrub_chars): Use LABELS_WITHOUT_COLONS directly rather than
2342 testing whether defined.
2343 * cond.c (ignore_input): Use NO_PSEUDO_DOT directly.
2344 * expr.c (operand): #ifdef unused case labels when TC_M68K undefined.
2345 * read.c: Use LABELS_WITHOUT_COLONS and NO_PSEUDO_DOT directly
2346 rather than testing whether defined.
2347 (s_mri): Set flag_m68k_mri only when TC_M68K defined.
2348 (parse_mri_cons): Declare and use only when TC_M68K.
2349 * config/tc-hppa.h (LABELS_WITHOUT_COLONS): Define as 1.
2350 * config/tc-m68k.h (NO_PSEUDO_DOT): Define as 1.
2351 * config/tc-m88k.h (NO_PSEUDO_DOT): Define as 1.
2353 * NEWS: Mention IBM 370 support.
2357 * config/tc-i386.c (md_assemble): When swapping operands for
2358 intel_syntax, assume everything that's not Imm or Disp is a
2363 * config/tc-i370.c, config/tc-i370.h: New files.
2364 * Makefile.am: Add support for Linux/IBM 370.
2365 * configure.in: Likewise.
2366 * app.c (do_scrub_begin): Don't lex single quote when TC_I370.
2367 * config/obj-elf.c: Include elf/i370.h
2368 (obj_elf_section): Don't do anything special for flag_mri if TC_I370.
2370 * Makefile.in: Regenerate.
2371 * configure: Regenerate.
2373 * doc/c-i370.texi: New file.
2374 * doc/all.texi: Include it.
2375 * doc/as.texinfo: And here.
2376 * doc/Makefile.am(CPU_DOCS): Add c-i370.texi.
2377 * doc/Makefile.in: Regenerate.
2381 * config/tc-d30v.c (parallel_ok): Use FLAG_NOT_WITH_ADDSUBppp to
2382 determine if an instruction can be used in parallel with an ADDppp
2383 or SUBppp instruction.
2387 * doc/c-mips.texi (MIPS Opts): Document -mgp32 and -mgp64.
2391 * config/tc-mips.c (mips_gp32): New variable.
2392 (macro_build) Use mips_gp32.
2394 (md_longopts): Add "-mgp32" and "-mgp64".
2395 (md_parse_option): Add OPTION_GP32 and OPTION_GP64.
2397 (OPTION_M7900): Change offset
2398 (OPTION_NO_M7900): Ditto.
2402 * config/obj-coff.c (add_lineno): Accept non-positive lineno with
2403 warning, and bump it to 1.
2408 * dwarf2dbg.c (print_stats): Add cast to force printf argument to
2413 * config/tc-mips.c (MF_HILO_INSN): Define.
2414 (mips_7000_hilo_fix): Declare.
2415 (append_insn): Conditionally insert nops after an mfhi/mflo insn.
2416 (md_parse_option): Check for 7000_HILO_FIX options.
2417 (OPTION_M7000_HILO_FIX): Define.
2418 (OPTION_NO_M7000_HILO_FIX): Define.
2419 * doc/c-mips.texi (-mfix7000): Describe.
2423 * listing.c (print_lines): Remove unused variable `end'.
2425 * config/tc-i386.c (md_assemble): Use `reloc()' to select reloc
2426 type for JumpInterSegment output. Use enum bfd_reloc_code_real for
2427 reloc_type when BFD_ASSEMBLER.
2428 (md_estimate_size_before_relax): Use enum bfd_reloc_code_real for
2429 reloc_type when BFD_ASSEMBLER. Move common code out of switch
2430 statement and quell signed vs. unsigned comparison warning.
2434 * config/tc-d10v.c (find_opcode): Add a symbol's value to
2435 the computed frag offset, rather than overwriting it.
2439 * config/tc-sh.c ("elf/sh.h"): Include.
2440 (sh_dsp, valid_arch, reg_x, reg_y, reg_efg): New static variables.
2441 (md.begin): Initialize target_arch.
2442 Only include opcodes in has table that match selected architecture.
2443 (parse_reg): Recognize register names for sh-dsp.
2444 (parse_at): Recognize post-modify addressing.
2445 (get_operands): The leading space is now optional.
2446 (get_specific): Remove FDREG_N support. Add support for sh-dsp
2447 arguments. Update valid_arch.
2448 (build_Mytes): Add support for SDT_REG_N.
2449 (find_cooked_opcode): New function, broken out of md_assemble.
2450 (assemble_ppi, sh_elf_final_processing): New functions.
2451 (md_assemble): Use find_cooked_opcode and assemble_ppi.
2452 (md_longopts, md_parse_option): New option: -dsp.
2453 * config/tc-sh.h (elf_tc_final_processing): Define.
2454 (sh_elf_final_processing): Declare.
2458 * config/tc-hppa.c (pa_build_unwind_subspace): Use subseg_new to create
2459 the unwinder subspace. Save the current seg/subseg before creating
2464 * config/tc-mcore.c (INST_BYTE0): Redefine to handle big and
2465 little endian targets.
2466 (INST_BYTE1): Redefine to handle big and little endian
2468 (cpu_type): New type: Select between M340 and M210.
2469 (parse_psrmod): New function: Parse the PSRCLR and PSRSET
2470 instructions of the M340.
2471 (md_assemble): Add support for the MULSH and OPSR classes of
2473 (md_atof): Add support for little endian targets.
2474 (md_parse_option): Add support for -EL, -EB and -mcpu command
2476 (md_convert_frag): Add support for little endian targets.
2477 (md_apply_fix3): Add support for little endian targets.
2478 (md_number_to_chars): Add support for little endian targets.
2482 * read.c (read_a_source_file): If TC_START_LABEL_WITHOUT_COLON is
2483 defined, use it to verify the symbol just read should be a label.
2487 * app.c (do_scrub_chars): Handle "||" for parallel instructions
2488 when DOUBLEBAR_PARALLEL is defined. Avoid stripping whitespace
2489 around colons when KEEP_WHITE_AROUND_COLON is defined.
2490 * doc/internals.texi (CPU backend): Document DOUBLEBAR_PARALLEL
2491 and KEEP_WHITE_AROUND_COLON.
2495 * read.c (s_rept): Call do_repeat, which abstracts the repeat
2497 (do_repeat): New. Abstract repeat logic so that a "break" can be
2499 (end_repeat): New. Provide support for a "break" out of the
2501 * read.h: Add prototypes for new functions.
2505 * doc/internals.texi: Document NUMBERS_WITH_SUFFIX macro.
2506 * as.h: Provide a default NUMBERS_WITH_SUFFIX definition (zero).
2507 * expr.c: Handle numbers with suffixes if NUMBERS_WITH_SUFFIX is
2512 * read.c: Added elseif to directives table.
2513 * read.h: Added prototype for s_elseif.
2514 * doc/as.texinfo: Added description for elseif.
2515 * cond.c (s_elseif): New function
2519 * listing.c (print_lines): Remove conditionals causing bug in
2524 * as.h: Define OCTETS_PER_BYTE and OCTETS_PER_BYTE_POWER
2526 * frags.c (frag_new): Calculate fr_fix in octets
2527 (frag_now_fix) Return offset as target address offset (bytes).
2528 (frag_now_fix_octets) New - Return offset in octets (8-bit
2530 * frags.h: Added prototype for frag_now_fix_octets().
2531 Distinguish between octets and bytes in field descriptions.
2532 * listing.c (calc_hex): Account for octets vs bytes when
2533 printing addresses/offsets.
2534 (print_lines) Ditto. Also, if LISTING_WORD_SIZE is not 1, and
2535 target is little-endian, print the octets in a word in big-endian
2536 order so that the display looks like a proper hexadecimal number,
2537 instead of having the octets reversed.
2538 * read.c (do_align): When recording alignment, alignment power
2539 should be in terms of target bytes (minimum addressible unit)
2541 (do_org) Convert ORG target address (byte) argument into an
2542 octet offset when generating a variable fragment.
2543 * symbols.c (resolve_symbol_value): Symbol final value
2544 converted to a target address offset (bytes) from its octet offset.
2545 * config/obj-coff.c (coff_frob_symbol): Symbol target address
2546 offset (bytes) is adjusted by the frag offset (octets) converted
2548 (coff_frob_section) Section alignment power is in terms of bytes;
2549 convert it to an octet alignment power when calculating size (and
2550 size mask) in octets. Don't modify the section size in order to
2551 "align" it for TI COFF, since that format has a different method
2552 for storing alignment information.
2556 * stabs.c (generate_asm_file): Escape backslashes in stabs file
2557 entries, matching the way GCC generates them. If not escaped, the
2558 filename is encoded incorrectly.
2562 * config/tc-arm.c (reg_table): Add support for ATPCS register
2567 * config/obj-coff.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Don't define if
2569 * config/tc-ppc.h [OBJ_XCOFF] (OBJ_COPY_SYMBOL_ATTRIBUTES):
2571 * config/tc-ppc.c (ppc_fix_adjustable): Don't look at the frag
2572 of a symbol when we really care about its value.
2576 * config/tc-mcore.c (md_assemble): Give warning message if
2577 operands passes to instruction are more than the spec.
2581 * config/tc-arm.c (armadjust_symtab): If the assembler is in
2582 Thumb mode but the label seen was not declared as '.thumb_func'
2583 then set the ST_INFO type to STT_ARM_16BIT mode. This allows
2584 correct disassembly of Thumb code bounded by non function labels.
2588 * Makefile.am (MULTI_CFILES): Add config/e-i386aout.c
2589 Add dependencies for e-i386aout.o. Fix 2 comment lines.
2591 * Makefile.in: Same here.
2594 * configure.in: Set bfd_gas for i386-aout when primary target
2595 is bfd. Handle i386aout emulation. Don't use te_file=multi, as
2596 we may need the primary te_file. Remove incorrect comment.
2598 * configure: Regenerate.
2600 * config/e-i386aout.c: New file.
2602 * as.c (USE_EMULATIONS): Move to before print_version_id.
2603 (struct emulation): Add i386aout.
2604 (show_usage): Split text strings. Reformat -a text. Add --em
2608 * obj.h (struct format_ops): Add s_get_other and s_get_desc.
2609 (aout_format_ops): New.
2612 * read.c (s_lcomm_internal): Rewrite OBJ_AOUT,OBJ_BOUT
2613 preprocessor conditional and add aout USE_EMULATIONS tests.
2614 (read_a_source_file): Don't pass error strings to printf as
2618 * gasp.c (exp_get_abs): Don't pass error strings to printf as
2620 (do_data): Same here.
2621 (process_file): And here.
2624 * symbols.c (colon): Rewrite "already defined" fatal message
2625 code for aout with USE_EMULATIONS.
2628 * config/obj-aout.c (OBJ_HEADER): Define.
2629 (obj_pseudo_table): Rename to aout_pseudo_table. Init all
2631 (obj_aout_frob_symbol): Expand S_GET_DESC, S_GET_TYPE,
2632 S_GET_OTHER, S_SET_TYPE macros since we don't need obj-multi
2634 (obj_aout_type): Expand S_SET_OTHER here too.
2635 (obj_read_begin_hook): Remove.
2636 (aout_pop_insert): New.
2637 (obj_aout_s_get_other): New.
2638 (obj_aout_s_get_desc): New.
2639 (aout_format_ops): New.
2642 * config/obj-aout.h (obj_pop_insert): Define so non-multi usage
2643 gets aout_pseudo_table.
2644 (aout_pseudo_table): Declare.
2645 (obj_read_begin_hook): Define.
2648 * config/obj-coff.c (obj_pseudo_table): Rename to
2650 (coff_pop_insert): Use coff_pseudo_table.
2651 (coff_sec_sym_ok_for_reloc): Remove.
2652 (coff_format_ops): Add 0 entries for s_get_size, s_set_size,
2653 and comment all zero entries and remove #if 0 code.
2656 * config/obj-coff.h (obj_pop_insert): Define.
2657 (coff_pseudo_table): Declare.
2660 * config/obj-ecoff.c (ecoff_format_ops): Add 0 entries for
2661 s_get_size, s_set_size. Comment all zero entries.
2664 * config/obj-elf.c (elf_s_get_other): New function.
2665 (obj_read_begin_hook): Rename to elf_obj_read_begin_hook.
2666 (obj_symbol_new_hook): Rename to elf_obj_symbol_new_hook.
2667 (elf_format_ops): Add elf_s_get_other, 0 s_get_size entry, and
2669 (obj_elf_parse_section_letters): Don't pass error strings to
2670 printf as format arg.
2673 * config/obj-elf.h (ECOFF_DEBUGGING): Define when
2675 (elf_s_get_other): Declare.
2676 (S_GET_OTHER) Define as elf_s_get_other if not already
2678 (S_SET_OTHER): Only define when not already defined.
2679 (elf_obj_read_begin_hook): Declare.
2680 (obj_read_begin_hook): Define.
2681 (elf_obj_symbol_new_hook): Declare.
2682 (obj_symbol_new_hook): Define.
2685 * config/obj-multi.h: Add copyright header and protect against
2686 multiple inclusion. Add * to all function pointers.
2687 (OBJ_HEADER): If defined, include it rather than other defines
2689 (obj_frob_file_after_relocs): Test for NULL.
2690 (obj_symbol_new_hook): Here too.
2691 (obj_sec_sym_ok_for_reloc): And here.
2692 (S_GET_OTHER): Define.
2693 (S_GET_DESC): Define.
2694 (ECOFF_DEBUGGING): Remove as it's done in obj-elf.h
2695 (OBJ_MAYBE_ELF): Update comment.
2697 * config/tc-i386.c (i386_immediate): Add OBJ_MAYBE_AOUT to
2698 OBJ_AOUT preprocessor conditional and handle emulation by
2699 testing OUTPUT_FLAVOR.
2700 (i386_displacement): Here too.
2701 (md_section_align): Similarly here.
2702 (i386_target_format): Conditionally compile when more than one
2703 of OBJ_MAYBE_{ELF,COFF,AOUT} defined. Add aout case.
2704 (i386_immediate): Fix error message for aout BFD_ASSMBLER.
2705 (i386_displacement): Here too.
2708 * config/tc-i386.h (AOUT_TARGET_FORMAT): Define for each TE_*.
2709 Define TARGET_FORMAT for aout only when not multi.
2712 * config/te-multi.h: Delete file as it's identical to te-generic.h
2716 * config/tc-i386.h (DWORD_MNEM_SUFFIX): Delete.
2717 * config/tc-i386.c (DWORD_MNEM_SUFFIX): Rename all occurrences to
2720 * config/tc-i386.h (INTEL_DWORD_MNEM_SUFFIX): Rename to
2722 * config/tc-i386.c (INTEL_DWORD_MNEM_SUFFIX): Here too. Fix some
2727 * config/tc-mips.c (mips_do_align): New function.
2728 * config/tc-mips.h (md_do_align): Define.
2732 * doc/c-arm.texi (ARM Options): Fix typo.
2733 (ARM-Chars): Correct description of `#'. Mention that `;' is a
2734 line separator for Linux.
2735 * doc/as.texinfo (Comments): Mention the ARM.
2739 * configure.in (arm*-*-conix*): New target.
2740 (arm*-*-linux-gnu*): Match instead of arm-*-linux* and
2742 * configure: Regenerate.
2746 * config/obj-elf.c (elf_pseudo_table): Define visibility pseudos.
2747 (obj_elf_visibility): New function.
2749 * doc/as.texinfo (Visibility): New node: document visibility
2754 * config/tc-i386.c (MATCH): Relax JumpAbsolute check. Emit a
2755 warning for absolute jump/call without `*' in non-intel mode. No
2756 need to set i.types[0] JumpAbsolute in intel mode.
2760 * config/tc-arm.c (arm_s_text): If OBJ_ELF, call the appropriate
2761 hook function when changing sections.
2762 (arm_s_data): Likewise.
2766 * config/tc-arm.c (md_parse_option): Add support for -marm720
2767 command line switch.
2771 * config/tc-mn10300.c (md_pseudo_table): Add ".am33" pseudo-op.
2772 (r_registers, xr_registers): Define.
2773 (r_register_name, xr_register_name): New functions.
2774 (md_assemble): Handle new am33 operand types and instruction
2776 (mn10300_insert_operand, check_operand): Likewise.
2780 * config/tc-arm.c (thumb_mode): Turn into a tristate variable.
2781 (s_force_thumb): Set thumb_mode to 2.
2782 (md_assemble): Do not complain about thumb instructions on a
2783 non-thumb target if thumb_mode is set to 2.
2787 * config/tc-alpha.c (toplevel): Include struc-symbol.h.
2788 (alpha_macro_arg): Add MACRO_{LITERAL,BASE,BYTOFF,JSR} cases.
2789 (O_...): Add new machine dependent expressions if we are handling
2790 explicit relocations.
2791 (alpha_reloc_op): New static table holding the explicit relocation
2793 (alpha_literal_hash): New static to hold the hash table for
2794 explicit relocations.
2795 (alpha_macros): Add support for explicit relocations.
2796 (md_begin): If explicit relocations, initialize hash table.
2797 (md_assemble): Don't print a second error if tokenize_arguments
2798 already printed an error message.
2799 (md_apply_fix): Add support for explicit relocations.
2800 (alpha_force_relocation): Ditto.
2801 (alpha_fix_adjustable): Ditto.
2802 (alpha_adjust_symtab): New function to support explicit
2804 (alpha_adjust_symtab_relocs): Ditto.
2805 (debug_exp): Debug stub compiled if DEBUG_ALPHA is defined.
2806 (tokenize_arguments): Add debug code if DEBUG_ALPHA is defined.
2807 Add support for explicit relocations. Return -2 if an error
2808 message was already printed.
2809 (find_macro_match): Add support for explicit relocations. Comment
2811 (emit_insn): Add support for explicit relocations.
2812 (assemble_tokens): Ditto.
2814 (load_expression): Ditto.
2817 (emit_ir_load): Ditto.
2818 (emit_loadstore): Ditto.
2821 (emit_sextX): Ditto.
2822 (emit_division): Ditto.
2823 (emit_jsrjmp): Ditto.
2824 (emit_retjcr): Ditto.
2826 * config/tc-alpha.h (RELOC_OP_P): Enable explicit relocations if
2828 (tc_adjust_symtab): If explicit relocations, call the function
2829 alpha_adjust_symtab.
2830 (TC_FIX_TYPE): Add fields to be able to move explicit lituse
2831 relocations next to the literal relocation they reference.
2832 (TC_INIT_FIX_DATA): Initialize the new fields.
2833 (TC_FIX_DATA_PRINT): Print the new fields if DEBUG5 is defined.
2837 * config/tc-hppa.c (pa_ip): Handle PA2.0 unit completers. Handle
2838 'B' operand for PA2.0 bb instruction.
2842 * config/tc-mcore.h (TC_FORCE_RELOCATION): Define for Mcore-pe
2845 * config/tc-mcore.c (tc_gen_reloc): Support generation of RVA
2847 (mcore_force_relocation): Force relocations to be generated for
2852 * config/tc-i386.c (i386_immediate): Disallow O_big immediates.
2853 (i386_displacement): Disallow O_big displacements.
2857 * config/tc-arm.c (do_mia,do_mar,do_mra,do_pld,do_ldrl,do_co_reg):
2858 Small improvements in error checking.
2859 (md_assemble): Support for unconditional ARM instructions.
2860 (md_parse_option): Support for -m[arm]v5e flag.
2864 * macro.c (buffer_and_nest): Do not check beyond the end of the
2869 * macro.c (buffer_and_nest): Look for seperator after TO and
2874 * app.c (do_scrub_chars): When in State 10, treat backslash
2875 characters in the same way as as symbol characters.
2879 * config/tc-alpha.c (alpha_align): Check, don't assert, that
2880 the previous label was in the current section before playing
2881 with auto-alignment.
2885 * config/tc-v850.c (v850_force_relocation): Force relocation
2887 (v850_pcrel_from_selection): Do not compute a pcrel offset if
2892 * expr.h (operatorT): Increase machine dependent operators to 16.
2893 * expr.c (op_rank): Ditto.
2897 * read.c (pseudo_set): Reject attempts to set the value of a
2900 * config/obj-elf.c (obj_elf_ident): Call md_flush_pending_output
2903 * config/obj-elf.c (elf_set_index): Add ATTRIBUTE_UNUSED.
2905 * config/obj-elf.c (elf_frob_file_after_relocs): Don't pass NULL
2906 to bfd_set_section_contents.
2910 * config/tc-mcore.h (TARGET_BYTES_BIG_ENDIAN): Change to false.
2914 * config/tc-mips.c (macro_build): Use OPCODE_IS_MEMBER.
2915 (mips_ip): Use OPCODE_IS_MEMBER.
2919 * config/tc-arm.c (reg_required_here): Improve comments.
2921 * config/tc-arm.c (thumb_opcode): Add "variants" field.
2922 (tinsns): Initialize variants field.
2924 * config/tc-arm.c (bad_args, bad_pc): Renamed to BAD_ARGS and
2925 BAD_PC respectively.
2929 * config/tc-arm.c (reloc_map[]): Fix compiler warning.
2930 * config/tc-arm.h: Fix compile time warnings.
2934 * tc_d10v.c (find_opcode): Allow ATSIGN to match expressions of the
2935 form @abs16, @(abs16) and @(abs16 + imm).
2939 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS) : New.
2940 (ISA_HAS_64_BIT_REGS) New.
2941 (gpr_interlocks,md_begin,reg_needs_delay,append_insn,
2942 mips_emit_delays,macro_build,load_register,load_addresss,
2943 macro,macro2,mips_ip,s_cprestore,s_cpadd): Simplify
2944 and/or use new ISA_xxx macros in expressions involving
2945 ISA, particularly mips_opts.isa.
2949 * expr.h (operatorT): Add machine dependent operators md1..md8.
2950 (expressionS): Make X_op 8 bits instead of 7. Add a X_md field
2951 for the machine dependent operators to use.
2953 * expr.c (op_rank): Add machine dependent operators.
2955 * config/tc-alpha.c (O_pregister): Define as a machine dependent
2957 (O_cpregister): Ditto.
2958 (md_begin): Change X_op test that field is wide enough to use
2959 O_max instead of O_alpha_max.
2960 (cpu_types): Fill in missing initializer.
2961 (alpha_num_macros): Make unsigned.
2962 (md_assemble): Make opnamelen be size_t.
2963 (md_apply_fix): Cast alpha_num_operands to int before testing.
2964 (alpha_force_relocation): Ditto.
2965 (alpha_fix_adjustable): Ditto.
2966 (alpha_fix_adjustable): Mark unused arguments ATTRIBUTE_UNUSED.
2967 (tc_gen_reloc): Ditto.
2968 (tc_get_register): Ditto.
2973 (s_alpha_ent): Ditto.
2974 (s_alpha_end): Ditto.
2975 (s_alpha_frame): Ditto.
2976 (s_alpha_prologue): Ditto.
2977 (s_alpha_file): Ditto.
2978 (s_alpha_gprel32): Ditto.
2979 (s_alpha_proc): Ditto.
2980 (s_alpha_set): Ditto.
2981 (s_alpha_base): Ditto.
2982 (s_alpha_align): Ditto.
2983 (s_alpha_arch): Ditto.
2984 (alpha_align): Ditto.
2985 (assemble_insn): Suppress unused variable warning.
2987 (assemble_insn): Don't assume X_op and X_unsigned are in a given
2988 order in the structure.
2989 (s_alpha_coff_wrapper): Avoid int/unsigned comparison.
2993 * config/tc-hppa.c (md_apply_fix): Make "fmt" an int.
2997 * config/tc-i386.c (i386_index_check): Correct #endif location.
3001 * as.c (show_usage): Document new options.
3002 (parse_args): Add --no-warn, --warn, --fatal-warnings,
3003 which become 'W', OPTION_WARN, and OPTION_WARN_FATAL.
3004 (parse_args): Parse the new options.
3005 (main): If there were warnings, and --fatal-warnings
3006 was specified, print an error.
3007 * as.h: New variable, flag_fatal_warnings, for new option.
3011 * config/tc-hppa.c (pa_ip): Add new codes 'cc', 'cd', 'cC', 'co',
3012 '@'. Change autoincrement completers to fall through to cache control
3015 * config/tc-hppa.c (pa_ip): Remove unused args. Add code to '?W'
3017 (pa_parse_addb_64_cmpltr): New function.
3019 * config/tc-hppa.c (pa_ip): Change error message.
3020 (pa_ip,pa_parse_cmpb_64_cmpltr,pa_parse_cmpib_64_cmpltr) Fix '?N' and
3021 '?Q' args to allow falling through.
3023 * config/tc-hppa.c (pa_ip): Implement conditional codes "?N", "?Q".
3024 Remove unused conditional codes.
3025 (pa_parse_cmpb_64_cmpltr,pa_parse_cmpib_64_cmpltr): New.
3029 * config/tc-d30v.c (CHAR_BIT): Define.
3030 (check_range): Fix bit operations to support integers bigger than
3035 * config/tc-d10v.c (check_range): Check range for RESTRICTED_NUM3
3041 Add support for m32rx.
3042 * config/tc-m32r.c (enable_m32rx): New static global.
3043 (enable_special,warn_explicit_parallel_conflicts,optimize): Ditto.
3044 (allow_m32rx): New function.
3045 (M32R_SHORTOPTS): Add `O'.
3046 (md_longopts): Add --m32rx plus several warning options.
3047 (md_parse_option): Handle new options.
3048 (md_show_usage): Print them.
3049 (md_begin): Enable m32rx.
3050 (OPERAND_IS_COND_BIT): New macro.
3051 (first_writes_to_seconds_operands): New function.
3052 (writes_to_pc,can_make_parallel,make_parallel): New functions.
3053 (target_make_parallel,assemble_two_insns): New functions.
3054 (md_assemble): Recognize "insn1 -> insn2" and "insn1 || insn2".
3055 If optimizing and m32rx, try to make consecutive insns parallel.
3059 * config/tc-mips.c (nopic_need_relax): Allow for the
3060 .sdata.foo sections generated by -fdata-sections,
3061 and for the .gnu.linkonce.s sections generated by C++.
3065 * config/tc-hppa.c (pa_ip): Replace 'B', 'M', 'l' and 'g' handling
3066 with cleaner code using completer prefixes. Add 'Y'.
3068 * config/tc-hppa.c (pa_ip): Add parens to silence compiler.
3072 * config/tc-hppa.c (pa_ip): Avoid ANSI specific initialization.
3073 (pa_ip, case 'm'): Failure to get a CBIT specifier just means the
3074 insn does not match and we should try the next insn in the table.
3078 * config/tc-arm.c (MULTI_SET_PSR): Rename to LDM_TYPE_2_OR_3.
3082 * config/tc-hppa.c (pa_ip): Fix thinkos in recent cleanup
3087 * config/tc-i386.c (md_shortopts): Check OBJ_MAYBE_ELF as well as
3088 OBJ_ELF. If ELF, add "sq".
3089 (md_parse_option): If ELF, ignore -s and -q.
3090 (md_show_usage): Mention ELF options.
3094 * config/tc-hppa.c (pa_ip): Handle 'J', 'K' and 'cc'
3097 * config/tc-hppa.c (pa_ip); Handle "fe", and 'cJ'.
3099 * config/tc-hppa.c (pa_ip): Handle 'd', '#' and 'cq'.
3101 * config/tc-hppa.c (struct pa_it): New field "trunc".
3102 (pa_ip): Hadnle 'h', 'm', '=', '{', and '_' operands.
3103 (pa_parse_ftest_gfx_completer): New function
3104 (pa_parse_fp_cnv_format): New function.
3106 * config/tc-hppa.c (pa_ip): Handle 'X' operand.
3107 (md_apply_fix): Handle 22bit pc-rel branches.
3109 * config/tc-hppa.c (pa_ip): Handle 'B' operand.
3111 * config/tc-hppa.c (pa_ip): Handle 'L' and 'M' operands.
3113 * config/tc-hppa.c (pa_ip): Handle 'l' operand.
3115 * config/tc-hppa.c (pa_ip): Handle 'g' operand.
3119 * config/tc-hppa.c (md_assemble): Fix dwarf2 line handling.
3120 (pa_ip): Handle 'fX'.
3124 * config/tc-hppa.c (pa_build_unwind_subspace): Do not build
3125 unwinds unless the function is in the text space.
3126 (pa_type_args): Set BSF_FUNCTION for an exproted data symbol.
3130 * config/tc-hppa.c (pa_ip): Move dwarf2 stuff from here.
3131 (md_assemble): To here. Tweak address generation.
3133 * config/tc-hppa.c: Include dwarf2dbg.h if OBJ_ELF. Declare
3135 (md_pseudo_table): Add .file and .line pseudo-ops for OBJ_ELF.
3136 (md_assemble): Call dwarf2_where for OBJ_ELF.
3137 (pa_ip): Call dwarf2_gen_line_info for OBJ_ELF.
3138 (pa_end_of_source): New function.
3139 * tc-hppa.h (md_end): Define for OBJ_ELF.
3143 * configure.in (Canonicalization of target names): Remove adding
3144 ${CONFIG_SHELL} in front of $ac_config_sub, since autoconfig 2.14
3145 generates $ac_config_sub with a ${CONFIG_SHELL} already.
3146 * configure: Regenerate.
3150 * config/te-interix.h (GLOBAL_OFFSET_TABLE_NAME): Define.
3154 * config/tc-i386.c (md_assemble): Handle "jmp/call constant" as a
3155 pc-relative jmp/call to an absolute symbol.
3156 (md_apply_fix3): When OBJ_ELF, don't add the values in twice for
3157 absolute section symbols.
3159 * config/tc-i386.c (md_assemble): Correct frag_var size. Tidy
3160 jump handling code and comments.
3164 * config/tc-i386.c (md_apply_fix3): Add horrible adjustments to
3165 the value if TE_PE and a global defined symbol.
3169 * write.c (dump_section_relocs): Call print_symbol_value_1 to
3170 print the symbol, rather than printing it here.
3174 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust
3175 BFD_RELOC_RVA relocations.
3177 * config/tc-i386.c (md_undefined_symbol): Compare the name against
3178 the macro GLOBAL_OFFSET_TABLE_NAME, rather than assuming that it
3181 * write.c (write_relocs): Call SET_SECTION_RELOCS if it is
3183 * config/obj-coff.h (SET_SECTION_RELOCS): Define.
3184 * doc/internals.texi (Object format backend): Document
3187 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust
3188 relocations against global symbols if TE_PE.
3190 * config/obj-coff.c (obj_coff_ident): Add BFD_ASSEMBLER version.
3191 (obj_pseudo_table): Always handle ".ident" with obj-coff_ident.
3193 * config/obj-coff.c (coff_frob_symbol): Prohibit weak common
3196 * config/obj-coff.c (obj_coff_endef): Don't merge labels, or
3197 symbols which do not have a constant value, or tags with
3198 non-tags. Remove the symbol from the list before adding it at the
3201 * config/obj-coff.c (obj_coff_endef) [BFD_ASSEMBLER]: Handle .ef
3202 C_FCN symbol differently if TE_PE.
3203 (obj_coff_line) [BFD_ASSEMBLER]: Always use the line number which
3204 appears in the pseudo-op, rather coff_line_base which is only set
3207 * config/obj-coff.c (obj_coff_loc): New static function.
3208 (obj_pseudo_table): Add "loc".
3210 * config/obj-coff.c (add_lineno): Check that the line number is
3213 * config/atof-ieee.c (atof_ieee): Change what_kind to int.
3214 * config/atof-vax.c (flonum_gen2vax): Change format_letter to
3216 (md_atof): Return NULL rather than 0.
3217 * config/tc-i386.c (md_atof): Change type to int.
3218 * expr.c (expr): Change first parameter to int.
3219 * config/obj-coff.c: Add declarations for static functions.
3220 (coff_frob_symbol): Use SYM_AUXENT.
3221 * config/tc-i386.h (flag_16bit_code): Don't declare.
3223 * config/obj-coff.c (obj_coff_section): Default to setting
3224 SEC_LOAD. Don't set SEC_DATA for 'w' modifier.
3226 * write.c (adjust_reloc_syms): Print adjusted fixup.
3228 * expr.c (integer_constant): Correct too_many_digits calculation
3233 * doc/c-arm.texi: Fix arguments of @var to not contain
3238 * config/tc-arm.c (s_thumb_set): Only support interworking for ELF
3240 (md_parse_option): Only support -k flag for ELF and COFF targets.
3244 * config/tc-hppa.c (pa_ip): Put strict register checks before
3245 call to pa_parse_number.
3247 * config/tc-hppa.c (pa_ip): Support 'Z' argument.
3251 * config/obj-coff.c: Add ATTRIBUTE_UNUSED as needed for
3256 * as.c (perform_an_assembly_pass): Set SEC_DATA for data_section.
3260 * config/tc-hppa.c (pa_ip): Get strict/non-strict mode from the
3261 candidate instruction. Require registers for register arguments
3262 when in strict mode. Require assemble-time constants for
3263 constants when in strict mode.
3264 (pa_get_absolute_expression): Require a constant when in strict
3269 * config/tc-m32r.c (md_longopts): Fix value for -Wnuh.
3273 * config/tc-pj.c: New file, supports picoJava in ELF.
3274 * config/tc-pj.h: Ditto.
3275 * configure.in (pjl*, pj*): New targets.
3276 * Makefile.am: Rebuild dependencies.
3277 (CPU_TYPES): Add pj.
3278 (TARGET_CPU_CFILES): Add config/tc-pj.c.
3279 (TARGET_CPU_HFILES): Add config/tc-pj.h.
3280 * doc/c-pj.texi: New file.
3281 * doc/as.texinfo: Add some PJ specifics.
3282 * doc/all.texi: Add PJ to the list of all architectures, sort them
3284 * doc/Makefile.in (CPU_DOCS): Add c-pj.texi.
3285 * configure, Makefile.in, doc/Makefile.in: Rebuild.
3289 * config/obj-multi.h: Include obj-elf.h if OBJ_MAYBE_ELF. Reformat.
3290 (obj_frob_file): Test for null pointer.
3291 (OBJ_COPY_SYMBOL_ATTRIBUTES): Here too.
3292 (OBJ_PROCESS_STAB): And here.
3293 (elf_obj_sy): Remove
3295 * config/obj-elf.h: #ifndef everything defined in obj-multi.h,
3296 except OBJ_PROCESS_STAB, which we #undef for ecoff.
3297 (elf_obj_sy): Remove #ifndef OBJ_SYMFIELD_TYPE.
3299 * config/obj-coff.c (no_func): Remove.
3300 (coff_format_ops): Change occurrences of no_func to 0, as we test
3301 for 0 in obj-multi.h.
3303 * configure.in: Enable bfd for i386-coff when primary target is
3304 bfd. Enable i386 elf,coff emulation support. Don't set
3305 USE_EMULATIONS=1 or te_file=multi unless there is more than one
3306 emulation to support.
3307 *configure: Regenerate.
3311 * config/tc-mcore.c (mcore_s_section): Do not dump literals if a
3312 .section .line directive is encountered.
3316 * config/tc-arm.c (md_section_align): Do not align sections in ELF
3319 * as.c (show_usage): Add --gdwarf2 to list of options displayed.
3320 * as.texinfo: Document --gdwarf2 command line option.
3321 Add additional documentation of ARM command line switches.
3325 * config/tc-i386.c (i386_intel_memory_operand): Combine
3326 i386_is_reg and parse_register calls. Remove END_STRING_AND_SAVE
3327 and RESTORE_END_STRING around parse_register calls.
3328 (i386_operand): Here too.
3329 (i386_is_reg): Remove.
3330 (parse_register): Move as_bad calls from within this function to
3336 * config/tc-i386.c (parse_register): Handle FP regs specially.
3337 (md_begin): Remove '(' and ')' from register_chars.
3341 * config/tc-m32r.c (md_parse_option): Delete unrecognized option
3342 error message (done elsewhere).
3346 * config/tc-hppa.c (pa_ip): Do not allow '*' in 32bit completers.
3350 * config/tc-hppa.c (pa_ip): Replace 'f' by 'v'. Prefix float register
3353 * config/tc-hppa.c (pa_ip): Add args q, %, and |.
3355 * config/tc-hppa.c (pa_ip): Absorb white space in instructions
3357 Add new completers. Fix bug in 64 bit condition handling.
3359 * config/tc-hppa.c (pa_ip): Add completer codes 'a', 'ch', 'cH',
3362 * config/tc-hppa.c (pa_ip): Place completers behind prefix 'c'.
3364 * config/tc-hppa.c (pa_ip): Add cases for '.', '~'. '$'. and '!'
3366 * config/tc-hppa.c (pa_ip): Add case for 'I'.
3370 * dwarf2dbg.c (MAX_SPECIAL_ADDR_DELTA): Correct typo in comment.
3371 (struct ls): Add frag field. Initialize it to zero.
3372 (out_end_sequence): New local text_frag. Set it while in text section.
3373 Replace address check with frag check. Set ls.frag to text_frag if
3374 out_set_addr called.
3375 (dwarf2_gen_line_info): Add explanatory comment. New local saved_frag.
3376 Set it before switching sections. Replace address check with frag
3377 check. Set ls.frag to saved_frag if out_set_addr called.
3381 * dwarf2dbg.c (out_end_sequence): If address changed, directly
3382 output "advance_pc" opcode instead of calling gen_addr_line().
3383 The latter has the undesired side-effect of creating a new row
3384 in the debug line info matrix.
3388 * dwarf2dbg.c (out_end_sequence): Correct comments. Set last to
3389 ls.last_filename if last is less than zero. Set ls.last_filename
3390 when allocating new entry.
3391 (dwarf2_gen_line_info): Save seg and subseg info before subseg_new
3396 * config/tc-i386.c (i386_index_check): Fix the displacement size
3397 when INFER_ADDR_PREFIX.
3401 * config/tc-arm.c (md_apply_fix3): If an offset is invalid,
3406 * config/tc-ppc.c (md_assemble): Trim @ha constant to 16 bits, to
3407 handle 0xffffNNNN constants correctly.
3411 * config/tc-arm.c (do_ldst): Look for register conflicts on stores
3416 * config/tc-arm.c (validate_offset_imm): Work on unsigned values.
3417 (md_apply_fix3): Always pass positive values to
3418 validate_offset_imm.
3422 * config/tc-arm.c (skip_whitespace): New macro.
3423 Formatting tidy ups.
3425 (md_apply_fix3): Store relocation offset in addend for ELF based
3427 (arm_force_relocation): Always generate relocs for Thumb function
3432 * config/tc-i386.c (md_assemble): Remove dead code. intel_syntax
3433 LONG_DOUBLE_MNEM_SUFFIX floating point is done in opcode/i386.h
3437 * config/tc-hppa.c (pa_build_unwind_subspace): Use sane section
3438 flags for the unwind subspace.
3440 * config/tc-hppa.c (UNWIND_SECTION_NAME): Define for ELF.
3441 (pa_build_unwind_subspace): Remove #if 0 wrapper. Select a
3442 suitable relocation based on the size of the target's pointer.
3443 Always Use subsegment zero for the unwinders.
3444 (pa_level): Handle "2.0w".
3448 * config/tc-d30v.c (write_2_short): Don't group repeat instructions
3449 with the following instruction unless this was specified.
3453 * config/tc-i386.h (SUB_SEGMENT_ALIGN): If TE_GO32, return 4 for
3454 certain sections, to match BFD changes.
3458 * Makefile.am (noinst_SCRIPTS): Change .gdbinit to $(GDBINIT).
3459 (EXTRA_SCRIPTS): Define to keep automake happy.
3460 * Makefile.in: Rebuild.
3464 * Makefile.am: Rename .dep* files to DEP*.
3465 (MKDEP): Rename from DEP. Change all uses. Use $${srcdir} rather
3466 than $(srcdir). Rename TCDEP targets to DEPTC. Rename OBJDEP
3468 * Makefile.in: Rebuild.
3472 * config/tc-sparc.c (sparc_ip): Allow assembly of %lo()+%reg.
3476 * Makefile.am: Change all uses of itbl-test-ops to itbl-tops to
3477 avoid problems on DOS filesystems.
3478 * Makefile.in: Rebuild.
3480 * doc/as.texinfo (Section): Document 's' flag for COFF version.
3484 * config/obj-coff.c (obj_coff_section): Handle 's' (shared)
3489 * configure.in: Define and substitute GDBINIT. Change AC_OUTPUT
3490 line to create ${GDBINIT} rather than .gdbinit.
3491 * configure, Makefile.in, doc/Makefile.in: Rebuild.
3495 * config/tc-hppa.c (pa_ip, case '?'): Add missing break.
3499 * config/tc-hppa.c (pa_ip): Add 64 bit condition completers.
3503 * config/tc-sparc.h (tc_fix_adjustable): Fix check for PIC local
3504 non-adjustable symbols.
3508 * config/tc-hppa.c (pa_ip): Change condition args to have '?' prefix.
3512 * config/tc-sh.c (md_assemble): Call as_bad when there are excess
3517 * config/te-interix.h: New file.
3518 * configure.in (i386-*-interix*): New target.
3519 * configure: Rebuild.
3523 * config/tc-hppa.c (pa_chk_field_selector): Allow 3 byte
3524 selectors for ELF too.
3525 (selector_table): Add "ltp" and "rtp" selectors.
3529 * config/tc-i386.c (i386_operand): No need to change
3530 operand_string pointer in segment reg case before goto
3531 do_memory_reference. Initialise displacement_string_start and
3532 displacement_string_end after do_memory_reference label.
3533 (i386_index_check): Add operand_string param, and print error
3534 message on failure here.
3535 (i386_intel_memory_operand): Instead of here.
3536 (i386_operand): And here.
3537 (INFER_ADDR_PREFIX): Enable.
3539 * doc/c-i386.texi (i386-16bit): Document .code16gcc.
3541 * config/tc-i386.h (DefaultSize): Define. Renumber following
3542 opcode_modifier defines.
3545 * config/tc-i386.c (stackop_size): New variable.
3546 (set_16bit_code_flag): Clear it here.
3547 (set_16bit_gcc_code_flag): New function.
3548 (md_pseudo_table): Add "code16gcc" entry.
3549 (md_assemble): Set i.suffix for insns with DefaultSize modifier.
3553 * config/obj-coff.c (coff_frob_symbol): Always update set_end with
3554 next_set_end even if the end symbol is being discarded.
3556 * gasp.c: Add ATTRIBUTE_UNUSED as needed for non-BFD_ASSEMBLER.
3557 * output-file.c, symbols.c, config/tc-i386.c: Likewise.
3558 * config/obj-coff.c: Likewise.
3559 (seg_info_type): Remove.
3560 (seg_info_off_by_4): Change to array of segT.
3561 (s_get_segment): Adjust accordingly.
3562 (obj_pseudo_table): Fully initialize sentinel entry.
3564 * config/tc-mips.c (append_insn): Correct INSN_SYNC test. From
3569 * config/tc-i386.c (f16_3): New. Fixes 16 bit 3 byte nop.
3573 * config/tc-i386.c: Indentation and white space changes.
3574 (i386_index_check): New function. Add INFER_ADDR_PREFIX code, but
3575 don't enable it by default.
3576 (i386_intel_operand): Remove redundant prototype.
3577 Move check on number of memory operands, and i.mem_operands++
3578 (i386_intel_memory_operand): To here.
3579 Remove i386_immediate code from here. Remove special case code
3580 for input and output using (%dx). Remove base/index checks and
3581 call i386_index_check instead. Save initial operand_string
3582 argument for error message.
3583 (i386_operand): Remove redundant prototype. Move base/index
3584 checks to i386_index_check.
3585 (i386_displacement): Move intel mode check for non-zero
3587 (i386_intel_memory_operand): To here.
3591 * config/tc-sparc.c (md_longopts): Add --no-undeclared-regs option.
3592 (sparc_ip): Warn if %g2 or %g3 register is used and not covered
3593 by .register pseudo-op if -64 and --no-undeclared-regs.
3594 (s_register, sparc_adjust_symtab): New functions.
3595 * config/tc-sparc.h (tc_adjust_symtab, sparc_adjust_symtab):
3596 Declare sparc_adjust_symtab as tc_adjust_symtab.
3597 * doc/c-sparc.texi: Add description of #ignore special literal
3598 for .register pseudo-op.
3602 * config/tc-arm.c (tc_gen_reloc): Record the vtable entry in
3603 the relocation's section offset.
3607 * write.c (fixup_segment): Fix generic error check overflow test.
3609 * config/tc-i386.c (pe): Change %d to %ld, %x to %lx, and cast
3610 X_add_number to long.
3614 * config/tc-hppa.c (pa_ip): Add 'J' and 'K' code
3619 * config/tc-sparc.h (tc_fix_adjustable): Don't adjust GOT, PLT, or
3624 * config/te-go32.h (COFF_LONG_SECTION_NAMES): Define.
3626 * configure.bat: Remove; obsolete.
3627 * config/go32.cfg: Likewise.
3631 * configure.in (i386-*-vxworks*): New target.
3632 * configure: Rebuild.
3636 * doc/c-sparc.texi: Document .register and .nword pseudo-ops.
3640 * config/tc-sparc.c (sparc_ip): Allow OLO10 relocations
3642 (output_insn): Put OLO10's secondary addend into tc_fix_data.
3643 (md_apply_fix3): Handle BFD_RELOC_SPARC_OLO10.
3644 (tc_gen_reloc): Return two relocs for OLO10, LO10 and SPARC13.
3645 * config/tc-sparc.h (RELOC_EXPANSION_POSSIBLE,
3646 MAX_RELOC_EXPANSION): Define.
3647 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FIX_DATA_PRINT): Likewise.
3651 * config/tc-i386.c (intel_float_operand): Add prototype, make static.
3652 (md_assemble): Localize *exp variable to if (fake_zero_displacement)
3653 block. Print a warning if an 8-bit or 16-bit constant
3654 displacement or immediate is truncated on output.
3655 (i386_immediate): Ensure Imm16 is always legal for a 16-bit mode
3657 (i386_operand): Disallow immediate jump absolute operand.
3661 * configure.in: Bump version number to 2.9.5.
3662 * configure: Rebuild.
3664 * dwarf2dbg.c (dwarf2_gen_line_info): Don't assume that long long
3669 * config/tc-hppa.c (md_pseudo_table): Add ".dword" pseudo-op.
3670 (cons_fix_new_hppa): Derive size of fixup from size of the object.
3674 * symbols.c (dollar_label_name): Prepend LOCAL_LABEL_PREFIX if it
3676 * config/tc-arm.h (LOCAL_LABEL_PREFIX): Define to '.' for ELF.
3678 * config/tc-arm.c (md_begin): Set F_SOFTFLOAT in the output file
3679 if -mno-fpu was given.
3680 (tc_gen_reloc): Fix typo. Delete bogus code related to GOTPC
3682 (cons_fix_new_arm): Remove misleading comments.
3686 * write.c (cvt_frag_to_fill): Use frag file and line in rs_org
3688 (relax_segment): Likewise. After giving a rs_org error, convert
3689 the frag to rs_align to avoid cascading errors.
3693 * config/tc-m68k.c: Add some ATTRIBUTE_UNUSED.
3697 * Many files: Changes to avoid gcc warnings: Add ATTRIBUTE_UNUSED
3698 as appropriate. Fill in structure initializations. Add variable
3699 initializations. Add casts.
3700 * dwarf2dbg.c (print_stats): Change i to size_t.
3701 * listing.c (listing_listing): Change list_line to unsigned int.
3705 * config/tc-ppc.h (tc_fix_adjustable) [OBJ_ELF]: Call S_IS_LOCAL
3706 rather than checking for \001 and \002 in symbol name.
3707 * config/tc-sparc.h (tc_fix_adjustable) [OBJ_ELF]: Likewise.
3711 * configure.in (hppa*-linux-gnu*): New target.
3712 * configure: Rebuilt.
3716 * doc/c-arm.texi (ARM Directives): Document .thumb_set directive.
3720 * config/tc-v850.c (v850_comm): Use symbol_get_obj() rather than
3721 accessing symbolP directly.
3725 * config/tc-hppa.h (tc_frob_symbol): Always punt "$global$" symbol
3730 * config/tc-arm.c (ARM_EXT_V5): Define.
3731 (ARM_ARCH_V5, ARM_ARCH_V5T): Define.
3732 (md_begin): Detect ARM v5 architectures.
3733 (md_parse_option): Accept arm v5 specification.
3734 (md_show_usage): Documment -marmv5 switch.
3736 * doc/c-arm.texi: Document -marmv5 command line option.
3738 * config/tc-arm.c (do_adrl): New function. Implement ADRL pseudo
3740 (validate_immediate_twopart): New function. Determine if a
3741 constant can be computed by two ADD instructions.
3742 (output_inst): Remove its command line parameter - it was never
3744 (md_apply_fix3): Support BFD_RELOC_ARM_ADRL_IMMEDIATE, used to
3745 implememt the ADRL pseudo op.
3746 (tc_gen_reloc): Generate a suitable error message if an ADRL
3747 instruction tries to generate a real reloc.
3749 * doc/c-arm.texi: Document NOP, ADR and ADRL pseudo ops.
3753 * config/tc-hppa.c (pa_ip): Convert the opcode and all completers
3758 * subsegs.c (subseg_text_p): Use 1/0 instead of true/false for
3759 non BFD_ASSEMBLER case.
3763 * config/obj-coff.c (obj_coff_section): Mark writable sections as
3768 * dwarf2dbg.c (dwarf2_gen_line_info): Don't call
3769 out_end_sequence() when the address decreases due to a new frag.
3770 (gen_dir_list): Set ls.file[i].dir to j + 1 (not j) because file
3771 numbering starts with 1.
3775 * config/tc-mcore.c (md_pseudo_table): Add .comm for ELF and allow
3777 (mcore_s_text): Call obj_elf_text for ELF target.
3778 (mcore_s_data): Call obj_elf_data for ELF target.
3779 (mcore_s_section): No longer ELF specific. Call obj_coff_section
3781 (mcore_s_bss): New function: Dump literal table before changing
3783 (mcore_s_comm): New function: Dump literal table before changing
3786 * config/obj-elf.c (obj_elf_common, obj_elf_data, obj_elf_text):
3787 No longer static functions.
3788 * config/obj-elf.h (obj_elf_common, obj_elf_data, obj_elf_text):
3789 Provide prototypes for these functions.
3793 * subsegs.c (subseg_text_p): Rewrite non BFD_ASSEMBLER case to use
3794 a list of names, to try obj_segment_name, and to try abbreviated
3795 names when using COFF without long section names.
3797 * config/tc-alpha.c: More use of symbol accessor functions.
3798 * config/tc-arc.c: Likewise.
3799 * config/tc-d30v.c: Likewise.
3800 * config/tc-fr30.c: Likewise.
3801 * config/tc-i860.c: Likewise.
3802 * config/tc-m88k.c: Likewise.
3803 * config/tc-mcore.c: Likewise.
3804 * config/tc-ns32k.c: Likewise.
3805 * config/tc-sparc.c: Likewise.
3806 * config/tc-v850.c: Likewise.
3808 * config/tc-arc.c (get_arc_exp_reloc_type): Change uses of
3809 sy_value with appropriate accessor functions.
3810 * config/tc-arm.c (md_apply_fix3): Likewise.
3811 * config/tc-d10v.c (AT_WORD_P): Likewise.
3812 * config/tc-v850.c (reg_name_search): Likewise.
3814 * config/obj-ecoff.c (obj_ecoff_set_ext): Change uses of bsym to
3815 use symbol_get_bfdsym instead.
3816 * config/tc-ppc.c (md_assemble): Likewise.
3817 * config/tc-v850.c (v850_comm): Likewise.
3821 * config/tc-arc.c (tc_gen_reloc): Use symbol_get_bfdsym to get at
3822 the symbol, rather than accessing the bsym member.
3823 * config/tc-d10v.c (tc_gen_reloc): Likewise.
3824 * config/tc-d30v.c (tc_gen_reloc): Likewise.
3825 * config/tc-mcore.c (tc_gen_reloc): Likewise.
3826 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
3827 * config/tc-mn10300.c (tc_gen_reloc): Likewise.
3828 * config/tc-ns32k.c (tc_gen_reloc): Likewise.
3829 * config/tc-tic30.c (tc_gen_reloc): Likewise.
3830 * config/tc-v850.c (tc_gen_reloc): Likewise.
3834 * tc-hppa.c (elf_hppa_reloc_type): Renamed from elf32_hppa_reloc_type.
3835 (hppa_gen_reloc_type): Conditionalize on BFD64.
3836 (tc_gen_reloc): Re-enable ELF relocations.
3837 * tc-hppa.h (TARGET_FORMAT): Handle elf64-hppa format.
3841 * config/tc-arm.c (ldst_extend): Add parentheses to avoid
3843 (do_ldst): Move assignment out of if condition.
3844 (md_apply_fix3): Add casts to avoid printf format warnings. Add
3845 parentheses to avoid warning.
3849 * config/tc-arm.c (arm_adjust_symtab): Use symbol_get_bfdsym()
3850 macro to get at the BFD symbol associated with a GAS symbol.
3854 * config/tc-ppc.c: Update for symbol handling changes.
3855 * config/obj-coff.c: Likewise.
3859 * tc-hppa.c: General cleanups of ELF support. No more spaces
3860 and subspaces for ELF.
3861 (GDB_DEBUG_SPACE_NAME): Delete definition for ELF.
3862 (GDB_STRINGS_SUBSPACE_NAME): Likewise.
3863 (GDB_SYMBOLS_SUBSPACE_NAME): Likewise
3864 (UNWIND_SECTION_NAME): Likewise.
3865 (space/subspace related structures): Conditionalize definitions
3867 (space/subspace directives and support routines): Conditionalize
3868 definitions and references/uses on OBJ_SOM.
3869 (label_symbol_struct): For ELF, track the symbol's segment. For
3870 SOM track its space.
3871 (pa_define_label, pa_undefine_label, pa_get_label): Corresponding
3873 (USE_ALIASES): Kill for both SOM & ELF.
3874 (pa_def_subspaces, pa_def_spaces): Corresponding changes.
3875 (pa_space, pa_subspace): Corresponding changes.
3876 (pa_spaces_begin): Corresponding chagnes.
3877 (md_begin): Do not muck around with space/subspace stuff for
3879 (md_apply_fix): Temporarily disable argument relocation stuff
3881 (tc_gen_reloc): Temporarily disable relocation generation for
3883 (pa_build_unwind_subspace): Similarly.
3887 * config/tc-arm.c (thumb_set): New pseudo op.
3888 (text, data, section): Override these pseudo ops with ARM
3890 (s_thumb_set): New function: Perform the same as a .set pseudo
3891 op, but also mark the alias'ed symbol as being a Thumb
3893 (arm_s_text): New function: Perform the same as the .text
3894 pseudo op, but dump the literal pool before changing
3896 (arm_s_data): New function: Perform the same as the .data
3897 pseudo op, but dump the literal pool before changing
3899 (arm_s_section): New function: Perform the same as the
3900 .section pseudo op, but dump the literal pool before changing
3902 (arm_cleanup): Do not reset the current section before dumping
3907 * config/tc-m32r.c (md_longopts): Fix OPTION_WARN_UNMATCHED and
3908 OPTION_NO_WARN_UNMATCHED entries.
3909 (md_parse_option): Generate a warning message if an unrecognised
3910 option is encountered.
3912 * config/tc-d10v.c (do_not_ignore_hash): New variable.
3913 (get_operands): When parsing an expression after an '@' symbol
3914 has been detected, do not ignore '#' symbols.
3915 (md_operand): Only ignore '#' symbols if do_not_ignore_hash is
3921 * configure.in (ppc-*-vxworks*): New target.
3922 * configure: Rebuild.
3926 * config/tc-arm.c (tc_gen_reloc): Fix handling of GOTPC relocs.
3930 * write.c (adjust_reloc_syms): Rather than never reducing reloc
3931 which refer to symbols in linkonce sections, permit reducing the
3932 relocs if the symbol is local.
3936 * subsegs.c (subseg_text_p): New function.
3937 * as.h (subseg_text_p): Declare.
3938 * read.c (do_align): Use subseg_text_p to set the default fill.
3939 * write.c (subsegs_finish): Likewise.
3940 * config/obj-coff.c (write_object_file): Likewise.
3941 * config/tc-i386.h (md_maybe_text): Don't define.
3942 (md_do_align): Use subseg_text_p to set the default fill.
3943 * config/tc-m32r.c (m32r_do_align): Likewise.
3944 * config/tc-sh.c (sh_do_align): Likewise.
3945 * config/tc-sparc.h (md_do_align): Likewise.
3949 * configure.in: (i[3456]86-*-freebsd*): Now defaults to ELF.
3950 * configure: Rebuild.
3954 * dwarf2dbg.c: Include elf/dwarf2.h with "", not <>.
3955 * Makefile.am: Rebuild dependencies.
3956 * Makefile.in: Rebuild.
3958 * config/tc-i386.c (i386_immediate): Remove unused label
3961 * struc-symbol.h: Put local_symbol code in ifdef BFD_ASSEMBLER.
3962 * symbols.c: Likewise.
3963 * config/obj-aout.c (obj_crawl_symbol_chain): Refer directly to
3964 sy_next field when taking address, rather than symbol_next.
3966 * dwarf2dbg.c: Change bfd_vma to addressT and bfd_signed_vma to
3968 (out_set_addr): Don't use BYTES_PER_ADDRESS. Instead, get the
3969 value from the output file architecture.
3970 (dwarf2_gen_line_info): Ifdef BFD_ASSEMBLER specific code.
3971 * dwarf2dbg.h: Change bfd_vma to addressT.
3975 * dwarf2dbg.h: Use PARAMS in function declarations.
3979 * write.c (fixup_segment): Don't add symbol value for i960 ELF.
3980 * config/tc-i960.c (s_leafproc): Don't call tc_set_bal_of_cal if
3982 (md_apply_fix): Simplify BFD_ASSEMBLER handling.
3986 * config/tc-i386.c (md_apply_fix3): Add default case to switch.
3988 * config/tc-sparc.c (md_pseudo_table): Remove pushsection and
3991 * config/tc-sparc.c (sparc_ip): Add default case to reloc switch.
3993 * read.c (read_a_source_file): Only declare inescape if
3996 * itbl-ops.c (itbl_disassemble): Change sprintf format strings to
3998 (find_entry_byval): Add parens to avoid warning.
4000 * as.c: If HAVE_ITBL_CPU, include "itbl-ops.h".
4002 * symbols.c (resolve_symbol_value): Don't permit subtraction of
4007 * config/tc-sparc.c (sparc_ip): Don't use side-effect expression
4010 * config/tc-sparc.c (synthetize_setuw, synthetize_setsw,
4011 synthetize_setx): New functions.
4012 (md_assemble): Broken the special cases into the above
4013 functions. Make compiler happy if sizeof(bfd_vma)==4.
4014 Fix sethi generated from set/setuw. If instructions have a relloc,
4015 always clear the fields to be relocated in the opcode.
4016 (sparc_ip): Remove special_case global variable.
4021 * itbl-ops.c (itbl_get_reg_val): Add pval parameter. Return
4022 indication of success rather than a value.
4023 (itbl_get_val): Likewise.
4024 (itbl_get_field): Use strcspn. Change delimiters to include
4026 * itbl-ops.h (itbl_get_reg_val): Update declaration.
4027 (itbl_get_val): Likewise.
4028 * config/tc-mips.c (mips_ip): Update call to itbl_get_reg_val.
4030 * symbols.c (copy_symbol_attributes): Convert local symbols to
4035 * config/tc-arm.c (md_parse_option): Add support for ARM920 and
4040 * config/tc-sparc.c (md_assemble): Fix up setx, support setsw.
4041 Optimize set if sizeof(bfd_vma) == 64.
4042 (sparc_ip): Fix sethi - without %hi() it should generate
4043 R_SPARC_22 reloc, not R_SPARC_HI22.
4044 (tc_gen_reloc): Handle BFD_RELOC_SPARC22.
4048 * config/tc-sparc.c (md_begin): Handle native wordsize aliases.
4049 (s_ncons): New function.
4050 (native_op_table): New table.
4051 (sparc_ip): Be more strict on %hi() etc.; prepare assembler for
4052 R_SPARC_OLO10 handling.
4056 * expr.h (struct expressionS): Revert last change; widen X_op.
4057 * config/tc-alpha.c (md_begin): Check the field is wide enough.
4061 * Makefile.am (TARGET_CPU_CFILES): Add config/tc-fr30.c.
4062 (TARGET_CPU_HFILES): Add config/tc-fr30.h.
4063 (TARG_ENV_HFILES): Add config/te-epoc-pe.h.
4064 * Makefile.in: Regenerated.
4066 * config/obj-elf.c (obj_elf_common): In MRI mode if called as
4067 `common' pass on to s_mri_common.
4068 (elf_pseudo_table): Pass 1 to obj_elf_common for `common'.
4072 * config/obj-elf.c (obj_elf_section): Don't free the return
4073 value of demand_copy_C_string.
4077 * dwarf2dbg.c (dwarf2_gen_line_info): Mirror the section symbol
4078 creation logic from obj_elf_create_section.
4080 * config/obj-elf.c (elf_pseudo_tab): Add pushsection/popsection.
4081 (section_stack): New.
4082 (special_sections): Make const.
4083 (obj_elf_section): Gut and rewrite parsing.
4084 (obj_elf_change_section): New function broken out of obj_elf_section.
4085 (obj_elf_parse_section_letters): Likewise.
4086 (obj_elf_section_word): Likewise.
4087 (obj_elf_section_type): Likewise.
4088 (obj_elf_previous): Treat as a toggle.
4089 (obj_elf_popsection): New.
4090 * config/tc-ppc.c (ppc_section_word): Take str+len not ptr_str.
4091 (ppc_section_type): Likewise.
4092 * config/tc-ppc.h: Likewise.
4094 * expr.h (struct expressionS): Don't make X_op a bitfield.
4095 * config/tc-alpha.c: Update for symbol handling changes.
4096 (md_apply_fix) [case GPREL]: Use now_seg instead of absolute_section.
4097 (load_expression, emit_ir_load, emit_loadstore, emit_jsrjmp): Likewise.
4101 * dwarf2dbg.c (*): Convert to K&R + prototypes.
4102 (dwarf2_gen_line_info): Kill unused variables.
4103 (dwarf2_finish): Likewise.
4104 (dwarf2_where): Likewise.
4105 (dwarf2_directive_file): If we've only got a string,
4106 hand off to s_app_file.
4107 * ecoff.c: Move the include of ecoff.h.
4108 * symbols.h (S_IS_FUNCTION): Prototype.
4110 * read.c (LEX_HASH): Supply a default.
4112 (s_globl): Update `c' after skipping whitespace.
4113 * read.h (LEX_END_NAME, is_name_ender): New.
4114 * expr.c (get_symbol_end): Respect it.
4118 * config/tc-hppa.c (md_begin): Convert local symbol dummy_symbol
4120 (tc_gen_reloc): Still need bfd_abs_symbol in some relocs.
4122 * config/tc-hppa.c: Update for symbol handling changes.
4126 * cgen.c: Update for symbol handling changes.
4127 * config/tc-m32r.c: Likewise.
4129 * config/tc-hppa.h: Update for symbol handling changes.
4130 * config/tc-hppa.c: Likewise.
4132 * config/tc-arm.h: Update for symbol handling changes.
4133 * config/tc-arm.c: Likewise.
4134 (symbol_make_empty): Remove. Just use symbol_create.
4136 * symbols.c (symbol_set_tc): Correct name.
4138 * Makefile.am: Rebuild dependencies.
4139 ($(OBJS)): Don't depend upon struc-symbol.h.
4140 (.dep1, .tcdep, .objdep): Create itbl-parse.h.
4141 * dep-in.sed: Don't remove struc-symbol.h.
4142 * Makefile.in: Rebuild.
4144 * doc/internals.texi (Symbols): Describe changes in symbol
4149 * dwarf2dbg.c (dwarf2_gen_line_info): Use section_symbol
4150 instead of doing the work by hand.
4154 * dwarf2dbg.c (INITIAL_STATE): New macro encapsulating initial
4155 state of line state-machine.
4156 (struct ls): Collect DWARF2 line state-machine state in new member
4157 SM. Add member EMPTY_SEQUENCE to keep track if a code sequence
4158 resulted in any DWARF2 directives.
4159 (reset_state_machine): New function.
4160 (out_end_sequence): Ditto.
4161 (dwarf2_gen_line_info): When switching sections or switching to a
4162 lower text address, call out_end_sequence() first to terminate the
4163 previous code sequence as code sequences MUST have monotonically
4164 increasing addresses.
4165 (dwarf2_finish): Call out_end_sequence() instead of open coding it.
4169 * as.c (parse_args): Add option -gdwarf2 to allow requesting
4170 DWARF2 debug info (line information only, at this point).
4171 * as.h: Update comment about supported debug formats.
4172 * dwarf2dbg.c, dwarf2dbg.h: New files.
4173 * Makefile.am (GAS_CFILES, HFILES, GENERIC_OBJS): Add them.
4175 * expr.c (operand): Don't use [ for parens if we want an index op.
4176 (op_encoding): Switch [ into O_index, if desired.
4177 (op_rank): Renumber with O_index on bottom.
4178 (expr): If O_index, match closing bracket.
4179 * expr.h (O_index): New.
4181 * read.c (read_a_source_file): Conditionally allow matched "
4182 in lines passed to md_assemble.
4184 * config/obj-elf.c (elf_pseudo_table): Add `common'.
4188 Add support for storing local symbols in a small structure to save
4189 memory when assembling large files.
4190 * as.h: Don't include struc-symbol.h.
4191 (symbolS): Add typedef.
4192 * symbols.c: Include struc-symbol.h.
4193 (local_hash): New static variable.
4194 (save_symbol_name): New static function, from symbol_create.
4195 (symbol_create): Call save_symbol_name.
4196 (local_symbol_count): New static variable.
4197 (local_symbol_conversion_count): Likewise.
4198 (LOCAL_SYMBOL_CHECK): Define.
4199 (local_symbol_make): New static function.
4200 (local_symbol_convert): New static function.
4201 (colon): Handle local symbols. Create local symbol for local
4203 (symbol_table_insert): Handle local symbols.
4204 (symbol_find_or_make): Create local symbol for local label name.
4205 (symbol_find_base): Check for local symbol.
4206 (symbol_append, symbol_insert): Check for local symbols.
4207 (symbol_clear_list_pointers, symbol_remove): Likewise.
4208 (verify_symbol_chain): Likewise.
4209 (copy_symbol_attributes): Likewise.
4210 (resolve_symbol_value): Handle local symbols.
4211 (resolve_local_symbol): New static function.
4212 (resolve_local_symbol_values): New function.
4213 (S_GET_VALUE, S_SET_VALUE): Handle local symbols.
4214 (S_IS_FUNCTION, S_IS_EXTERNAL, S_IS_WEAK, S_IS_COMMON): Likewise.
4215 (S_IS_DEFINED, S_IS_DEBUG, S_IS_LOCAL, S_GET_NAME): Likewise.
4216 (S_GET_SEGMENT, S_SET_SEGMENT, S_SET_EXTERNAL): Likewise.
4217 (S_CLEAR_EXTERNAL, S_SET_WEAK, S_SET_NAME): Likewise.
4218 (symbol_previous, symbol_next): New functions.
4219 (symbol_get_value_expression): Likewise.
4220 (symbol_set_value_expression): Likewise.
4221 (symbol_set_frag, symbol_get_frag): Likewise.
4222 (symbol_mark_used, symbol_clear_used, symbol_used_p): Likewise.
4223 (symbol_mark_used_in_reloc): Likewise.
4224 (symbol_clear_used_in_reloc, symbol_used_in_reloc_p): Likewise.
4225 (symbol_mark_mri_common, symbol_clear_mri_common): Likewise.
4226 (symbol_mri_common_p): Likewise.
4227 (symbol_mark_written, symbol_clear_written): Likewise.
4228 (symbol_written_p): Likewise.
4229 (symbol_mark_resolved, symbol_resolved_p): Likewise.
4230 (symbol_section_p, symbol_equated_p): Likewise.
4231 (symbol_constant_p): Likewise.
4232 (symbol_get_bfdsym, symbol_set_bfdsym): Likewise.
4233 (symbol_get_obj, symbol_set_obj): Likewise.
4234 (symbol_get_tc, symbol_set_tc): Likewise.
4235 (symbol_begin): Initialize local_hash.
4236 (print_symbol_value_1): Handle local symbols.
4237 (symbol_print_statistics): Print local symbol statistics.
4238 * symbols.h: Include "struc-symbol.h" if not BFD_ASSEMBLER.
4239 Declare new symbols.c functions. Move many declarations here from
4241 (SYMBOLS_NEED_BACKPOINTERS): Define if needed.
4242 * struc-symbol.h (SYMBOLS_NEED_BACKPOINTERS): Don't set.
4243 (struct symbol): Move bsym to make it clearly the first field.
4244 Remove TARGET_SYMBOL_FIELDS.
4245 (symbolS): Don't typedef.
4246 (struct broken_word): Remove.
4247 (N_TYPE_seg, seg_N_TYPE): Move to symbol.h.
4248 (SEGMENT_TO_SYMBOL_TYPE, N_REGISTER): Likewise.
4249 (symbol_clear_list_pointers): Likewise.
4250 (symbol_insert, symbol_remove): Likewise.
4251 (symbol_previous, symbol_append): Likewise.
4252 (verify_symbol_chain, verify_symbol_chain_2): Likewise.
4253 (struct local_symbol): Define.
4254 (local_symbol_converted_p, local_symbol_mark_converted): Define.
4255 (local_symbol_resolved_p, local_symbol_mark_resolved): Define.
4256 (local_symbol_get_frag, local_symbol_set_frag): Define.
4257 (local_symbol_get_real_symbol): Define.
4258 (local_symbol_set_real_symbol): Define.
4260 * write.c (write_object_file): Call resolve_local_symbol_values.
4261 * config/obj-ecoff.h (OBJ_SYMFIELD_TYPE): Define.
4262 (TARGET_SYMBOL_FIELDS): Don't define.
4263 * config/obj-elf.h (OBJ_SYMFIELD_TYPE): Add local field. If
4264 ECOFF_DEBUGGING, add ECOFF fields.
4265 (ELF_TARGET_SYMBOL_FIELDS, TARGET_SYMBOL_FIELDS): Don't define.
4266 * config/obj-multi.h (struct elf_obj_sy): Add local field. If
4267 ECOFF_DEBUGGING, add ECOFF fields.
4268 (ELF_TARGET_SYMBOL_FIELDS, TARGET_SYMBOL_FIELDS): Don't define.
4269 (ECOFF_DEBUG_TARGET_SYMBOL_FIELDS): Don't define.
4270 * config/tc-mcore.h: Don't include struc-symbol.h.
4271 (TARGET_SYMBOL_FIELDS): Don't define.
4272 (struct mcore_tc_sy): Define.
4273 (TC_SYMFIELD_TYPE): Define.
4274 * Many files: Use symbolS instead of struct symbol. Use new
4275 accessor functions rather than referring to symbolS fields
4278 * read.c (s_mri_common): Don't add in value of line_label.
4280 * config/tc-mips.c (md_apply_fix): Correct parenthesization when
4281 checking for SEC_LINK_ONCE.
4283 * config/tc-sh.h (sh_fix_adjustable): Declare.
4285 * app.c (input_buffer): New static variable.
4286 (app_push): Save saved_input in allocated buffer.
4287 (app_pop): Restored saved_input.
4288 (do_scrub_chars): Change get parameter to take char * and int as
4289 arguments. Change GET macro to pass input_buffer to get
4290 function. Don't save input into allocated buffer.
4291 * as.h (do_scrub_chars): Update declaration.
4292 * input-file.c (input_file_get): Change to take char * and int.
4293 Read data into passed in buffer. Remove static buffer.
4294 * read.c (scrub_from_string): Change to take char * and int. Copy
4295 data into passed in buffer.
4297 * hash.h: Neaten. Declare hash_traverse.
4298 * hash.c: Complete rewrite based on BFD hashing code.
4299 * gasp.c (chunksize): New variable.
4300 * macro.c (macro_expand_body): Call hash_jam with NULL rather than
4305 * config/tc-arm.c (md_apply_fix3): Add pipeline offset into reloc
4306 addend unless the target uses an old ABI.
4310 -Wchar-subscripts cleanup
4311 * listing.c (listing_newline): Use unsigned char variable, so
4312 calls to isascii,iscntrl are correct.
4313 * atof-generic.c (atof_generic): Cast arg to isdigit, et. al. with
4315 * ecoff.c (ecoff_directive_ent,ecoff_stab): Ditto.
4316 * config/obj-elf.c (obj_elf_vtable_inherit): Ditto.
4317 * config/tc-mips.c (mips_ip,mips16_ip): Ditto.
4318 (my_getSmallExpression,get_number,s_mips_ent): Ditto.
4322 * config/tc-m68k.c (m68k_ip): Check for disallowed index register
4324 (arch_coldfire_p): New #define.
4325 (m68k_ip, m68k_init_after_args): Use arch_coldfire_p.
4329 * config/tc-m68k.c (install_operand): Add places `n', `o'.
4331 * config/tc-m68k.c (m68k_ip): Add formats `E', `G', `H'.
4332 (install_operand): Add place `N'.
4333 (init_table): Add registers ACC, MACSR, MASK.
4335 * config/m68k-parse.h (m68k_register): Add ACC, MACSR, MASK.
4337 * config/tc-m68k.c: Change mcf5200 --> mcf.
4338 (archs): Add mcf5206e, mcf5307.
4339 (m68k_ip): Add format `u'.
4340 (install_operand): Add place `m', `M', `h'.
4341 (init_table): Add upper/lower registers.
4343 * config/m68k-parse.h (m68k_register): Add upper/lower registers.
4347 * config/tc-i960.c: Several minor changes to add ELF and
4348 BFD_ASSEMBLER support.
4349 * config/tc-i960.h: Likewise.
4350 * configure.in (i960-*-elf*): New target.
4351 * aclocal.m4, configure: Rebuild.
4355 * config/tc-i386.c (md_apply_fix3): Only do 1999-05-17 fx_pcrel
4356 reloc changes when defined(BFD_ASSEMBLER).
4360 * config/tc-i386.c (tc_gen_reloc): Remove F and MAP macros.
4362 * write.c (write_print_statistics): Output to file, not stderr.
4364 * expr.c (generic_bignum_to_int32,64): Prototype.
4366 * read.c (s_lcomm_internal, sizeof_sleb128, sizeof_uleb128,
4367 output_sleb128, output_uleb128, output_big_sleb128,
4368 output_big_uleb128, output_big_leb128): Prototype.
4369 (output_big_sleb128, output_big_uleb128): Make inline.
4370 (output_big_leb128): Remove inline
4373 * config/tc-i386.c (md_apply_fix3): Convert BFD_RELOC_16 with
4374 fx_pcrel set to BFD_RELOC_16_PCREL. Similarly for BFD_RELOC_8.
4375 Handle BFD_RELOC_16_PCREL and BFD_RELOC_8_PCREL. Return changed
4376 value for correct overflow check in write.c:fixup_segment.
4377 * write.c (fixup_segment): Move bitfield overflow checks to after
4378 the md_apply_fix call.
4379 * config/obj-coff.c (fixup_segment): Likewise.
4380 * doc/internals.texi (CPU backend): Mention md_apply_fix modifying
4381 valueT *val argument.
4385 * config/atof-ieee.c (gen_to_words): Correctly round a
4386 denormalized number. Fix off-by-one in range checking for
4387 exponent in a denormal.
4391 * config/tc-mcore.c (parse_reg): Accept 'sp' as a valid register
4396 * configure.in (i386-*-rtemself*, sh-*-rtemself*): New targets.
4400 * config/tc-i386.h (InvMem): New flag. Add to AnyMem.
4401 (ReverseRegRegmem): Remove.
4402 (ImmExt): New flag. Renumber some of the opcode_modifier bits.
4403 * config/tc-i386.c (md_assemble): Test for PIII SIMD and AMD
4404 3DNow! via ImmExt opcode_modifier. Remove ReverseRegRegmem
4408 * config/tc-i386.h (RegXMM): New for P/III.
4409 * config/tc-i386.c: Add support for P/III.
4413 * config/tc-ppc.c (md_parse_option): Recognize -mppc64bridge.
4414 (md_begin): Allow ppc32 insns in ppc64bridge mode.
4415 (ppc_insert_operand): Accept SIGNOPT in ppc64 mode.
4419 * config/tc-i386.c (i386_immediate): Skip whitespace before
4420 complaining about junk after expression.
4421 (i386_displacement): Likewise.
4425 * symbols.c (symbol_find_base): Use memcpy instead of strcpy.
4426 Don't copy before downcaseing.
4430 * tc-m68k.c: Include elf/m68k.h.
4431 (m68k_elf_final_processing): New routine.
4432 * tc-m68k.h (elf_tc_final_processing m68k_elf_final_processing):
4437 * config/tc-hppa.c (md_apply_fix): Handle 22 bit fmt insn like a
4442 * config/tc-mcore.c (mcore_s_section): Dump literals before
4447 * config/tc-mcore.c (md_apply_fix3): Insert reloc addend into insn
4452 * config/tc-fr30.h (TC_FIX_TYPE): Delete, cgen fields moved to write.h.
4453 (TC_INIT_FIX_DATA): Delete.
4454 * config/tc-m32r.h (TC_FIX_TYPE): Delete, cgen fields moved to write.h.
4455 (TC_INIT_FIX_DATA): Delete.
4456 * write.h (struct fix): New member fx_cgen, ifdef USING_CGEN.
4457 * write.c (fix_new_internal): Initialize fx_cgen member.
4458 * cgen.c (gas_cgen_record_fixup,gas_cgen_record_fixup_exp): Update.
4459 (gas_cgen_md_apply_fix3): Update.
4460 * config/tc-m32r.c (md_cgen_lookup_reloc): Update.
4461 (md_cgen_record_fixup_exp): Update.
4462 (FX_OPINFO_R_TYPE): Update.
4464 * frags.c (frag_var,frag_variant): Initialize fr_cgen here.
4465 * config/tc-fr30.h (TC_FRAG_INIT): Delete.
4466 * config/tc-m32r.h (TC_FRAG_INIT): Delete.
4467 * frags.h (struct frag): Make opindex, opinfo ints.
4469 * config/tc-fr30.c (FX_OPINFO_R_TYPE): Delete, unused.
4473 * aclocal.m4, configure: Updated for new version of libtool.
4477 * config/tc-mcore.c (md_apply_fix3): Renamed function from
4479 (md_apply_fix3): Do not fix up absolute relocations against
4482 * config/tc-mcore.h (MD_APPLY_FIX3): Define.
4486 * config/tc-mcore.c (md_pseudo_table): Add intercepts for section
4487 changes and data-in-text directives.
4488 (mcore_cons): New function: intercept cons() operations.
4489 (mcore_float_cons): New function: intercept float_cons()
4491 (mcore_stringer): New function: intercept stringer() operations.
4495 * obj.h (struct format_ops): Change generate_asm_lineno field to
4497 * config/obj-ecoff.h (OBJ_GENERATE_ASM_LINENO): Don't define.
4499 * config/tc-alpha.c (find_opcode_match): Add default case to
4501 (find_macro_match): Likewise.
4502 (load_expression): Parenthesize && within ||.
4504 * config/tc-alpha.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
4508 * config/tc-mcore.c (md_pseudo_table): Add overrides for .bss
4509 .text .data .section pseudo ops.
4510 (mcore_s_section): New function. Dump lits before changing secs.
4511 (mcore_s_text): New function. Dump lits before changing secs.
4512 (mcore_s_data): New function. Dump lits before changing secs.
4516 * config/tc-mips.c (mips_32bitmode): New.
4517 (md_begin): Set mips_32bitmode if needed.
4518 (mips_elf_final_processing): Don't set EF_MIPS_ARCH.
4519 Set EF_MIPS_32BITMODE.
4523 * config/obj-coff.c (c_section_symbol): Fix typo in previous
4528 * config/tc-mcore.h (LOCAL_LABELS_FB): Define to 1.
4532 * tc-hppa.c (pa_get_absolute_exression): Try to handle "5 %r3"
4533 expressions correctly.
4538 * config/tc-mips.c (mips_elf_final_processing): Set EF_MIPS_ARCH.
4542 * tc-hppa.c (pa_ip, case '3'): New case for PA2.0 fmpyfadd
4543 and fmpynfadd instructions.
4547 * as.h (environ): Declare it, if needed.
4548 * as.c (dump_statistics): Don't declare environ.
4549 * configure.in (environ): Detect declaration.
4550 * configure, config.in: Rebuild
4552 * config/tc-i386.c (i386_immediate): Accept @GOT relocations.
4553 (i386_displacement): Allocate enough space for replacement buffer.
4554 Clean up replacement buffer initialization.
4558 * subsegs.c (section_symbol): Don't create a new symbol if one
4559 already exists; instead, use the existing one, but set its segment
4560 and frag data if it hasn't already been defined.
4561 * config/obj-coff.c (c_section_symbol): Likewise.
4565 * tc-alpha.c (load_expression): Call as_bad instead of abort.
4569 * config/tc-mcore.c: New File: Support routines for MCore
4571 * config/tc-mcore.h: New File: Definitions for MCore assembler.
4572 * config/obj-coff.c: Add support for mcore-pe target.
4574 * Makefile.am: Add support for MCore targets.
4575 * Makefile.in: Regenerate.
4576 * configure.in: Add support for MCore targets.
4577 * configure: Regenerate.
4579 * doc/all.texi: Set MCORE.
4580 * doc/as.texinfo: Document MCore specific command line options.
4582 * write.h: Prevent multiple inclusion.
4586 * asintl.h (LC_MESSAGES): Never define.
4587 * as.c (main): Don't pass LC_MESSAGES to setlocale if the system
4589 * gasp.c (main): Don't pass LC_MESSAGES to setlocale if the system
4592 * Makefile.am (m68k-parse.c): If configuring in the source
4593 directory, copy m68k-parse.y into the local directory before
4594 running ylwrap, to remove spurious differences when generating
4596 * Makefile.in: Rebuild.
4598 * config/tc-sparc.h (md_do_align): Just allocate the number of
4599 bytes necessary, rather than always allocating 1024.
4603 * listing.c (listing_newline): Add cast to avoid warning.
4604 * read.c (generate_lineno_debug): Add cases to switch. Reindent.
4605 * config/tc-i386.c (i386_scale): Add return value.
4606 (build_displacement_string): Remove unused local temp_disp2.
4607 (i386_intel_memory_operand): Add parentheses to avoid warning.
4608 (i386_intel_operand): Remove unused local end_of_operand_string.
4609 (i386_operand): Remove unused local operand_modifier.
4610 (i386_operand): Add parens to avoid warning.
4614 * configure.in: Add mips*-*-vxworks* target; have it define
4616 * configure, config.in: Rebuild.
4620 * configure.in (emulations): Add support for arm-epoc-pe.
4621 * configure: Regenerate.
4622 * config/te-epoc-pe.h: New file. Define macros specific to
4624 * config/tc-arm.h: Select epoc-pe-arm target format if configured
4625 for arm-epoc-pe target.
4629 * tc-mips.c (md_apply_fix): Adjust value for linkonce sections.
4633 * tc-hppa.c (pa_parse_nonneg_cmpsub_cmpltr): Clean up code to
4634 detect ",n" without a condition.
4635 (pa_parse_neg_cmpsub_cmpltr): Likewise.
4640 * tc-hppa.c (pa_ip, case '~'): The condition for a branch on bit
4641 instruction is encoded with one bit.
4646 * doc/internals.texi (CPU backend): Mention that
4647 line_separator_chars should not include newline. From thi
4652 * config/tc-fr30.c (md_begin): Update call to fr30_cgen_cpu_open.
4653 * config/tc-m32r.c (md_begin): Update call to m32r_cgen_cpu_open.
4657 * tc-alpha.c (md_assemble): Allow '6' in an opcode.
4661 * tc-hppa.c (pa_ip, case 'a'): Do not call pa_parse_..._cmpsub_cmpltr.
4666 * tc-hppa.c (pa_ip, case 'd'): Do not allow ",n".
4670 * app.c (do_scrub_begin): Change '-' back to a symbol char
4671 so we can use multiple opcodes on a line again.
4673 * config/tc-d30v.c: By default, warn if a symbol has
4674 the same name as a register. Plus some minor
4675 updates from the branch.
4679 * config/tc-d30v.c (md_apply_fix3): Handle BFD_RELOC_8,
4680 BFD_RELOC_16 and BFD_RELOC_64.
4684 * expr.c (expr): Add missing else.
4688 * config/tc-arm.c (md_apply_fix3): Improve error message.
4692 * Makefile.am (CPU_TYPES): Add fr30.
4693 (cgen.o): Add $(CGEN_CPU_PREFIX)-desc.h dependency.
4694 (fr30,m32r dependencies): Update.
4695 * Makefile.in: Rebuild.
4697 * cgen.c (gas_cgen_record_fixup): Update use of operand->type.
4698 (gas_cgen_record_fixup_exp): Ditto.
4699 (gas_cgen_finish_insn): Call cgen_operand_lookup_by_num.
4700 (gas_cgen_md_apply_fix3): Ditto. Update call to set_vma_operand.
4701 * config/tc-fr30.c (md_begin): Update call to fr30_cgen_cpu_open.
4702 (md_cgen_lookup_reloc): Update use of operand->type.
4703 * config/tc-m32r.c (md_begin): Update call to fr30_cgen_cpu_open.
4704 (md_convert_frag): Call cgen_operand_lookup_by_num.
4705 (md_cgen_lookup_reloc): Update use of operand->type.
4706 (m32r_cgen_record_fixup_exp): Ditto.
4710 * config/tc-mips.c (md_show_usage): Fix message.
4714 * doc/c-arm.texi (ARM Syntax): Document new command line switches
4715 and LDR reg,=<expr> instruction.
4717 * config/tc-arm.c: Add support for -mcpu=arm810, -mcpu=arm9 and
4722 * doc/c-arm.texi (ARM-Chars): Fix typo in use of '@'.
4726 This patch was created by: Scott Bambrough
4730 Special cased '@' character. The '@' character is used as the
4731 ARM assembler comment character, as a special character
4732 and in ELF .symver pseudo-op's, and as a special character in
4733 .type and .section pseudo-ops.
4734 (symver_pseudo): New static variable.
4735 (symver_state): New static variable.
4736 (struct app_save): Add field 'symver_state'.
4737 (app_push): Save global symver_state int struct app_save.
4738 (app_pop): Restore global symver_state from struct app_save.
4739 (do_scrub_chars): Special case handling of '@' character in
4742 * configure.in: Modified to recognize armv* uname syntax from ARM
4744 * configure: Regenerated.
4746 * config/obj-elf.c (obj_elf_section): Allow '%' as well as '@' as
4747 a prefix to the section's type.
4748 (obj_elf_type): Allow '%' as well as '@' and '#' as prefixes to
4749 the type's typename.
4751 * config/tc-arm.h: Add support for PIC generation:
4752 (pic_code): New boolean.
4753 (obj_relocate_extern): Define.
4754 (TC_RELOC_RTSYM_LOC_FIXUP): Define
4755 (TC_CONS_FIX_NEW): Define.
4756 (tc_fix_adjustable): Define.
4757 (GLOBAL_OFFSET_TABLE_NAME): Define.
4759 * config/tc-arm.c: Add support for PIC generation:
4760 (line_seperator_chars): Allow ';' as a seperator for Linux.
4761 (is_immediate_prefix): New macro.
4762 (arm_parse_reloc): New function.
4763 (s_arm_elf_cons): New function.
4764 (do_branch): Special case for BFD_RELOC_ARM_PLT32.
4765 (md_undefined_symbol): Special case handling for the Global Offset
4767 (md_apply_fix3): Handle PIC relocs.
4768 (tc_gen_reloc): Handle PIC relocs.
4769 (md_parse_option): Add support for '-k' command line switch to
4770 enable PIC generation.
4771 (cons_fix_new_arm): New function.
4772 (s_arm_elf_cons): New function.
4776 * configure.in: Add comments for uses of AC_DEFINE.
4777 * acinclude.m4: Likewise.
4778 * acconfig.h: Remove.
4779 * aclocal.m4: Rebuild.
4780 * configure: Rebuild.
4781 * Makefile.in: Rebuild.
4782 * config.in: Rebuild.
4786 * config/tc-mips.c (mips_ip: case 'o'): Fix assertion failure for
4787 non-constant offset from a base register.
4791 * config/tc-alpha.c (md_show_usage): Put \ before newline in
4796 * config/tc-alpha.c (cpu_types): Enable EV6 PALcode with -m21264.
4797 (emit_insn): Look for pc-relative and no-overflow specifiers on
4798 internal relocation types.
4802 * doc/c-mips.texi (MIPS Opts): Updated list of -mNNNN and
4805 * config/tc-mips.c: Remove all the mips_NNNN variables; just use
4807 (mips_4650, mips_4010, mips_4100): Variables removed.
4808 (hilo_interlocks, gpr_interlocks, append_insn, macro_build, macro,
4809 macro2, mips16_macro, mips_ip): Test mips_cpu, not the mips_NNNN
4811 (md_begin): Don't bother initializing the mips_NNNN variables;
4812 mips_cpu is set, and that's good enough now.
4813 (md_parse_option): Have the -mNNNN options set mips_cpu instead of
4814 the mips_NNNN variable. The -no-mNNNN flags are now no-ops.
4815 (show): New function, to handle wrapping in the CPU lists.
4816 (md_show_usage): Update lists of -mcpu and -mNNNN switches.
4820 * config/tc-i386.c (i386_intel_operand): Ignore `SHORT' rather
4821 than treat as an immediate specifier.
4825 * config/tc-i386.c: Prototype many functions.
4826 (set_intel_syntax): Accept `prefix'/`noprefix' specifiers.
4827 (i386_immediate): Remove unused second argument.
4828 (i386_intel_operand): Fix i386_is_reg typo.
4829 (i386_operand): Use allow_naked_reg.
4830 (output_invalid): Make operand int for K&R.
4834 * Makefile.am (EXTRA_as_new_SOURCES): Uncomment--fixed by automake
4836 * Makefile.in: Rebuild.
4840 * Makefile.am (DISTCLEANFILES): Change cgen-opc.h to cgen-desc.h.
4842 (EXTRA_as_new_SOURCES): Comment out.
4843 (.tcdep): <arch>-opc.h renamed to <arch>-desc.h.
4844 * Makefile.in: Rebuild.
4845 * doc/Makefile.in: Rebuild.
4846 * configure.in: Require autoconf 2.13. Redo using_cgen handling.
4847 Delete call to AM_CYGWIN32. Replace AM_EXEEXT with AC_EXEEXT.
4848 (AC_OUTPUT): <arch>-opc.h renamed to <arch>-desc.h.
4849 * configure: Rebuild.
4850 * aclocal.m4: Rebuild.
4851 * config.in: Rebuild.
4852 * cgen.c: Include cgen-desc.h, not cgen-opc.h.
4853 (*): CGEN_OPCODE_DESC renamed to CGEN_CPU_DESC.
4854 (gas_cgen_cpu_desc): Renamed from gas_cgen_opcode_desc.
4855 CGEN_INSN_ATTR renamed to CGEN_INSN_ATTR_VALUE.
4856 CGEN_OPERAND_ATTR renamed to CGEN_OPERAND_ATTR_VALUE.
4857 (gas_cgen_record_fixup): Remove unnecessary != 0 test.
4858 (gas_cgen_record_fixup_exp): Ditto.
4859 (gas_cgen_finish_insn): Ditto. Refer to operand table via cpu
4860 descriptor, not global variable.
4861 (gas_cgen_md_apply_fix3): Refer to operand_table via cpu
4862 descriptor, not global variable. Refer to insert_operand handler
4863 via cpu descriptor, not global function.
4864 * cgen.h (*): CGEN_OPCODE_DESC renamed to CGEN_CPU_DESC.
4865 * config/tc-fr30.c: Include opcodes/fr30-desc.h.
4866 (*): gas_cgen_opcode_desc renamed to gas_cgen_cpu_desc.
4867 CGEN_INSN_ATTR renamed to CGEN_INSN_ATTR_VALUE.
4868 Update call to CGEN_OPERAND_TYPE,CGEN_INSN_OPERANDS.
4869 * config/tc-m32r.c: Ditto.
4870 (assemble_two_insns): Update calls to cgen_lookup_get_insn_operands.
4871 (md_assemble): Ditto.
4872 (md_convert_frag): Update call to CGEN_OPERAND_ENTRY.
4876 * config/tc-arm.c (md_apply_fix3): Fix handling of label1 - label2
4877 relocations for ELF targets.
4881 * configure.in: Add support for StrongARM target.
4882 * configure: Regenerate.
4886 * config/tc-arm.h: Tidy OBJ_ELF and OBJ_COFF definitions.
4888 * config/tc-arm.c (md_apply_fix3): Fix BFD_RELOC_ARM_PCREL_BRANCH
4893 * config/tc-alpha.c (md_show_usage): Document pca56 and ev6 options.
4897 * config/tc-i386.h (LONG_DOUBLE_MNEM_SUFFIX): Define.
4898 (INTEL_DWORD_MNEM_SUFFIX): Define.
4901 (DWORD_PTR): Define.
4902 (XWORD_PTR): Define.
4904 (OFFSET_FLAT): Define.
4906 (NONE_FOUND): Define.
4909 * config/tc-i386.c (set_intel_syntax): New routine.
4910 (intel_syntax): Declare.
4911 (allow_naked_reg): Declare.
4912 (md_pseudo_table): Support .intel_syntax and .att_syntax.
4913 (intel_float_operand): New routine.
4914 (md_assemble): Handle INTEL_DWORD_MNEM_SUFFIX.
4915 Handle brackets as well as parens. Call i386_intel_operand for
4916 intel syntax. Reverse operands if appropriate. Handle new
4917 suffixes. Handle movzx and movsx.
4918 (i386_is_reg): New routine.
4919 (i386_immediate): New routine.
4920 (i386_scale): New routine.
4921 (i386_displacement): New routine.
4922 (i386_operand_modifier): New routine.
4923 (build_displacement_string): New routine.
4924 (i386_parse_seg): New routine.
4925 (i386_intel_memory_operand): New routine.
4926 (i386_intel_operand): New routine.
4927 (i386_operand): Call i386_displacement, i386_immediate,
4928 i386_scale, etc. instead of handling inline.
4929 (parse_register): Handle registers without prefix.
4933 * configure: Regenerate.
4934 * configure.in (arm-*-oabi): New.
4935 (thumb-*-oabi): New.
4936 * config/tc-arm.c (target_oabi): Declare.
4937 (md_apply_fix3): Support REL relocs.
4938 (md_parse_option): Handle -oabi.
4939 (elf32_arm_target_format): New routine.
4940 (md_longopts): Add OPTION_OABI.
4941 * config/tc-arm.h: Redefine TARGET_FORMAT.
4946 * write.c (write_relocs): Handle out of range error.
4948 * config/tc-fr30.c (fr30_fix_adjustable): New function.
4949 (fr30_force_relocation): Default to 0.
4951 * config/tc-fr30.h (obj_fix_adjustable): Define.
4952 (TC_FORCE_RELOCATION): Define.
4954 * cgen.c (gas_cgen_md_apply_fix3): Do not apply fixes to VTABLE
4959 * config/tc-d30v.c (write_2_short): Do not generate a sequential
4960 merge of two instructions if the left instruciton kills the right.
4964 * Makefile.in: Regenerate.
4965 * configure.in: Redo test for using cgen.
4966 * configure: Regenerate.
4970 * config/obj-coff.h (obj_adjust_symtab): Prevent accidental
4971 redefinition of this macro.
4975 * config/tc-mips.c (mips_frob_file): Disable "Unmatched %hi reloc"
4980 * config/tc-mips.c (append_insn): For mips16, insert a nop between
4981 a read of HI or LO and an immediatly following branch.
4985 * config/tc-mips.c (md_begin): Another correction to the setting of
4990 * config/tc-mips.c (md_begin): Correct type-o in setting of
4995 * config/tc-m32r.c (md_assemble): Emit a NOP after a relaxable 16
4996 bit insn when optimizing, so that parallelised instructions will
4997 start on a 32 bit boundary.
5001 * config/tc-mips.c (mips_eabi64): New.
5002 (md_begin): Set mips_eabi64.
5003 (mips_elf_final_processing): Use it.
5007 * config/tc-mips.c (mips_elf_final_processing):
5008 Correct setting of ABI in e_flags.
5012 * config/tc-fr30.c (md_assemble): Warn about invalid instructions
5017 * config/tc-mips.c (md_begin,md_parse_option): Handle vr4111.
5021 * cgen.c (gas_cgen_md_apply_fix3): Mark as an error, rather than a
5022 warning, values that don't fit in the field.
5026 * config/tc-mips.c (mips_abi_string): New.
5027 (md_parse_option,md_longopts): Add mabi.
5028 (mips_elf_final_processing): Set e_flags based on mabi flag.
5032 * config/tc-mips.c (md_parse_option): Handle vr4111.
5036 * config/tc-h8300.c (build_bytes): Change message given if the
5037 instruction requires H8/300H mode and we're not in Hmode, to
5038 suggest that it may be the operand modes that are the problem, not
5039 necessarily the opcode.
5043 * config/tc-fr30.c: Add line separator character.
5047 * configure.in (hppa-*-mpeix*): New target.
5048 * config/obj-som.h (obj_som_compiler): Declare.
5049 * config/obj-som.c (compiler_seen): New static variable.
5050 (obj_som_compiler): New function.
5051 * config/tc-hppa.c: Update tc_data uses for change to bfd/som.h.
5052 (md_pseudo_table): Add "compiler" if OBJ_SOM.
5053 (pa_type_args): Set hppa_priv_level.
5054 (pa_compiler): New static function if OBJ_SOM.
5055 * configure: Rebuild.
5059 * read.c (output_leb128): Don't mark as inline.
5063 * config/tc-ppc.c (ppc_vbyte): Prototype and new function for
5064 AIX .vbyte unaligned data support.
5065 (md_pseudo_table): Add 'vbyte' to list of valid pseudos.
5066 (ppc_elf_validate_fix): Add eh_frame to list of ELF relocatable
5071 * config/tc-d30v.c (md_assemble, do_assemble): Improve erroneous
5076 * config/tc-arm.c (elf32_arm_force_relocation): Check for
5077 BFD_RELOC_ARM_PCREL_BRANCH.
5081 * configure.in: Define TARGET_BYTES_{BIG,LITTLE}_ENDIAN after
5082 checking the target type.
5083 (mips-dec-bsd*): Set endian to little.
5084 * configure: Rebuild.
5086 COFF weak symbol support, based on patches from Mark Elbrecht
5088 * config/obj-coff.h (S_IS_WEAK): Define if not BFD_ASSEMBLER.
5089 * config/obj-coff.c (obj_coff_weak): New static function.
5090 (obj_coff_endef) [both versions]: Handle weak symbols.
5091 (coff_frob_symbol): Likewise.
5092 (yank_symbols): Likewise.
5093 (obj_pseudo_table): Add "weak".
5095 * configure.in (m68k-*-gnu*): New target. From Aymeric Vincent
5097 * aclocal.m4: Rebuild with current tools.
5098 * configure: Rebuild.
5100 * config/tc-alpha.c (emit_ldgp): Give an error message rather than
5101 an assertion failure for a case we can't handle when OBJ_ECOFF.
5103 * expr.c (operator): And with 0xff to avoid problems with signed
5108 * config/tc-fr30.c (md_cgen_lookup_reloc): Generate
5109 BFD_RELOC_FR30_48 instead of BFD_RELOC_FR30_32.
5113 * config/tc-fr30.c (md_cgen_lookup_reloc): Enable relocs for
5118 * config/tc-fr30.c (md_pcrel_from_section): Restore previous
5119 calculation of pcrel point.
5123 * config/tc-fr30.c (md_pcrel_from_section): Fix calculation of
5128 * config/tc-d10v.c (md_assemble): Make static 'etype' have file
5130 (d10v_cleanup): Only generate previous insn if a multiline insn is
5135 * config/tc-fr30.c (md_cgen_lookup_reloc): Add support for
5140 * config/tc-arm.c (md_parse_option): Add support for -marm7xxx and
5141 -marm6xxx command line switches.
5145 * Makefile.am (DEP): Use $(srcdir)/../mkdep.
5146 (itbl-ops.o): Delete duplicate dependencies.
5147 Rebuild dependencies.
5148 Add fr30 dependencies.
5149 * Makefile.in: Rebuild.
5153 * config/tc-fr30.c (md_cgen_lookup_reloc): Updated to match latest
5155 * listing.c: Ignore line terminator characters found inside
5160 * po/gas.pot: Regenerated.
5164 * config/tc-fr30.c (fr30_is_colon_insn): New name for
5165 fr30_is_label_start(). Also checks for delay slot insns.
5167 * config/tc-fr30.c (fr30_is_label_start): New function: Handle
5168 FR30 instructions which contain a colon in the mnemonic.
5170 * config/tc-fr30.h (TC_START_LABEL): Define this macro.
5174 * config/tc-fr30.c: Removed currently superflous code.
5178 * config/tc-fr30.h: New file.
5179 * config/tc-fr30.c: Tweaking so that it will compile.
5183 * config/tc-d10v.h (obj_fix_adjustable): Define.
5184 (TC_FORCE_RELOCATION): Define.
5185 (d10v_force_relocation): Declare.
5186 * config/tc-d10v.c (tc_gen_reloc): Handle Vtable relocs.
5187 (md_apply_fix3): Handle Vtable relocs.
5188 (d10v_fix_adjustable): New.
5189 (d10v_force_relocation): New.
5193 * config/tc-d30v.c: Change default behaviour to ignore potential
5194 conflicts between register name and symbol names.
5198 * configure.in: Add fr30-*-*.
5199 * config/tc-fr30.c: New file.
5200 * Makefile.in: Regenerated.
5201 * config.in: Regenerated.
5202 * configure: Regenerated.
5203 * doc/Makefile.in: Regenerated.
5204 * po/gas.pot: Regenerated.
5208 * config/tc-m32r.c (assemble_two_insns): Ensure both insns
5213 * app.c (do_scrub_begin): Set characters above 127 to be symbol
5215 (do_scrub_chars): Add some casts to unsigned char to avoid
5216 unwanted sign extension.
5217 * read.c (lex_type): Set characters about 127 to be symbol
5219 * config/tc-i386.c (md_begin): Set identifier_chars and
5220 operand_chars for values above 127.
5224 * configure.in: detect cygwin* instead of cygwin32*
5225 * configure: regenerate
5229 * listing.c: Add support for producing a listing from piped
5234 * config/tc-mips.c (hilo_interlocks): Remove mips_3900.
5235 (append_insn): Account for the tx39's multiply behavior.
5239 * config/tc-m32r.c (assemble_two_insns): Rename assemble_two_insns
5240 from assemble_parallel_insns. Add support for '->' to indicate
5241 explicitly serializing the instructions.
5242 (md_assemble): Ditto.
5246 * config/tc-sh.c (sh_fix_adjustable): Adjust EXTERN and
5251 * cgen.c (gas_cgen_md_apply_fix3): Revert last change.
5255 * config/tc-mips.c : support frame and regmask/fregmask when
5256 MIPS_STABS_ELF is specified.
5260 * config/tc-sh.c (sh_fix_adjustable): Only include if OBJ_ELF.
5261 (md_apply_fix): Don't return 1 for VTABLE relocs.
5262 * config/tc-sh.h (obj_fix_adjustable): Define only if OBJ_ELF.
5266 * doc/c-i386.texi: Replace occurences of "opcode" with
5267 "instruction mnemonic", "instruction", or "mnemonic" when
5268 referring to the name of an instruction. Use "opcode" when
5269 referring to the sequence of machine bytes.
5271 * config/tc-i386.c (opcode_chars): Rename to mnemonic_chars.
5272 (is_opcode_char): Rename to is_mnemonic_char.
5273 (md_assemble and i386_operand): Correct error messages from
5274 "opcode" to "instruction mnemonic"
5275 Rename throughout opcode[] -> mnemonic[], opp -> mnem_p,
5276 MAX_OPCODE_SIZE -> MAX_MNEM_SIZE,
5277 DWORD_OPCODE_SUFFIX -> DWORD_MNEM_SUFFIX,
5278 WORD_OPCODE_SUFFIX -> WORD_MNEM_SUFFIX,
5279 BYTE_OPCODE_SUFFIX -> BYTE_MNEM_SUFFIX,
5280 SHORT_OPCODE_SUFFIX -> SHORT_MNEM_SUFFIX
5281 LONG_OPCODE_SUFFIX -> LONG_MNEM_SUFFIX
5283 * config/tc-i386.h (*_MNEM_SUFFIX): Rename from *_OPCODE_SUFFIX.
5285 * config/tc-i386.c (i386_operand): Check for garbage after
5290 * config/tc-i386.c (md_apply_fix3): Change handling of PCREL reloc
5291 for BFD_ASSEMBLER to only change value when COFF if TE_PE.
5295 * config/tc-sh.h (obj_fix_adjustable): Define.
5296 * config/tc-sh.c (sh_force_relocation): Handle VT relocs.
5297 (md_apply_fix): Likewise.
5298 (tc_gen_reloc): Likewise.
5299 (sh_fix_adjustable): New.
5303 * cgen.c (gas_cgen_finish_insn): Update handling of CGEN_INT_INSN_P.
5304 * cgen.h (gas_cgen_finish_insn): Update prototype.
5305 * config/tc-m32r.c (m32r_insn): CGEN_INT_INSN -> CGEN_INT_INSN_P.
5306 cgen_insn_t -> CGEN_INSN_INT.
5307 (make_parallel): Update handling of CGEN_INT_INSN_P.
5308 (assemble_parallel_insn): Ditto.
5309 (target_make_parallel): New function.
5310 (md_assemble): Use it.
5314 * config/tc-m32r.c (m32r_force_relocation): Fix typo.
5318 * config/tc-sh.c (md_assemble): Make sure the entire opcode is
5319 converted into lower case.
5323 * cgen.c (gas_cgen_md_apply_fix3): Handle VTABLE relocs.
5324 (gas_cgen_tc_gen_reloc): Likewise.
5325 * config/tc-m32r.h (obj_fix_adjustable): Define.
5326 * config/tc-m32r.c (m32r_fix_adjustable): New.
5327 (m32r_force_relocation): Handle VTABLE relocs.
5331 * doc/c-arm.texi (ARM Directives): Document .ltorn directive.
5335 * config/tc-m32r.c (assemble_parallel_insn): Convert second opcode
5336 to lower case before parsing.
5338 * config/tc-d30v.c (parallel_ok): Ignore conflicts when explicitly
5339 parallel insns modift buts in the PSW as a side effect.
5343 * config/tc-d30v.c (find_format): Test for missing flag and
5346 (md_apply_fix3): Fix error messages to avoid
5347 assumption about presence of a symbol.
5349 (parallel_ok): Disallow parallel instructions that both modify the
5352 (find_format): Generate a warning if an odd numbered register is
5353 used as the first register in a mutli-register instruction.
5357 * config/tc-d30v.c (md_apply_fix3): Do not assume that bad
5358 relocations are always associated with a symbol.
5362 * tc-sparc.h (TC_FORCE_RELOCATION): Define.
5363 (elf32_sparc_force_relocation): Declare.
5364 * tc-sparc.c (md_apply_fix3): Handle vtable relocs.
5365 (tc_gen_reloc): Handle vtable relocs.
5366 (elf32_sparc_force_relocation): New.
5370 * symbols.c (S_IS_FUNCTION): New.
5371 * config/tc-v850.h (obj_fix_adjustable): Define.
5372 (TC_FORCE_RELOCATION): Define.
5373 (v850_force_relocation): Declare.
5374 * config/tc-v850.c (tc_gen_reloc): Use offset instead
5375 of fx_addnumber for VTABLE reloc addends.
5376 (md_apply_fix3): Handle VTABLE relocs.
5377 (v850_fix_adjustable): New.
5378 (v850_force_relocation): New.
5382 * tc-hppa.c (fp_operand_format): Add some additional formats.
5383 (pa_ip): Do not automatically promote into pa2.0 mode.
5384 (pa_level): Handle ".level 2.0".
5388 * config/tc-i386.c (md_assemble): Handle AMD_3DNOW_OPCODE.
5389 * config/tc-i386.h (template.extension_opcode): Change to
5390 unsigned int to allow full range of 8-bit opcode suffixes.
5391 (None): Redefine as 0xffff.
5394 * doc/c-i386.texi (i386-SIMD): New section.
5398 * read.c (discard_rest_of_line): New function.
5399 * read.h: Declare it.
5400 * config/tc-alpha.c (s_alpha_mask, s_alpha_frame): Use it.
5404 * config/tc-d10v.c (find_symbol_matching_register): New function.
5405 (find_opcode): Cope with the case where a register name matches
5410 * config/tc-v850.c (md_pcrel_from): Rename to
5411 v850_pcrel_from_section.
5412 (v850_pcrel_from_section): Do not resolves symbols in other
5415 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
5419 * config/tc-d10v.c (find_opcode): Generate an error if a register
5420 is supplied for an operand that should not be a register.
5424 * config/tc-d30v.c (write_2_short): But do allow delayed branch
5425 instructions to have another instruction in the right bin.
5429 * config/tc-d30v.c (write_2_short): Do not allow instructions in
5430 the right container if the left container holds a branch
5435 * config/tc-d30v.c (reg_name_search): Only warn if a name matches
5436 both a register name and symbol name.
5437 (find_format): Allow correct parsing of MVTSYS and MVFSYS insns.
5441 * config/tc-d30v.c (write_2_short): Implement EITHER_BUT_PREFER_MU
5442 execution unit class.
5444 (reg_name_search): If a name matches a register and a symbol,
5445 prefer the register.
5446 (find_format): Disallow flag registers when a general purpose
5447 register is required.
5448 If a number is required, but a register has been given, check to
5449 see if a symbol with the same name as the register exists, and if
5450 so, use that symbol.
5454 * config/obj-elf.h (ECOFF_DEBUGGING): Add missing parens.
5458 * config/tc-d30v.c (find_format): Do not accept flag registers as
5459 general purpose registers.
5460 (find_format): If an immediate value is expected at a given place
5461 in a format, but a register name has been provided instead, check
5462 to see if that register name matches the name of a predefined
5463 symbol and if it does, then use the symbol instead.
5464 (reg_name_search): If a register name matches a symbol name,
5465 prefer the register name to the symbol name.
5469 * config/tc-m32r.c (m32r_do_align): After inserting NOPs, reset
5470 the previous insn to empty.
5474 * config/tc-ppc.c (md_apply_fix3): Do not break string into two
5475 pieces, forcing the use of an ANSI compiler.
5479 * config/tc-m68k.h (TC_FORCE_RELOCATION): New macro. Force vtable
5481 * config/tc-m68k.c (md_apply_fix_2): Do nothing for vtable relocs.
5485 * config/obj-elf.c (obj_elf_vtable_inherit): Handle arm
5487 (obj_elf_vtable_entry): Likewise.
5488 * config/tc-arm.h: Define TC_FORCE_RELOCATION for OBJ_ELF.
5489 * config/tc-arm.c (md_apply_fix3): Handle VTABLE relocations.
5490 (tc_gen_reloc): Likewise.
5491 (arm_fix_adjustable): Likewise.
5492 (elf32_arm_force_relocation): New.
5493 (armelf_frob_symbol): Remove coff-style symbol support.
5497 * config/tc-i386.c (i386_operand): Fix typo in last patch.
5501 * config/tc-arm.c (arm_adjust_symtab): Move #ifdef
5502 OBJ_COFF so that routine is defined for a.out format.
5506 * config/tc-i386.c (i386_operand): Detect non-segment registers
5507 used as segment prefixes.
5511 * ehopt.c (check_eh_frame): Check the size of the FDE, and don't
5512 optimize across FDE boundaries.
5514 * config/obj-coff.c (obj_coff_section): Preserve any link once
5515 flags when setting the section flags.
5519 * config/tc-arm.h (obj_adjust_symtab): Fixed typo.
5520 * config/tc-arm.c (armelf_adjust_symtab): Reformatted.
5524 * config/tc-sparc.c (in_signed_range): Sign extend 32-bit words
5529 * frags.c (frag_grow): Include the size of the frag struct in the
5532 * subsegs.c (subseg_set_rest): Adjust the seginfo frchain start
5533 if the new subseg comes before the old.
5537 * config/tc-sparc.c (sparc_ip): Allow all digits in an instruction
5538 to handle edge8 and edge16.
5542 * config/obj-elf.c (obj_elf_vtable_inherit): Print error message
5543 before we clobber the symbol involved.
5547 * config/tc-arm.c: Remove OBJ_ELF definitions for
5548 S_GET_STORAGE_CLASS and S_SET_STORAGE_CLASS. Only
5549 use arm_adjust_symtab for OBJ_COFF.
5550 (armelf_adjust_symtab): New Routine.
5551 * config/tc-arm.h: Define obj_adjust_symtab to
5552 armelf_adjust_symtab for OBJ_ELF.
5556 * configure.in: Make all i386-elf targets use bfd_gas.
5557 * config/tc-i386.c (tc_i386_force_relocation): New.
5558 (tc_i386_fix_adjustable): Don't fix vtable relocs.
5559 (md_apply_fix3): Likewise.
5560 (tc_gen_reloc): Handle them.
5561 * config/tc-i386.h (TC_FORCE_RELOCATION): Always define, calling
5562 tc_i386_force_relocation.
5566 * config/tc-arm.c (md_show_usage): Improve formatting of --help output.
5570 * config/tc-d30v.c (md_assemble): Copy previous opcode over
5571 current opcode after writing the first insturction of a reverse
5576 * read.h (generate_lineno_debug): Add prototype.
5577 * read.c (generate_lineno_debug): Make non-static.
5581 * config/tc-i386.c (md_assemble): Only warn for address/data size
5586 * config/tc-arm.c (arm_fix_adjustable): Do not adjust relocations
5587 against Thumb function names, as the linker needs this information.
5591 * expr.c (operand): Check also that there is no advance in operand
5592 after atof_generic in order to decide "is it label 0f or floating
5597 * config/tc-m32r.c: Replace double dash prefix to M32R specific
5598 command line options with a single dash.
5599 * doc/c-m32r.texi: Replace double dash prefix with a single dash.
5603 * tc-arm.h: Define obj_fix_adjustable for OBJ_ELF.
5604 * tc-arm.c (arm_fix_adjustable): New routine.
5608 * read.c (s_align, s_comm, s_mri_common, s_fail, s_globl, s_space,
5609 s_float_space, s_struct, cons_worker): Move ignore_rest_of_line or
5610 demand_empty_rest_of_line before mri_comment_end.
5611 (equals): Check garbage after expression before
5612 mri_comment_end in MRI mode.
5616 * config/tc-mips.c (macro): Correct M_SGE_I/M_SGEUI_I case for a
5617 small immediate constant to use the constant itself rather than
5622 * config/tc-hppa.c (pa_enter): Call as_bad rather than abort.
5623 (pa_leave): Likewise.
5627 * config/tc-i386.c (md_assemble): Emit a warning for stand-alone
5629 (i386_operand): Fix an error message.
5633 * doc/c-arm.texi (ARM Directives): Document .req directive.
5635 * config/tc-arm.c (reg_required_here): Display erroneous string if
5636 the register name could not be decoded.
5637 Do not set inst.instruction if the sift is -1.
5641 * config/tc-alpha.c (tc_gen_reloc): Bias WEAK symbols just as
5646 * config/tc-d30v.c (d30v_align): Always perform alignment request,
5647 even if it is belived to be unnecessary.
5651 config/tc-i386.c (i386_operand): Size immediate constants by
5652 suffix (erroneously removed as part of July 7 change).
5656 * config/obj-elf.h: Check for redefinition of obj_frob_symbol.
5657 * config/tc-arm.c: Define S_GET_STORAGE_CLASS and S_SET_STORAGE_CLASS.
5658 (armelf_frob_symbol): New Routine.
5659 * config/tc-arm.h: Define obj_frob_symbol if OBJ_ELF.
5663 * config/tc-alpha.c (alpha_fix_adjustable): Don't adjust weak syms.
5667 * config/tc-arm.c (md_begin): Set BFD private flags depending upon
5668 command line switches passed to assembler.
5672 * cgen.h (GAS_CGEN_MAX_FIXUPS): GAS_ prepended, all uses updated.
5673 (gas_cgen_opcode_desc): Declare.
5674 (gas_cgen_parse_operand): Declare.
5675 (*): Prepend gas_ to gas specific fns to denote them as such.
5677 * cgen.c (gas_cgen_opcode_desc): New global
5678 (gas_cgen_init_parse): Renamed from cgen_asm_init_parse.
5679 (queue_fixup): Renamed from cgen_queue_fixup.
5680 (*): Prepend gas_ to gas specific fns to denote them as such.
5682 (gas_cgen_md_apply_fix3): Update call to insert_operand.
5683 (gas_cgen_finish_insn): Renamed from cgen_asm_finish_insn.
5684 * config/tc-m32r.c (md_begin): Remove use of CGEN_SYM.
5685 Open opcode table and initialize it.
5686 (make_parallel): Use gas_cgen_opcode_desc.
5687 (assemble_parallel_insn): Ditto. Remove use of CGEN_SYM.
5688 (md_assemble): Ditto.
5692 * as.h (debug_info_type): Add entries for unspecified and dwarf*.
5693 * ecoff.c (ecoff_generate_asm_lineno): Take no arguments; call
5694 as_where ourselves. Provide a stub for !ECOFF_DEBUGGING.
5695 * ecoff.h: Move ECOFF_DEBUGGING protection inside GAS_ECOFF_H.
5696 Move ecoff_generate_asm_lineno outside ECOFF_DEBUGGING protection.
5697 * read.c (generate_lineno_debug): Tidy ECOFF bits. Use
5698 DEBUG_UNSPECIFIED rather than DEBUG_NONE for initial test.
5699 * config/obj-elf.h (ECOFF_DEBUGGING) [TC_ALPHA]: Define to a variable.
5700 (SEPARATE_STAB_SECTIONS): Conditionalize on value of ECOFF_DEBUGGING.
5701 (INIT_STAB_SECTION): Likewise.
5702 (OBJ_PROCESS_STAB): Likewise.
5704 * config/tc-alpha.c (md_longopts): New options -mdebug/-no-mdebug.
5705 (md_parse_option): Watch for them.
5706 (alpha_cur_ent_sym, alpha_flag_mdebug): New variables.
5707 (md_begin): Kill neverdef code.
5708 (s_alpha_ent, s_alpha_end, s_alpha_mask, s_alpha_frame): New.
5709 (s_alpha_prologue): Watch alpha_cur_ent_sym.
5710 (s_alpha_coff_wrapper): New.
5711 (md_pseudo_table): Trap all ECOFF pseudos.
5715 Start of changes to remove mdebug section from mips*-elf
5716 Based on MIPS_STAB_ELF definition
5717 * acconfig.h: undef if not configured
5718 * config.in: undef if not configured
5719 * config/mips-elf.h: only set ECOFF debugging if not stabs-in-elf
5720 * config/tc-mips.c (s_ent): set BSF_FUNCTION
5721 * stabs.c (s_stab_generic): flush frag
5725 * configure.in: (arm-*-elf): Handle.
5726 (thumb-*-elf): Handle.
5727 * configure: Regenerate.
5728 * read.c (stringer): Fix typo in comment.
5729 * write.c (fixup_segment): Don't add symbol value to addend if
5731 * config/tc-arm.c (md_section_align): Don't align dwarf debug
5733 (tc_gen_reloc): Always set the reloc addend to fixp->fx_offset
5738 * config/tc-d30v.c ({cur,prev}_left_kills_right_p): New variables.
5739 (write_2_short): Emit warning if new flag is set.
5740 (do_assemble): Set flags if left instruction is one of special
5741 "right-instruction-killer" type.
5745 * config/tc-sparc.c (md_number_to_chars, cons_fix_new_sparc):
5746 Always output words in debug_info section as big endian.
5747 (sparc_target_format): Choose correct bfd target.
5748 (md_apply_fix3): Rename BFD_RELOC_SPARC_32LE to BFD_RELOC_SPARC_REV32.
5752 * config/tc-mn10300.c (md_assemble): Fix "errmsg" initialization
5753 to work with internationalization code. Issue an error when two
5754 operands match that are not allowed to match.
5758 * configure.in (install_tooldir): Allow target to specify whether
5759 it wants to be installed in $(tooldir)/bin.
5760 * configure: Regenerate.
5761 * Makefile.am (install-exec-local): Set install-exec-tooldir
5762 dependency via configure.
5763 * Makefile.in: Regenerate.
5767 * Makefile.am (install-exec-local): Split into two ...
5768 (install-exec-bindir,install-exec-tooldir): New rules.
5769 * Makefile.in: Regenerate.
5773 * Makefile.am (install-exec-local): Don't remove the file before
5774 checking whether $(bindir) == $(tooldir)/bin. From Maciej
5776 * Makefile.in: Rebuild.
5780 * cgen.c: Include libiberty.h.
5781 (cgen_md_apply_fix3): Update call to md_cgen_lookup_reloc.
5782 (cgen_tc_gen_reloc): Use xmalloc, not bfd_alloc.
5783 * cgen.h (cgen_md_apply_fix3,cgen_tc_gen_reloc): Declare.
5784 (md_cgen_lookup_reloc)): Declare.
5785 (md_cgen_record_fixup_exp): Declare.
5786 * config/tc-m32r.h (md_pcrel_from_section): Declare.
5787 (m32r_relax_frag): Declare.
5788 (cgen_md_apply_fix3): Decls moved to cgen.h.
5789 (cgen_record_fixup_exp,cgen_tc_gen_reloc): Ditto.
5790 (m32r_cgen_record_fixup_exp): Delete decl.
5791 * config/tc-m32r.c (m32r_cpu_desc): #if 0 out.
5792 (assemble_nop): Delete.
5793 (expand_debug_syms): Delete unused `exp'.
5794 (md_cgen_lookup_reloc): Renamed from CGEN_SYM (lookup_reloc).
5795 Add default case for -Wall.
5796 (m32r_cgen_record_fixup_exp): Add default case for -Wall.
5797 (md_atof): Delete unused wordP.
5801 * config/tc-mn10300.c (md_assemble): Make sure "errmsg" has a non-NULL
5806 * doc/as.texinfo: Add documentation for .end, .exitm, .fail,
5807 .ifc, .ifeqs, .ifge, .ifgt, .ifle, .iflt, .ifnc, .ifne, .ifnes,
5808 .print, .purgem, and .struct. Remove documentation for
5813 * cgen.c (cgen_md_apply_fix3): set_operand renamed to set_vma_operand.
5814 Update call to insert_operand.
5818 * config/tc-m32r.c (ms_show_usage): Formatting changes.
5822 * config/tc-i386.c (md_assemble): Don't get confused by trailing
5823 whitespace after a prefix operator.
5827 * configure.in (i386-*-beos{pe,elf,}*): Recognize.
5831 * config/tc-sparc.c (log2): New static function.
5832 (s_reserve): Use log2 to convert alignment before calling
5834 (s_common): Use log2 to convert alignment before calling
5835 record_alignment and frag_align.
5836 (sparc_cons_align): Use log2.
5840 * config/tc-sparc.c (s_reserve): Set symbol size if OBJ_ELF.
5841 (s_common): Likewise.
5843 * config/tc-sparc.c (sparc_handle_align): Reindent a bit. Correct
5844 initialization of waddr.
5845 (sparc_elf_final_processing): Add default case to switch.
5849 * doc/c-i386.texi: Fix a typo. Use the term 80-bit real rather
5850 than temporary real.
5854 * write.c (subsegs_finish): Don't align the segments if there were
5857 * config/obj-coff.c (c_symbol_merge): Correct number of bytes when
5858 copying aux information.
5860 * expr.c (make_expr_symbol): Catch attempts to turn an O_big
5861 expression into a symbol.
5865 * config/tc-i386.c (mode_from_disp_size): Change arg and return
5866 type to unsigned int.
5867 (md_assemble): Change type used to store offsets from unsigned
5869 (i386_operand): Switch error check to only call RESTORE_END_STRING
5870 once after parse_register.
5874 * config/tc-v850.c (md_show_usage): Changed format to match that
5877 * as.c (show_usage): Changed format to match that of gcc, ld, etc.
5881 * config/tc-m68k.c (tc_m68k_fix_adjustable): Don't adjust vtable
5883 (md_apply_fix_2): Force the symbol of the vtable reloc to be
5888 * doc/Makefile.am (MAINTAINERCLEANFILES): Define.
5889 * doc/Makefile.in: Rebuild.
5893 * config/tc-mips.c (mips_ip, case 'i' and 'j'): Mask off high bits
5894 for %lo expressions.
5895 (mips_ip, case 'u'): Move range check after code to mask
5896 off bits in %hi/%lo expressions. Mask off high bits for
5901 * doc/Makefile.am (gasver.texi): New target.
5902 (as.info, as.dvi): Depends upon gasver.texi.
5903 * doc/as.texinfo: Include gasver.texi. Mention version number on
5904 title page and in top node.
5905 * doc/Makefile.in: Rebuild.
5909 * listing.c (listing_listing): For EDICT_LIST, skip all lines up to
5910 but not including the line containing the edict.
5911 * listing.h (LISTING_EOF): New.
5912 * input-scrub.c (input_scrub_next_buffer): Call it.
5916 * config/tc-i386.c (i386_operand): Don't set the size of an
5917 immediate address based solely on the suffix and the mode.
5919 * config/tc-i386.c (md_assemble): Add assertion to make sure
5920 overlap2 does not set Imm.
5922 * config/tc-i386.c (space_chars): Remove. The scrubber converts
5923 sequences of whitespace to a single space.
5924 (is_space_chars): Just compare with space.
5925 (md_begin): Don't initialize space_chars.
5926 (md_assemble): Just skip a single whitespace character.
5927 (i386_operand): Rewrite base-index parsing to use new
5928 parse_register, and to skip white space. Skip white space in a
5929 number of other places too. Don't give error message if
5930 parse_register fails.
5931 (parse_register): Change reg_string parameter to be non-const.
5932 Add end_op parameter. Skip white space after the `%', and return
5933 end of register string. Give error message here rather than
5939 * expr.c (op_encoding): Make const.
5940 (expr_set_precedence): New function.
5941 (expr_begin): Don't set operator rankings, just call
5942 expr_set_precedence.
5943 * expr.h (expr_set_precedence): Declare.
5944 * read.c (s_mri): Call expr_set_precedence.
5948 * doc/as.texinfo (Statements): Remove paragraph discussing
5949 continuing lines with a backslash. This hasn't worked for years,
5954 * obj-vms.c: Add C++ support with ctors/dtors sections. Add weak
5956 (Ctors_Symbols, Dtors_Symbols): New symbol chains.
5957 (ps_CTORS, ps_DTORS): New section types.
5958 (vms_fixup_xtors_section): New function
5959 (Ctors_Psect, Dtors_Psect): Define.
5960 (IS_GXX_XTOR): Define
5961 (global_symbol_directory): Change check of gxx_bug_fixed to 0.
5962 Filter static constructors/destructors and add to
5963 Ctors_Symbols/Dtors_Symbols chain.
5964 (vms_write_object_file): Write Ctors_Symbols/Dtors_Symbols to
5965 appropriate section.
5967 * tc-alpha.h (TARGET_FORMAT): Rename "evax-alpha" to "vms-alpha".
5968 * makefile.vms: Merge vax/vms support.
5972 * config/obj-elf.c (obj_elf_vtable_inherit, obj_elf_vtable_entry): New.
5973 (elf_pseudo_table): Add them.
5974 * config/tc-mips.c (mips_force_relocation): Force vtable relocs.
5975 (md_apply_fix): Accept them.
5976 (mips_fix_adjustable): Don't adjust them.
5977 (tc_gen_reloc): Mung BFD_RELOC_VTABLE_ENTRY for Rel.
5978 * config/tc-ppc.c (md_apply_fix3): Accept vtable relocs.
5979 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Force vtable relocs.
5980 (tc_fix_adjustable): Don't adjust them.
5984 * Makefile.am (CGEN_CPU_PREFIX): New variable.
5986 * Makefile.in: Regenerate.
5987 * configure.in: AC_SUBST cgen_cpu_prefix.
5988 * configure: Regenerate.
5992 * config/tc-sh (COND_JUMP_DELAY, COND12_DELAY_LENGTH): Define.
5993 Changed all users of COND12_DELAY.
5997 * config/tc-mn10300.c (set_arch_mach): New function.
5998 (md_pseudo_table): Add pseudo-ops to set the current machine type.
5999 (md_begin): Default to mn10300 mode.
6000 (md_assemble): Only accept instructions for the core mn10300
6001 chip and the active machine type.
6005 * subsegs.h (segment_info_type): Give the struct a name.
6006 * config/tc-h8300.h (tc_reloc_mangle): Add prototype.
6007 * config/tc-h8500.h (tc_reloc_mangle): Declare.
6008 * config/tc-sh.h (sh_coff_reloc_mangle): Add prototype.
6009 * config/tc-w65.h (tc_reloc_mangle): Declare.
6010 * config/tc-z8k.h (tc_reloc_mangle): Declare.
6014 * config/tc-v850.c (v850_comm): Restore old section
6015 after common processing.
6019 * config/obj-vms.c (Create_VMS_Object_File): Force binary file.
6023 * config/tc-h8300.c (do_a_fix_imm, build_bytes): Replace cast to
6024 char with code that explicitly sign-extends.
6028 * config/tc-v850.c (md_begin): Restore text section as the current
6029 section after creating call table sections.
6030 * config/obj-coff.h (SYM_AUXINFO): New macro to conceal ugly
6033 * config/obj-coff.c (c_symbol_merge): Replace complex expresion
6034 with call to macro SYM_AUXINFO.
6038 * Makefile.am (install-exec-local): Don't let EXEEXT interfere
6039 with the program transform name.
6040 * Makefile.in: Rebuild.
6044 * config/obj-coff.c (c_symbol_merge): Fix copying of auxiliary
6049 * config/tc-i386.c (i386_operand): Be prepared for a space between
6050 the open parenthesis and the start of the register operand,
6051 because of the June 16 change.
6055 * config/tc-sh.c (md_apply_fix): Handle weak symbols correctly if
6060 * config/tc-d30v.c (d30v_align): Always perform alignment request,
6061 even if it is belived to be unnecessary.
6065 * write.c (adjust_reloc_syms): Never adjust relocs against weak
6067 * config/tc-mips.c (md_apply_fix): Adjust accordingly.
6071 * config/tc-mn10300.c (mn10300_insert_operand): Do not hardcode the
6072 shift amount for a repeated operand. The shift amount for the
6073 repeated copy comes from the size of the operand.
6077 * config/tc-h8300.c (get_operand): Fix typos in ldm/stm support.
6081 * config/tc-mips.c (md_show_usage): Fix -mipsN usage.
6085 * app.c (do_scrub_begin): If tc_symbol_chars is defined, treat all
6086 characters in it as LEX_IS_SYMBOL_COMPONENT.
6087 * config/tc-i386.h (tc_symbol_chars): Define.
6088 (extra_symbol_chars): Declare.
6089 * config/tc-i386.c (extra_symbol_chars): Define.
6090 (comment_chars): Don't use '/' as comment start if TE_LINUX.
6091 (line_comment_chars): Set to '/' if TE_LINUX.
6092 * doc/c-i386.texi (i386-prefixes): Update.
6093 * doc/internals.texi (CPU backend): Document tc_symbol_chars.
6097 * po/Make-in (all-yes): If maintainer mode, depend on .pot file.
6098 ($(PACKAGE).pot): Unconditionally depend on POTFILES.
6102 * config/tc-d10v.c (md_apply_fix3): Checking displacement
6103 constraint in instructions REP & REPI.
6107 * config/tc-d30v.c (md_apply_fix3): Catch BFD_RELOC_8,
6108 BFD_RELOC_16, BFD_RELOC_64 and issue appropriate error messages.
6110 (check_range): If the operand is shifted, then shift the number
6111 before checking its range.
6113 * write.c (adjust_reloc_syms): Add more checks for NULL pointers.
6115 * config/tc-v850.c (v850_comm): Set SEC_COMMON bit on special
6120 * config/tc-v850.c (v850_comm): Create special sections as needed.
6124 * config/tc-d10v.c (write_2_short): Addition of swapping
6125 instructions for sequential and reverse sequential order when
6126 given order is not possible.
6130 * Makefile.am: Rebuild dependencies.
6131 (DEP_INCLUDES): Fix reference to intl build directory.
6132 * Makefile.in: Rebuild.
6136 * doc/c-i386.texi: Update 16 bit documentation.
6138 * config/tc-i386.h: Change Data16 to Size16, Data32 to Size32,
6139 IgnoreDataSize to IgnoreSize as they are used for address size as
6141 * config/tc-i386.c: Likewise. Add code to reject addr32/data32 in
6142 32-bit mode, similarly addr16/data16 and variants.
6146 * config/tc-d30v.c (md_assemble): Fix handling of reverse
6147 sequential word multiply instructions.
6149 (do_assemble): Add extra command line argument, to allow mul32
6150 attribute to be preserved across parallel insns.
6151 (md_assemble): Insert NOPs between explicitly parallel insns which
6152 contain an 32 bit multiply and a 16 multiply.
6156 * config/tc-i386.c: REPNE renamed to REPNE_PREFIX_OPCODE, and
6159 * config/tc-i386.c (reloc): Add braces.
6161 * config/tc-i386.c (struct _i386_insn): Rename bi to sib to be
6162 consistent with Intel naming.
6163 * config/tc-i386.h (base_index_byte): Rename to sib_byte. Don't
6164 use bitfields in sib_byte.
6165 (modrm_byte): Don't use bitfields here either.
6167 * config/tc-i386.c (current_templates): Add const.
6168 (parse_register): Add const to return, param, and char *s.
6169 (i386_operand): Add const to reg_entry *r.
6170 * config/tc-i386.h (templates): Add const to start, end.
6172 Inspired by code for 16 bit gas support from Martynas Kunigelis
6174 * config/tc-i386.c (md_assemble): Add full support for 16 bit
6175 modrm, and Jump, JumpByte, JumpDword, JumpInterSegment insns.
6176 (uses_mem_addrmode): Remove.
6177 (md_estimate_size_before_relax): Add support here too.
6178 (md_relax_table): Rewrite interface to md_relax for 16 bit
6180 (BYTE, WORD, DWORD, UNKNOWN_SIZE): Remove.
6181 (opcode_suffix_to_type): Remove.
6182 (CODE16, SMALL, SMALL16, BIG, BIG16): Define.
6183 (SIZE_FROM_RELAX_STATE): Modify to suit above.
6184 (md_convert_frag): Likewise.
6185 (i386_operand): Add support for 16 bit base/index regs,
6186 immediates, and displacements. Remove some unnecessary casts, and
6187 localise end_of_operand_string, displacement_string_start,
6188 displacement_string_end variables. Add GCC_ASM_O_HACK.
6189 * config/tc-i386.h (NO_BASE_REGISTER_16): Define.
6191 * config/tc-i386.c (prefix_hash): Remove.
6192 (md_begin): Rewrite without obstacks. Remove prefix hash table
6193 handling. Rewrite lexical table handling.
6194 (i386_print_statistics): Don't print prefix statistics.
6195 (md_assemble): Rewrite instruction parser so that line is not
6196 converted to lower case. Don't do a hash_find for prefixes,
6197 instead recognise them via opcode modifier.
6198 (expecting_operand, paren_not_balanced): Localise variables.
6199 * config/tc-i386.h (IsPrefix): Define.
6200 (prefix_entry): Remove.
6202 * config/tc-i386.h (PREFIX_SEPERATOR): Don't define.
6203 * config/tc-i386.c (PREFIX_SEPARATOR): Define here instead, using
6204 '\\' in case where comment_chars contains '/'.
6206 * config/tc-i386.c (MATCH): Ensure given operand and template
6207 match for JumpAbsolute. Makes e.g. `ljmp table(%ebx)' invalid;
6208 you must write `ljmp *table(%ebx)'.
6211 * config/tc-i386.c (BFD_RELOC_16, BFD_RELOC_16_PCREL): Define
6212 as 0 ifndef BFD_ASSEMBLER.
6213 (md_assemble): Allow immediate operands without suffix or
6214 other reg operand to default in size to the current code size.
6218 * config/tc-v850.c (md_begin): Restore creation of
6219 .call_table_text and .call_table_data sections.
6223 * config/tc-d30v.c (md_assemble): Set execution type to unknown
6224 after emitting a word of noops.
6228 * config/tc-i386.c (mode_from_disp_size): Disp16 is mode 2.
6229 (i386_operand): Simplify checks for valid base/index combinations.
6230 Disallow `in 4(%dx),%al'.
6232 * config/tc-i386.c (struct _i386_insn): Make regs, base_reg, and
6234 (add_prefix): Change parameter from char to int.
6236 * config/tc-i386.h (Ugh): Define opcode modifier.
6237 * config/tc-i386.c (md_assemble): Print warnings for Ugh insns.
6239 * config/tc-i386.c (md_assemble): Rewrite MATCH and
6240 CONSISTENT_REGISTER_MATCH macros to check register types more
6241 thoroughly. Check for illegal suffix/operand combinations
6242 when matching insns with operands. Handle new `s' suffix, and
6243 associated FloatMF opcode modifier for float insns with memory
6245 * config/tc-i386.h (FloatMF): Define new opcode modifier.
6246 (No_sSuf, No_bSuf, No_wSuf, No_lSuf): Likewise.
6247 (SHORT_OPCODE_SUFFIX, LONG_OPCODE_SUFFIX): Define.
6248 * config/tc-i386.c: Rename WORD_PREFIX_OPCODE to
6249 DATA_PREFIX_OPCODE throughout.
6251 * config/tc-i386.c (REGISTER_WARNINGS): Define.
6252 (md_assemble): Rewrite suffix/register operand checking code to be
6253 more thorough. Remove Abs8,16,32. Change occurrences of Mem to
6254 AnyMem, the better to grep.
6256 (i386_operand): Don't set Mem bits in i.types[this_operand] when
6257 given a memory operand. Don't set Abs bits either.
6258 (type_names): Remove Mem*, Abs*.
6259 * config/tc-i386.h (Mem8, Mem16, Mem32, Abs8, Abs16, Abs32): Don't
6260 define opcode_modifiers as these cases are handled by Disp8,
6261 Disp16, Disp32 and suffix checks.
6262 (COMES_IN_BOTH_DIRECTIONS): Remove.
6263 (FloatR): Define. It's OK to share the bit with ReverseRegRegmem.
6265 * config/tc-i386.c (md_assemble): Don't emit operand size prefix
6266 if IgnoreDataSize modifier given. Remove ShortformW modifier
6267 test. Add test for ShortForm in W base_opcode modification.
6268 Merge Seg2ShortForm and Seg3ShortForm code.
6269 * config/tc-i386.h (ShortFormW): Remove.
6270 (IgnoreDataSize): Define.
6274 * config/tc-d30v.c (md_assemble): Store previous segment state
6275 with previous instruction.
6279 * config/tc-i386.c (SCALE1_WHEN_NO_INDEX): Define.
6280 (ebp, esp): Remove static variables.
6281 (MATCH): Remove test for InOutPortReg.
6282 (i386_operand): Properly handle InOutPortReg here instead.
6283 Disallows `inb (%dx,2)', `inb %es:(%dx)' and `mov (%dx),%ax'
6284 (md_assemble): Simplify and correct modrm and sib generation.
6285 (i386_operand): Add warning for scale without index.
6286 Rewrite checks for valid base/index combinations.
6288 * config/tc-i386.c (END_STRING_AND_SAVE): Protect arguments of
6289 macros and enclose in do while(0).
6290 (RESTORE_END_STRING): Likewise.
6291 (md_assemble): Add one to printed operand number so we start
6292 from 1 not 0. Add some more gettext invocations.
6293 (i386_operand): Fix `%%s' -> `%%%s'. Inc printed operand
6296 * config/tc-i386.h (WAIT_PREFIX, LOCKREP_PREFIX, ADDR_PREFIX,
6297 DATA_PREFIX, SEG_PREFIX): Define.
6298 * config/tc-i386.c (struct _i386_insn): Remove wait_prefix field.
6299 (check_prefix): Remove function.
6300 (add_prefix): New function. Add prefix to i.prefix as well as
6302 (md_assemble): Changes for add_prefix. Remove hack for wait
6303 prefix, instead always output prefixes in fixed order. Test
6304 for jcxz/loop when selecting between word & dword operations,
6305 and add address size prefix rather than operand size prefix.
6306 Remove operand -> address size hack when emitting jcxz/loop.
6307 (i386_operand): Remove O_Absent check as it's done in expr.
6311 * configure.in: Recognize m5200 as a cpu_type of m68k.
6312 * aclocal.m4: Rebuild with current libtool.
6313 * configure: Rebuild.
6317 * config/tc-m68k.c (md_estimate_size_before_relax): Add more calls
6318 to relaxable_symbol to prevent references to external symbol from
6323 * config/tc-m68k.c (relaxable_symbol): If TARGET_OS is "elf", all
6324 symbols are relaxable.
6328 * config/tc-v850.c (md_begin): Don't create special
6329 sections by default.
6333 * config/tc-mips.c (macro): For div and udiv, close the
6334 reorder block as soon as possible.
6339 * macro.c (macro_mri_mode): New function.
6340 * macro.h (macro_mri_mode): Declare.
6341 * read.c (s_mri): Call macro_mri_mode when switching in and out of
6346 * config/tc-alpha.c (s_alpha_comm): Allow alignment parameter in
6349 * config/tc-alpha.c (s_alpha_comm): Defer restoring character
6350 until after xstrdup in OBJ_EVAX case.
6354 * config/tc-vax.c (md_create_short_jump): Fix off by two bug in
6355 offset calculation. Also, use VAX_BRW from vax-inst.h instead
6356 of hardcoded magic number.
6357 (md_create_long_jump): Use VAX_JMP and VAX_ABSOLUTE_MODE macros.
6361 * read.c (do_s_func): New function.
6363 * read.h (do_s_func): Add prototype.
6367 * config/tc-m32r.c (m32r_do_align): Only fill code sections with
6368 nops if fill pattern not specified.
6373 * config/te-go32.h (TE_GO32): Define.
6374 * config/tc-i386.h (LOCAL_LABEL): Don't define if TE_GO32.
6378 Implement .func/.endfunc pseudo-ops.
6379 * read.h (stabs_generate_asm_func,stabs_generate_asm_endfunc): Declare.
6381 * read.c (potable): Add .func,.endfunc.
6382 (s_func): New function.
6383 * stabs.c (stabs_generate_asm_func,stabs_generate_asm_endfunc): New
6385 (in_doc_func_p,current_function_label): New static globals.
6386 (stabs_generate_asm_lineno): Emit function relative stabs if in .func.
6390 * config/tc-a29k.h (WORKING_DOT_WORD): Define.
6391 * config/tc-alpha.h (WORKING_DOT_WORD): Define.
6392 * config/tc-arm.h (WORKING_DOT_WORD): Define.
6393 * config/tc-h8300.h (WORKING_DOT_WORD): Define.
6394 * config/tc-h8500.h (WORKING_DOT_WORD): Define.
6395 * config/tc-hppa.h (WORKING_DOT_WORD): Define.
6396 * config/tc-i860.h (WORKING_DOT_WORD): Define.
6397 * config/tc-i960.h (WORKING_DOT_WORD): Define.
6398 * config/tc-tic30.h (WORKING_DOT_WORD): Define.
6399 * config/tc-w65.h (WORKING_DOT_WORD): Define.
6400 * config/tc-z8k.h (WORKING_DOT_WORD): Define.
6401 * config/tc-a29k.c: Don't define md_short_jump_size,
6402 md_long_jump_size, md_create_short_jump or md_create_long_jump.
6403 * config/tc-alpha.c: Likewise.
6404 * config/tc-alpha.h: Likewise.
6405 * config/tc-arm.c: Likewise.
6406 * config/tc-h8300.c: Likewise.
6407 * config/tc-h8500.c: Likewise.
6408 * config/tc-hppa.c: Likewise.
6409 * config/tc-i860.c: Likewise.
6410 * config/tc-i960.c: Likewise.
6411 * config/tc-ppc.c: Likewise.
6412 * config/tc-sh.c: Likewise.
6413 * config/tc-sparc.h: Likewise.
6414 * config/tc-tic30.c: Likewise.
6415 * config/tc-w65.c: Likewise.
6416 * config/tc-z8k.c: Likewise.
6420 * config/tc-vax.c (_): Delete this macro used for placeholder
6421 values in vax_operand_width_size; it conflicts with the _() macro
6422 used for internationalization.
6426 * symbols.c (symbol_find_base): Fix case insensitive symbol name
6430 * struc-symbol.h (struct broken_word): Add seg and subseg fields.
6431 * read.c (emit_expr): Initialize seg and subseg fields of a new
6433 * write.c (write_object_file): Switch to the appropriate segment
6434 and subsegment when processing a broken word.
6436 * config/tc-m68k.c (mri_assemble): New static function.
6437 (build_mri_control_operand): Call mri_assemble rather than
6439 (s_mri_else, s_mri_break, s_mri_next, s_mri_for): Likewise.
6440 (s_mri_endf, s_mri_endw): Likewise.
6444 * read.c (s_org): Call md_flush_pending_output if it is defined.
6446 * config/tc-sparc.c (md_show_usage): Add \n\ to new string.
6450 * config/tc-sparc.c (OPTION_LITTLE_ENDIAN_DATA): New.
6451 (md_parse_option): Add for same.
6452 (sparc_md_end): Set bfd_mach_sparc_sparclite_le.
6453 (md_apply_fix3, tc_gen_reloc): Allow BFD_RELOC_SPARC_32LE.
6454 (cons_fix_new_sparc): Added to create BFD_RELOC_SPARC_32LE.
6456 * config/tc-sparc.h (cons_fix_new_sparc): Added.
6460 * config/tc-arm.c (find_real_start): Relax definition of local
6465 * config/tc-d30v.c (d30v_align): Apply address adjustment to all
6466 symbols at the given address, not just the last one specified.
6470 * config/tc-sparc.c (sparc_handle_align): Use number_to_chars_bigendian
6471 or number_to_chars_littleendian to write data.
6475 * config/tc-v850.c (md_assemble): Remove artificially created
6476 register name symbols.
6480 * write.c (fixup_segment): Change "segment" to "section" in
6485 * write.c (fixup_segment): Change sym1-sym2 message again.
6489 * write.c (fixup_segment): Improve error message for sym1-sym2
6490 errors when sym1 is in a different segment from sym2.
6494 * config/tc-m32r.c (warn_unmatched_high): New static local.
6495 (OPTION_WARN_PARALLEL): Rename from OPTION_WARN.
6496 (OPTION_NO_WARN_PARALLEL): Rename from OPTION_NO_WARN.
6497 (md_longopts): Recognize --{no-,}warn-unmatched-high.
6498 (md_parse_option): Likewise.
6499 (md_show_usage): Likewise.
6500 (m32r_frob_file): Likewise.
6502 * read.c (generate_file_debug,generate_lineno_debug): New functions.
6503 (read_a_source_file): Call them.
6504 * read.h (stabs_generate_asm_file): Declare.
6505 * stabs.c (stabs_generate_asm_file): New function.
6506 (generate_asm_file): New function.
6507 (stabs_generate_asm_lineno): Move file name handling into
6512 * config/tc-d30v.c (cur_mul32_p, prev_mul32_p): Make static.
6513 (d30v_current_align, d30v_current_align_seg): New variables.
6514 (d30v_last_label): New variable.
6515 (d30v_align, s_d30v_align, s_d30v_text): New functions.
6516 (s_d30v_data, s_d30v_section): Likewise.
6517 (md_pseudo_table): Call them.
6518 (md_begin): Initialize d30v_current_align_seg.
6519 (md_assemble): Call d30v_align when needed by known current alignment.
6520 (d30v_frob_label, d30v_cons_align): New functions.
6521 * config/tc-d30v.h (md_do_align): Remove.
6522 (tc_frob_label): Call d30v_frob_label.
6523 (md_cons_align): New.
6525 * config/tc-d30v.c (find_format): Convert complex expressions to
6526 expression symbols before processing. Clean up code formatting.
6530 * po/Make-in (install-info): New target.
6534 * config/tc-mn10200.c (md_assemble): Handle "bra" just like "jmp"
6536 * config/tc-mn10300.c (md_assemble): Likewise.
6540 * Makefile.am: Update with `make dep-am'.
6541 (HFILES): Add cgen.h.
6542 (cgen.o): Depend on cgen.h.
6543 * Makefile.in: Regenerate.
6545 * cgen.c (cgen_md_apply_fix3): Don't pass newline to as_warn_where.
6549 * gasp.c (grab_label): Permit a label to be a preprocessor
6550 variable by permitting a label to start with a backslash.
6554 * config/tc-mips.c (validate_mips_insn): Removed hack
6555 for previously inaccessible bitfields in some INSN_TRAP
6560 * config/tc-d30v.c (do_assemble): Abort with error message
6561 if opcode operands do not match.
6565 * config/tc-mips.c (macro_build, validate_mips_insn): Implement
6566 'q' operand format for 20-bit "break"/"sdbbp" instructions.
6567 (mips_ip): Truncate overflowed "break" 'c' operand. Implement
6568 similar new 'q' operand.
6572 * cgen.c (cgen_asm_finish_insn): Fix typo.
6577 * cgen.c: Include it.
6578 (MAX_FIXUPS): Renamed to CGEN_MAX_FIXUPS.
6579 (cgen_asm_finish_insn): Result is now void. New arg `result'.
6580 All callers updated.
6581 * config/tc-m32r.c: Include cgen.h.
6582 (m23r_insn): New members num_fixups,fixups.
6586 * config/tc-hppa.c (md_apply_fix): Slightly rework some code
6587 to avoid compiler warning.
6591 * makefile.vms: Run dec c with /nodebug. Pass CC value when
6594 * makefile.vms (OBJS): Add ehopt.obj
6598 * doc/c-vax.texi: Correct and extend vax/vms documentation.
6602 * config/tc-d30v.c (do_assemble): Accept a new parameter requesting
6603 a short format insn.
6604 (md_assemble): Set it for explicitly packed insns.
6608 * config/obj-coff.c (c_symbol_merge): Do not take address of
6609 native fields when performing the memcpy.
6613 * config/tc-mips.c (macro,macro2): Implement
6614 M_DMULO_I, M_MULO_I, M_DMULOU_I, and M_MULOU_I.
6618 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Changed to keep
6619 relocations against globally visible symbols.
6620 * config/tc-m68k.c (relaxable_symbol): New macro.
6621 (m68k_ip, md_estimate_size_before_relax): Use it.
6622 (tc_m68k_fix_adjustable): Also handle weak symbols.
6626 * config/tc-i386.h (TC_RELOC_RTSYM_LOC_FIXUP): Keep relocs for all
6627 references to externally visible symbols.
6628 * config/tc-i386.c (md_apply_fix3): When OBJ_ELF, don't add the
6629 values in twice for a PC relative reloc if the symbol is
6632 * config/tc-sparc.h (tc_fix_adjustable) [OBJ_AOUT]: When PIC,
6633 don't adjust a PC relative reloc against an externally visible
6635 * config/tc-sparc.c (md_apply_fix3): When generating a.out PIC,
6636 for a PC relative fixup against an externally visible defined
6637 symbol, arrange to store object file and addend values as though
6638 the symbol were not defined.
6639 (tc_gen_reloc): Likewise.
6643 * read.c (sizeof_leb128): Referenced externally by write.c so
6648 * config/tc-m32r.c ({,expand_}debug_sym): New functions to record
6649 and expand a 'debug' symbol associated with the next instruction
6650 that does not cause a short instruction to be filled with a NOP.
6651 (md_pseudo_table): Add support for .debugsym.
6652 (assemble_parallel_insn): Add calls to expand_debug_sym as
6654 (md_assemble): Ditto.
6658 * as.c (main): Conditionally call setlocale.
6659 * gasp.c (main): Likewise.
6660 * asintl.h: Include <locale.h> if HAVE_LOCALE_H.
6661 (LC_MESSAGES): Now can be defined even when ENABLE_NLS.
6665 * config/tc-d30v.c (md_show_usage): Correct gettext typo.
6669 * config/tc-hppa.c: Change all calls to bzero to use memset.
6670 (pa_ip): Add cast to avoid warning.
6671 (tc_gen_reloc, md_apply_fix): Likewise.
6672 (pa_find_space_by_number): Likewise.
6673 (hppa_force_relocation): Likewise.
6674 (pa_block): Change i to unsigned int.
6675 * config/obj-som.h (obj_som_copyright): Declare.
6679 * ecoff.c (ecoff_build_lineno): Do not use dummy first_lineno
6680 for line numbers for assembly source.
6684 * configure.in: Change version number to 2.9.4
6685 * configure: Rebuild.
6689 * cgen.c (cgen_asm_finish_insn): New arg relax_p. All callers updated.
6693 * ecoff.h: Change symbolS in function declaration to struct
6698 * config/tc-sh.c (parse_reg): Add casts to avoid warnings.
6699 (md_convert_frag): Fix i18n typo.
6703 * ecoff.c (ecoff_get_cur_proc_sym): New function.
6704 * ecoff.h: Protoype it.
6705 * config/tc-alpha.c [ELF] (s_alpha_prologue): New function.
6706 [EVAX] (s_alpha_prologue): Delete.
6707 (md_pseudo_table): Update.
6711 * config/tc-i960.c (md_assemble): Change bp_error_msg from static
6712 array to local pointer.
6713 (get_args, parse_expr): Add casts to avoid warnings.
6717 * read.c (s_set): Cast xmalloc return value to fragS *.
6718 * config/tc-m68k.c (m68k_ip): Function made static to match
6719 previous forward declaration.
6720 (insert_reg, init_regtable, md_convert_frag_1): Likewise.
6724 * config/tc-v850.c: Add internationalisation macros to error
6727 * config/tc-m32r.c (can_make_parallel): Add internationalisation
6728 macros to error strings.
6732 * config/tc-ppc.c (ppc_fix_adjustable): Correct test of whether a
6733 reloc is in the TOC csect.
6734 (md_apply_fix3): Correct gettext typo.
6738 * config/tc-arm.c (find_real_start): Ignore symbols starting with
6739 .L - they are local labels and the branches are not really
6740 function calls but rather far jumps.
6744 * po/Make-in (MKINSTALLDIRS): Don't look in $(top_srcdir).
6748 * config/tc-i386.c (md_assemble): Print operand number rather than
6749 using ordinal_names.
6750 (i386_operand): Likewise.
6754 * Makefile.am (INTLLIBS): Define to work around apparent automake
6756 All Makefiles: Regenerated.
6758 * Many files: Added gettext invocations around user-visible
6760 * acconfig.h (ENABLE_NLS, HAVE_CATGETS, HAVE_GETTEXT, HAVE_STPCPY,
6761 HAVE_LC_MESSAGES): Define.
6762 * dep-in.sed: Added asintl.h.
6763 * po/Make-in: New file.
6764 * gasp.c (main): Call setlocale, bindtextdomain, and textdomain.
6766 * read.c (Z_): Renamed from `_'.
6767 * Makefile.am (SUBDIRS): Added po.
6768 (POTFILES): new macro.
6769 (po/POTFILES.in): New target.
6770 ($(OBJS)): Added asintl.h.
6772 (INCLUDES): Added -DLOCALEDIR, -I$(top_srcdir)/../intl.
6773 (as_new_LDADD): Added $(INTLLIBS).
6774 (as_new_DEPENDENCIES): Added $(INTLDEPS).
6775 (gasp_new_LDADD): Added $(INTLLIBS).
6776 (gasp_new_DEPENDENCIES): New macro.
6777 * configure, aclocal.m4: Rebuilt.
6778 * configure.in: Call CY_GNU_GETTEXT. Generate po/Makefile.in and
6780 (ALL_LINGUAS): Define.
6781 * macro.c: Include "asintl.h".
6782 * as.c (main): Call setlocale, bindtextdomain, and textdomain.
6783 * as.h: Include "asintl.h".
6784 * config/tc-i386.c (ordinal_names): Removed.
6785 (md_assemble): Changed error text to avoid ordinal_names.
6786 (i386_operand): Likewise.
6787 (reloc): Added as_bad to avoid i18n problems.
6788 (tc_gen_reloc): Likewise.
6789 * config/tc-arm.c (bad_args): Now a #define.
6791 * config/obj-vms.c (VMS_stab_parse): Changed type of
6793 (global_symbol_directory): Unified strings to avoid i18n
6795 * config/tc-m68k.c (get_reloc_code): Added some as_bad calls to
6796 avoid i18n problems.
6797 * config/tc-ns32k.c (reloc): Added as_bad to avoid i18n problems.
6798 * config/tc-ppc.c (md_apply_fix3): Added as_bad_where to avoid
6800 * config/tc-sh.c (md_convert_frag): Added as_bad to avoid i18n
6802 * config/tc-v850.c (md_assemble): Changed C++ comment into C
6804 * config/tc-vax.c (md_assemble): Added as_warn to avoid i18n
6806 * as.c (print_version_id): Added an fprintf to avoid i18n
6808 * cond.c (cond_finish_check): Added as_bad call to avoid i18n
6810 * expr.c (expr): Added as_warn call to avoid i18n problems.
6811 * messages.c (as_assert): Changed code to avoid i18n problems.
6812 (as_abort): Likewise.
6813 * read.c (pseudo_set): Added as_bad call to avoid i18n problems.
6814 (s_space): Likewise.
6815 * po/Make-in, po/POTFILES.in, po/gas.pot: New files.
6819 * config/tc-i386.c (check_prefix): New static function, split out
6821 (struct _i386_insn): Add wait_prefix field.
6822 (md_assemble): Remove wait_prefix local variable. Use
6823 check_prefix when adding a prefix.
6825 * config/tc-i386.c (current_templates): New static variable.
6826 (md_assemble): Remove current_templates local variable.
6827 (md_assemble, i386_operand): Improve error and warning messages in
6828 many places. Add RESTORE_END_STRING in many places before error
6829 return. Clarify some comments.
6831 * config/tc-i386.c (struct _i386_insn): Change seg field to a two
6833 (md_assemble): Parse string instruction operands, looking for
6834 segment override prefixes. Check for invalid segment prefixes on
6836 (i386_operand): i.seg[] and max mem_operand changes for string
6838 * config/tc-i386.h (EsSeg): Define.
6840 * config/tc-i386.h (regKludge): Define.
6841 (iclrKludge, imulKludge): Don't define.
6842 * config/tc-i386.c (md_assemble): Merge imulKludge and iclrKludge
6843 code. Move ReverseRegRegmem fudges into Modrm case. Reorder
6844 opcode_modifier checks to look for more common cases first. Add
6845 default_seg for IsString case.
6849 * configure.in: Call AM_PROG_LEX rather than AC_PROG_LEX and
6851 * configure: Rebuild with new automake and libtool.
6852 * aclocal.m4, Makefile.in: Likewise.
6854 * doc/Makefile.am (as.dvi): New target.
6855 * doc/Makefile.in: Rebuild.
6859 * configure.in: Added sparc86x support.
6861 * configure: Rebuild.
6863 * config/tc-sparc.c (lookup_arch): Added arch_type to struct
6865 (md_parse_option): Warn if -EL is not supported for this architecture.
6867 * config/tc-sparc.h (SPARC_BIENDIAN) Always define.
6871 * config/tc-mips.c (mips_ip): Note when we use get match on
6872 the full instruction name.
6876 * symbols.c (resolve_symbol_value) [O_symbol]: Also store the symbol
6877 back into the expression to handle add/sub simplification correctly.
6881 * config/tc-mips.c (hilo_interlocks): Remove 4300.
6885 * config/tc-arm.c (do_msr): Support undocumented 'msr cpsr_flg,
6890 * symbols.c (max_indent_level): New global.
6891 (print_symbol_value_1): Use it.
6892 * expr.h (expr_build_dot): Declare.
6893 * expr.c (expr_build_dot): New function.
6897 * symbols.c (print_binary): New function.
6898 (print_expr_1): Call it.
6902 * config/tc-m68k.c (m68k_ip, case "#B"): Install the offset of the
6903 operand in the opcode.
6907 * config/tc-i386.h: Reorder operand flags and opcode modifier
6908 flags for clarity. Remove unused definitions: Unknown,
6909 ImmUnknown, DispUnknown, NoModrm.
6910 * config/tc-i386.c (type_names): Add missing Debug type.
6911 (md_assemble): Better duplicate prefix checking. Quicker string
6912 instruction check via new opcode_modifier flag.
6916 * doc/as.texinfo (Invoking): Clarify -Wa example.
6920 * config/tc-mips.c (mips_pseudo_table): Add weakext entry.
6921 (s_mips_weakext): Define.
6922 * ecoff.c (ecoff_directive_weakext): Don't define if defined(TC_MIPS).
6923 * config/obj-ecoff.c (obj_pseudo_table): Don't add weakext if
6928 * config/tc-mn10200.c (tc_gen_reloc): The difference of two symbols
6929 is an error if the value can not be computed at assembly time.
6930 * config/tc-mn10300.c (tc_gen-reloc): Likewise.
6934 * gasp.c (main): Set next field of new include_path structure to
6937 * read.c (s_mri_sect): Call as_bad rather than abort for an
6938 unsupported MRI target.
6942 * config/tc-arm.c (arm_validate_fix): New function. Determine if
6943 the destination of a branch instruction should be altered.
6944 (find_real_start): New function: Locate the real, Thumb coded
6945 start of a Thumb function.
6946 (do_t_branch23): Alter the destination of branches to Thumb
6949 * config/tc-arm.h: Define TC_VALIDATE_FIX.
6953 * config/tc-arm.c (decode_shift): Handle addressing mode 2 w/rrx
6958 * doc/as.texinfo: Use @itemx for a secondary item in a table.
6959 * doc/c-hppa.texi: Likewise.
6963 * Makefile.am: Rebuild dependencies.
6964 * Makefile.in: Rebuild.
6966 * Makefile.am (DEP_INCLUDES): New variable.
6967 (.dep1): Change to work when srcdir is not an absolute path.
6968 (.tcdep, .objdep, .dep2, dep.sed): Likewise.
6969 * Makefile.in: Rebuild.
6973 * config/tc-i386.h, config/tc-i386.c: Revert March 24
6974 LinearAddress patch.
6976 * configure.in: Set version to 2.9.1.
6977 * configure: Rebuild.
6979 * Branched binutils 2.9.
6983 * config/tc-i386.h (FWait): Define.
6984 * config/tc-i386.c (md_assemble): Emit fwait prefix before any
6985 other prefixes. Check FWait flag in opcode table to see which
6986 instructions require an fwait prefix.
6990 * stabs.c (get_stab_string_offset): Always create a stab string
6995 Fix some gcc -Wall warnings:
6996 * atof-generic.c (atof_generic): Add casts to avoid warnings.
6997 * ehopt.c (eh_frame_code_alignment): Likewise.
6998 * expr.c (integer_constant, operand): Likewise.
6999 * frags.c (frag_align): Likewise.
7000 * gasp.c (level_0, change_base, doinstr): Likewise.
7001 * hash.c (hash_ask): Likewise.
7002 * listing.c (listing_page, calc_hex, print_lines): Likewise.
7003 (debugging_pseudo): Likewise.
7004 * macro.c (define_macro, check_macro): Likewise.
7005 * read.c (read_a_source_file, s_align, s_float_space): Likewise.
7006 (ignore_rest_of_line, float_cons): Likewise.
7007 * symbols.c (decode_local_label_name): Likewise.
7008 * write.c (record_alignment, cvs_frag_to_fill): Likewise.
7009 (fixup_segment, number_to_chars_bigendian): Likewise.
7010 (number_to_chars_littleendian): Likewise.
7011 * config/atof-ieee.c (gen_to_words): Likewise.
7012 * config/tc-sparc.c (md_begin, md_assemble): Likewise.
7013 (sparc_ip, parse_keyword_arg, s_common): Likewise.
7014 * read.c (output_big_sleb128): Initialize locals to avoid
7016 (output_big_uleb128, equals): Likewise.
7017 * atof-generic.c (atof_generic): Change number_of_digits_* locals
7018 to unsigned int. Change zeros to unsigned int.
7019 * cond.c (s_if): Add return to default case.
7020 * frags.c (frag_now_fix): Change return type to addressT.
7021 * frags.h (frag_now_fix): Update declaration.
7022 * listing.c (file_info_struct): Change linenum to unsigned int.
7023 (struct list_info_struct): Change hll_line to unsigned int.
7024 (print_source): Update format string.
7025 * read.c (emit_expr): Change scan to unsigned int, and don't
7026 bother to initialize it.
7027 * symbols.c (dollar_label_count): Change to unsigned long.
7028 * write.c (adjust_reloc_syms): Remove unused label reduce_fixup.
7029 * config/tc-sparc.c (sparc_memory_model): Only define if OBJ_ELF.
7030 * config/tc-sparc.c (tc_gen_reloc): Add return to default case.
7034 * config/tc-m68k.c (m68k_ip): Check legal addressing modes for
7035 mcf5200 just as we do for m68000.
7036 (m68k_init_after_args): Likewise.
7037 (md_estimate_size_before_relax): Likewise.
7041 * config/tc-v850.c (md_assemble): Store relocation addend in
7042 fixup instead of instruction.
7046 * config/tc-i386.c (md_assemble): Swap template arguments to
7047 CONSISTENT_REGISTER_MATCH macro in reverse direction test.
7048 This macro is currently symmetric, so passing them the wrong
7049 way didn't cause any problem, but may if the macro is changed
7051 After copying template to i.tm, use i.tm. rather than t-> to
7052 access fields, and make t a const*
7053 Move i.tm.operand_types[] swap to immediately after the copy.
7057 * expr.h (expr_build_uconstant): Add prototype.
7058 (expr_build_unary,expr_build_binary): Add prototypes.
7059 * expr.c (expr_build_uconstant): New function.
7060 (expr_build_unary,expr_build_binary): New functions.
7064 * gasp.c (IS*): Cast argument to unsigned char, not unsigned int.
7065 * macro.c (macro_expand_body): Increase buffer size.
7066 * messages.c (as_warn): Likewise.
7067 (as_warn_where, as_bad, as_bad_where): Likewise.
7072 * Makefile.am (DISTSTUFF): New variable.
7073 (diststuff): New target.
7074 * Makefile.in: Rebuild.
7078 * config/tc-m32r.h (md_cleanup, md_elf_section_change_hook): Call
7079 m32r_elf_section_change_hook.
7081 * config/tc-m32r.c (m32r_elf_section_change_hook): New function to
7082 emit a nop if a section ends with a 16 bit instruction.
7086 * config/obj-coff.c (obj_coff_bss): Compile unconditionally. Call
7087 s_lcomm rather than obj_coff_lcomm.
7088 (obj_pseudo_table): Compile .bss pseudo-op unconditionally.
7092 * config/tc-i386.h (LinearAddress): Define.
7093 * config/tc-i386.c (md_assemble): If LinearAddress is set for the
7094 instruction, don't use a default segment.
7098 * configure.in: (sh*-*-rtems*): Switched from ELF to COFF.
7099 * configure: Rebuild.
7103 * aclocal.m4, configure: Rebuild with libtool 1.2.
7107 * config/tc-arm.c (md_apply_fix3): fix code to test the range of
7108 PC relative branches. Patch courtesy of Jonathan Walton.
7113 * configure.in (emulations): Add thumb-pe target.
7115 * configure (emulations): Add thumb-pe target.
7119 * itbl-lex.l (yywrap): Don't define if already defined as a
7124 * depend.c (quote_string_for_make): New function.
7125 (wrap_output): Use it.
7129 * config/obj-elf.c (obj_elf_section): Set bss flag in seg_info
7130 structure if type is SHT_NOBITS. [Bug fix courtesy of rth]
7134 * config/tc-alpha.c (md_shortopts, md_longopts, md_parse_option):
7135 Recognize -GN and -relax.
7136 (md_begin): Initialize gp size from -G switch.
7137 (alpha_force_relocation): Always force if -relax.
7138 (alpha_align): Take a new argument that will specify when to
7139 emit an R_ALPHA_ALIGN relocation (though we don't do that now).
7140 Change all callers. Emit nop alignment padding as nop+unop pair.
7144 * config/obj-elf.c [TC_ALPHA]: Include <elf/alpha.h>.
7145 * config/tc-alpha.h (ELF_TC_SPECIAL_SECTIONS): New.
7149 * config/tc-d30v.c (write_2_short): Delayed jsr instructions don't
7150 require padding to the next long word boundary.
7154 * cgen.c: #include symcat.h.
7155 * config/tc-m32r.c: Likewise.
7159 * config/tc-mips.c (mips_ip, case 'P'): Make 'P' arguments be
7160 absolute expressions instead of '$' prefixed register names.
7164 * read.c (s_set): Record file and line info for symbols when -as.
7165 (pseudo_set): Don't overwrite that dummy fragment.
7169 * config/tc-ppc.c (md_pseudo_table): Add "section".
7170 (ppc_named_section): New static function.
7174 * tc-ppc.c (ppc_biei): Cache the last symbol we inserted
7175 so we don't have to scan the entire list.
7179 * config/tc-d30v.c (parallel_ok): For the explicitly parallel
7180 case, allow the parallel instructions to modify the same flag
7185 * listing.c (list_symbol_table): Categorize symbols by
7186 undefined_section rather than sy_frag->line == NULL.
7190 * Makefile.am (install-exec-local): Install properly when ln
7191 fails or tooldir == prefix.
7195 * cgen.c (cgen_md_apply_fix3): Delete call to validate_operand.
7196 Test result of insert_operand for error.
7200 * Makefile.am (AUTOMAKE_OPTIONS): Add cygnus.
7201 * configure, Makefile.in, aclocal.m4: Rebuild with automake 1.2e.
7202 * doc/Makefile.am (AUTOMAKE_OPTIONS): Define.
7203 * doc/Makefile.in: Rebuild.
7207 * config/tc-mips.c (macro_build): Handle operand type 'C'.
7208 (macro): Fix handling of M_COP[0-3].
7213 * macro.c (ISSEP): Only treat '<' and '>' as separator characters
7214 if macro_alternate or macro_mri.
7215 (getstring): Remove support for byte constants between < and >.
7216 (get_any_string): '<' only starts a string if macro_alternate or
7218 (macro_expand_body): Permit keyword parameters following
7219 positional parameters.
7222 * configure.in (alpha*-*-netbsd*): New target.
7223 * config/te-nbsd.h (LOCAL_LABELS_FB): Define.
7224 * configure: Rebuild.
7226 * as.h (flag_warn_suppress_instructionswap): Move from here...
7227 * config/tc-d10v.c (flag_warn_suppress_instructionswap): ...to
7228 here, and make static.
7230 * ehopt.c (eh_frame_code_alignment): Only use seg_info if
7231 BFD_ASSEMBLER or MANY_SEGMENTS.
7233 * as.c (show_usage): Update bug-gnu-utils address.
7234 * gasp.c (show_usage): Likewise.
7235 * doc/as.texinfo (Bug Reporting): Likewise.
7239 * config/tc-mips.c (load_address): Don't use mips III or mips IV
7240 insns regardless of the size of a pointer if we're in mips I or
7242 (macro, macro2, s_cprestore, s_cpadd): Likewise.
7246 Fix rac to accept only a0:
7247 * tc-d10v.c (parallel_ok, find_opcode):
7248 Split OPERAND_ACC into OPERAND_ACC0 and OPERAND_ACC1.
7249 Introduce OPERAND_GPR.
7253 * read.c (s_fill): Handle non-constant repeat counts by creating
7254 an rs_space fragment.
7258 * config/tc-i386.c (i386_operand): Change error added Jan 2 1998
7259 from as_bad to as_warn.
7263 * as.c: (perform_an_assembly_pass): Use [TEXT|DATA|BSS]_SECTION_NAME
7264 * as.h: Define default values of [TEXT|DATA|BSS]_SECTION_NAME
7265 * config/obj-elf.c (elf_begin): Use [TEXT|DATA|BSS]_SECTION_NAME
7269 * ehopt.c (eh_frame_code_alignment): If not BFD_ASSEMBLER, use
7270 seg_fix_rotP rather than fix_root from seg_info.
7274 * expr.c: Add support for 0x1_2_3_4 bignums.
7278 * configure.in: Change -linux* to -linux-gnu*.
7279 * configure: Rebuild.
7281 * app.c (do_scrub_begin): Treat \r as whitespace.
7285 * Makefile.am: Update dependencies.
7286 * Makefile.in: Rebuild.
7290 * configure, aclocal.m4: Rebuild with new libtool.
7294 * config/tc-mips.c (md_begin): If mips_cpu is set, then use it as
7295 the argument to bfd_set_arch_mach.
7296 (load_address): Use bfd_arch_bits_per_address to determine the
7297 bit size of an address instead of looking at the isa level.
7298 (macro, macro2, s_cprestore, s_cpadd): Likewise.
7302 * config/tc-v850.c (md_parse_option): Add -mv850any command line option.
7306 * ehopt.c: New file.
7307 * as.h (enum _relax_state): Add rs_cfa.
7308 (check_eh_frame, eh_frame_estimate_size_before_relax): Declare.
7309 (eh_frame_relax_frag, eh_frame_convert_frag): Declare.
7310 * read.c (emit_expr): Call check_eh_frame.
7311 * write.c (cvt_frag_to_fill): Handle rs_cfa.
7312 (relax_segment): Likewise.
7313 * Makefile.am: Rebuild dependencies.
7314 (GAS_CFILES): Add ehopt.c.
7315 (GENERIC_OBJS): Add ehopt.o.
7316 * doc/internals.texi (Frags): Document rs_cfa.
7318 * as.c (show_usage): Mention --traditional-format.
7319 (parse_args): Accept --traditional-format.
7320 * as.h (flag_traditional_format): Declare.
7321 * output-file.c (output_file_create): If flag_traditional_format,
7322 set BFD_TRADITIONAL_FORMAT on stdoutput.
7323 * doc/as.texinfo, doc/as.1: Document --traditional-format.
7325 * config/tc-mips.c (append_insn): Make sure that if we have a
7326 fixup for an unmatched %hi reloc, it does not associated with a
7329 * configure, Makefile.in, aclocal.m4: Rebuild with new libtool.
7330 * doc/Makefile.in: Likewise.
7334 * config/tc-m32r.c (check_for_side_effects): New function.
7335 (can_make_parallel): Add checks for instruction side effects
7336 clashing with the other instruction.
7337 (assemble_parallel_insn): Improve warning messages. Return error
7338 message from non-swapped instruction order.
7342 * config/tc-arm.c: Rename arm_after_pass_hook() to arm_cleanup().
7344 * config/tc-arm.h: Replace md_after_pass_hook definition with a
7345 md_cleanup definition. This moves the forced literal output to
7346 the end of the source pass, and avoids macro's inserting literals
7347 into the code immediately after the macro expansion.
7351 * config/tc-sparc.h (tc_fix_adjustable) [OBJ_ELF]: A reloc against
7352 a gas internal symbol is adjustable.
7353 * config/tc-ppc.h (tc_fix_adjustable): Likewise.
7355 * as.h: If gcc version greater than 2.6, use `__format__' and
7356 `__printf__' in function attributes, rather than `format' and
7361 * config/tc-sparc.c: Only include elf/sparc.h if OBJ_ELF.
7365 Add tms320c30 support:
7366 * config/tc-tic30.h: New file.
7367 * config/tc-tic30.c: New file.
7368 * config/obj-coff.h: If TC_TIC30, include coff/tic30.h and define
7369 TARGET_FORMAT as "coff-tic30".
7370 * configure.in (tic30-*-*aout*, tic30-*-*coff*): New targets.
7371 * Makefile.am: Rebuild dependencies.
7372 (CPU_TYPES): Add tic30.
7373 (CPU_OBJ_VALID): tic30-aout is valid.
7374 (TARGET_CPU_CFILES): Add config/tc-tic30.c.
7375 (TARGET_CPU_HFILES): Add config/tc-tic30.h.
7376 * configure, Makefile.in: Rebuild.
7380 * config/tc-v850.c (md_assemble): Improvements to error messages.
7384 * config/tc-ppc.c (md_apply_fix3): Change BFD_RELOC_HI16 and
7385 BFD_RELOC_HI16_S to store the high bits of any value.
7387 * config/tc-ppc.h (tc_fix_adjustable): Undo change of Fri Jun 27.
7388 (TC_RELOC_RTSYM_LOC_FIXUP): Don't let the
7389 assembler calculate relocations to any external symbol at all.
7390 * config/tc-ppc.c (md_apply_fix3) [OBJ_ELF]: Correct bugs
7391 involving generation of pc-relative relocs.
7392 (md_pcrel_from_section) [OBJ_ELF]: The job this code used to do
7393 has been moved to md_apply_fix3.
7395 * config/tc-ppc.c (md_apply_fix3): Fix test for too-far branch.
7396 (ppc_elf_suffix): Warn about 'identifier+constant@got' syntax,
7397 which actually means (the address of identifier's GOT entry) +
7398 constant, which is not particularly useful.
7402 * read.h (include_dirs): Declare.
7403 (include_dir_count,include_dir_maxlen): Declare.
7407 * configure.in: Correct check for shared opcodes library.
7408 * configure: Rebuild.
7410 * listing.c (buffer_line): If we can't open the file, set at_end.
7411 (listing_print): Remove unused local variable fi.
7413 * config/m68k-parse.y (reglistpair): Handle register list in
7416 * config/vms-conf.h: Don't undef VERSION.
7420 * Makefile.am (CONFIG_OBJS): New variable, containing part of old
7422 (GENERIC_OBJS): New variable, with the rest of the old OBJS
7424 (OBJS): Now just $(CONFIG_OBJS) and $(GENERIC_OBJS).
7425 ($(srcdir)/make-gas.com): Rename from make-gas.com.
7426 (stamp-mk.com): Replace $(OBJS) with $(GENERIC_OBJS).
7427 (EXTRA_DIST): Define.
7428 * vmsconf.sh: Handle {targ-cpu, obj-format, atof-targ} modules
7429 explicitly rather than via the list of object files.
7430 (gcc-as.opt): New file created when make-gas.com is run.
7431 * config-gas.com: Create {targ-cpu.h, obj-format.h, targ-env.h,
7432 itbl-cpu.h} to #include appropriate file rather than copying that
7434 * config/vms-conf.h: Synchronize with current config.in.
7435 * Makefile.in: Rebuild.
7439 * config/tc-d30v.c (do_assemble): Added flag_explicitly_parallel.
7440 (parallel_ok): Relaxed parallel subinstruction dependency check.
7444 * as.h (flag_warn_suppress_instructionswap): added new flag.
7445 * tc-d10v.c (md_parse_option,md_longopts): added "--nowarnswap"
7446 command line argument.
7447 * tc-d10v.c (write_2_short): emit "Swapping instructions"
7448 warning only if flag_warn_suppress_instructionswap is false.
7452 * configure.in (i386-*-mingw32*): New target.
7453 * configure: Rebuild.
7457 * symbols.c (resolve_symbol_value): Don't set the segment if it
7458 hasn't changed, and this is OBJ_AOUT without BFD_ASSEMBLER.
7460 * config/obj-aout.h (S_IS_LOCAL): Correct typo--pass argument to
7465 as.h (unlink): Reverse 13-Feb-97 change; use of unlink vs remove
7466 depends upon HAVE_{UNLINK,REMOVE} values rather than host
7471 * config/obj-coff.h (RESOLVE_SYMBOL_REDEFINITION): Define.
7475 * config/tc-v850.c (v850_insert_operand): Display instruction when
7476 an error is encountered.
7480 * configure.in (i386-*-sco3.2v5*): Defaults to ELF now.
7481 (i386-*-sco3.2v5*coff): New target.
7482 (i386-*-sco3.2*): New target.
7483 * configure: Rebuild.
7487 * config/tc-v850.c: Tidy error message production.
7491 * config/tc-arm.c (md_apply_fix3): Add new variable newimm to hold
7492 validate_immediate return value in the right type for comparisons
7497 * listing.c (MAX_BYTES): Use listing variables not constants.
7498 (data_buffer): No longer an array, but a pointer.
7499 (calc_hex): sizeof(data_buffer) -> MAX_BYTES.
7500 (listing_listing): Allocate data_buffer.
7504 * as.c (parse_args): Add --listing-lhs-width, --listing-lhs-width2,
7505 --listing-rhs-width, --listing-cont-lines.
7506 (show_usage): Update.
7507 * listing.c (listing_lhs_width, listing_lhs_width_second): New vars.
7508 (listing_lhs_cont_lines, listing_rhs_width): New vars.
7509 (print_lines): Use the variables instead of the constants.
7510 (listing_listing): Likewise.
7511 * listing.h: Declare the new vars.
7515 * as.c (parse_args): Add --keep-locals alias for -L.
7516 Add --strip-local-absolute.
7517 (show_usage): Update.
7518 * as.h (flag_strip_local_absolute): New flag.
7519 * symbols.c (S_IS_LOCAL): Use it.
7520 * config/obj-aout.h (S_IS_LOCAL): Likewise.
7521 * config/obj-bout.h (S_IS_LOCAL): Likewise.
7522 * config/obj-coff.h (S_IS_LOCAL): Likewise.
7526 * config/tc-m32r.c: Detect if explicitly parallel instructions
7527 might have an io conflict and issue a warning message.
7531 * cgen.c (cgen_save_fixups, cgen_restore_fixups,
7532 cgen_swap_fixups): Functions to save, restore and swap the fixup
7533 chain with a backup copy.
7534 (cgen_asm_finish_insn): Returns address of constructed insn.
7538 * listing.c (file_info_struct): Remove FILE, add POS.
7539 (last_open_file_info, last_open_file): New; a one entry FILE* cache.
7540 (file_info): Don't open the file.
7541 (buffer_line): Check for the file in the last_open cache, updating
7543 (print_source): Don't reference file_info->file.
7544 (listing_listing): Likewise.
7545 (listing_print): Close the file in the cache, if any.
7549 * read.c (dwarf_file_string): New file static variable.
7550 (emit_expr): Look for constant sequence that leads up to a file
7551 name in DWARF debugging output.
7552 (stringer): Use dwarf_file_string to decide whether to accept a
7553 string as a file name.
7557 * tc-m68k.c (m68k_ip): Remove absl->reglst MRI hack.
7558 (crack_operand): Add reg->reglst MRI hack.
7559 (r_seg): Put reglst symbols in reg_section.
7560 (m68k_frob_symbol): Frob reglst symbols into absolute_section.
7564 * tc-sh.c (get_specific): Handle SGR & DBR.
7568 * tc-h8300.c (parse_reg): Take the length of the symbol into
7569 account when attempting to match a register name.
7570 * tc-h8500.c (parse_reg): Likewise.
7574 * cgen.c: Formatting changes to improve readability.
7578 * config/tc-mips.c (macro): Rework division code to avoid unfilled
7583 Based on a patch from Jim Wilson
7584 * config/tc-d30v.c (do_assemble): Remove non-ansi default case.
7585 (tc_gen_reloc): Handle cross section PC relative relocs
7590 * config/tc-mips.c (mips_ip): Don't test pinfo flags if INSN_MACRO.
7594 * cgen.c: #include setjmp.h. Clean up pass over `struct foo' usage.
7595 (expr_jmp_buf): New static local.
7596 (cgen_parse_operand): Allow use of longjmp in parsing to handle errors.
7597 (cgen_md_operand): New function.
7598 * tc-m32r.c: Clean up pass over `struct foo' usage.
7599 (md_estimate_size_before_relax): Use CGEN_INSN_MNEMONIC.
7603 * symbols.c (S_SET_SEGMENT): Don't set the segment for section syms.
7604 (S_IS_EXTERNAL, S_IS_LOCAL): Correct parenthetication.
7608 * config/tc-i386.c (i386_operand): Give an error if there are
7609 unrecognized characters after an expression.
7611 For older changes see ChangeLog-9697