3 * config/tc-arm.c (fix_new_arm): Use make_expr_symbol to handle a
6 * symbols.c (resolve_symbol_value): If both left and right
7 operands are undefined, warn about both of them.
12 * config/obj-elf.c (elf_pseudo_table): Add "symver".
13 (obj_elf_symver): New static function.
14 * config/obj-elf.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Copy the st_other
17 * write.c (relax_segment): Make type and printf format agree.
19 * read.c (get_line_sb): Don't end the line on a semicolon inside a
25 * config/tc-d30v.c, config/tc-d30v.h: New files.
29 * configure.in: Add case for d30v.
34 * config/tc-alpha.h (md_operand): Define with a null expansion,
35 like all the other targets.
36 * doc/internals.texi (CPU backend): Add missing word in
37 md_flush_pending_output description. Fix typo in md_convert_frag
40 * config/tc-tic80: Minor comment additions/changes.
45 * config/tc-m68k.c (LOCAL_LABEL): Macro redefined if TE_DELTA.
46 (tc_canonicalize_symbol_name): Macro defined if TE_DELTA.
47 * config/obj-coff.c (obj_coff_def): Use
48 tc_canonicalize_symbol_name if defined.
49 (obj_coff_tag, obj_coff_val): Likewise.
50 * expr.c (operand): Reject '~' as operator if is_name_beginner.
55 does not actually work, though:
56 * configure.in (i386-sequent-bsd*): New target.
58 * config/tc-dynix.h: New file.
59 * config/tc-i386.h: Define TARGET_FORMAT if TE_DYNIX.
61 * read.c (do_align): Add max parameter. Change all callers.
62 Remove useless static variables.
63 (s_align): New static function. Do common portion of
64 s_align_bytes and s_align_ptwo.
65 (s_align_bytes, s_align_ptwo): Just call s_align.
66 * frags.c (frag_align): Add max parameter. Change all callers.
67 (frag_align_pattern): Likewise.
68 * frags.h (frag_align, frag_align_pattern): Update declarations.
69 * write.c (relax_segment): Limit alignment change to fr_subtype.
70 Fix some types to be addressT.
71 * config/obj-coff.c (size_section): Likewise.
72 * config/obj-ieee.c (size_section): Likewise.
73 * config/tc-d10v.h (md_do_align): Add max parameter.
74 * config/tc-i386.h (md_do_align): Likewise.
75 * config/tc-m88k.h (md_do_align): Likewise.
76 * config/tc-m88k.c (m88k_do_align): Likewise.
77 * config/tc-sh.h (md_do_align): Likewise.
78 * config/tc-sh.c (sh_do_align): Likewise.
79 * as.h: Improve comments on rs_align and rs_align_code.
80 * doc/as.texinfo: Document new alignment arguments.
81 * doc/internals.texi (Frags): Document use of fr_subtype field for
82 rs_align and rs_align_code.
86 * config/tc-mips.c: Changed opcode parsing.
90 * config/{tc-alpha.h, tc-arc.h, tc-d10v.h, tc-generic.h, tc-i960.h,
91 tc-mn10200.h, tc-mn10300.h, tc-sh.h, tc-v850.h, tc-vax.h, tc-w65.h}:
92 Add default definition of zero for TARGET_BYTES_BIG_ENDIAN.
93 * config/{tc-arm.h, tc-hppa.h, tc-i386.h, tc-mips.h, tc-ns32k.h,
94 tc-ppc.h, tc-sparc.h}: Move definition of TARGET_BYTES_BIG_ENDIAN
95 to a location consistent with the rest of the target include files.
96 * config/tc-i386.c: Remove misleading comment.
97 * doc/internals.texi (CPU backend): Add description of function
100 * config/tc-tic80.c: Add code to insert predefined symbols into the
101 symbol table so they can be parsed by the standard expression parser.
102 Remove custom code that use to parse them.
103 * config/tc-tic80.h: Move definition of TARGET_BYTES_BIG_ENDIAN
104 to a location consistent with the rest of the target include files.
109 * as.h: GNU c provides unlink() function.
111 Unify section handling on openVMS/Alpha:
112 * config/tc-alpha.c(s_alpha_link): Remove.
113 (s_alpha_section): New function.
114 Remove case-hacking of symbols
115 Add .code_address pseudo-op.
116 (BFD_RELOC_ALPHA_CODEADDR): New relocation.
117 (s_alpha_code_address): New function.
118 (alpha_ctors_section, alpha_dtors_section): New sections for C++
119 static constructors/destructors.
120 Add debug code for crash debugs, to be removed when traceback code
121 is added to object code.
122 (s_alpha_name): New function for .name pseudo-op.
123 (alpha_print_token): New function to print token expressions with
124 alpha specific extensions.
126 * makefile.vms: Allow compilation with current gcc snapshot.
130 * doc/Makefile.in (TEXI2DVI): Set to just name of program.
131 (DVIPS): Set to dvips.
132 (ps, as.ps, gasp.ps): New targets.
133 (internals.info, gasp.dvi, internals.dvi): Set both TEXINPUTS
134 and MAKEINFO env variables.
135 (internals.ps): Use DVIPS macro.
136 (clean): Remove core and backup files.
137 (distclean): Remove temporary files from building internals.
139 * doc/internals.texi (Frags): Fix typo.
140 (GAS processing): Ditto.
141 (CPU backend): Ditto.
142 * ecoff.c (init_file): Use TARGET_BYTES_BIG_ENDIAN value directly.
143 * mpw-config.in: Define TARGET_BYTES_BIG_ENDIAN as 1.
144 * read.c: Remove ugly hack that dealt with config files not
145 correctly defining TARGET_BYTES_BIG_ENDIAN.
146 (target_big_endian): Use TARGET_BYTES_BIG_ENDIAN directly.
147 * config/arm-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
148 * config/arm-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
149 * config/mips-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
150 * config/mips-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
151 * config/ppc-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
152 * config/ppc-sol.mt: Replace TARGET_BYTES_LITTLE_ENDIAN
153 with TARGET_BYTES_BIG_ENDIAN defined to 0.
154 * config/tc-arm.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
155 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
156 whether it is defined or not.
157 * config/tc-mips.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN.
158 * config/tc-ppc.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
159 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
160 whether it is defined or not.
162 * config/tic80.h (TARGET_FORMAT): Define to coff-tic80.
163 (TARGET_BYTES_BIG_ENDIAN): Define to 0.
168 * write.c (write_relocs): Correct text in as_fatal error message,
169 bfd_perform_relocation -> bfd_install_relocation.
173 * config/tc-m68k.c (LEX_TILDE): Define if TE_DELTA.
174 * read.c (LEX_TILDE): Define if not defined.
175 (lex_type): Use LEX_TILDE.
176 * expr.c (get_symbol_end): Check first char with is_name_beginner,
181 * config/tc-sparc.c (md_show_usage): Add missing backslash at end
184 * config/tc-mips.c (mips16_extended_frag): Correct base address
185 for an extended PC relative instruction.
186 (md_convert_frag): Likewise.
188 * config/tc-mips.c (prev_nop_frag): New static variable.
189 (prev_nop_frag_holds): New static variable.
190 (prev_nop_frag_required): New static variable.
191 (prev_nop_frag_since): New static variable.
192 (append_insn): If we aren't reordering, and prev_nop_frag is not
193 NULL, and we don't need any nops, then decrease the size of
194 prev_nop_frag. Don't insert nops because of instructions in
195 noreorder sections. Remember whether the previous instructions
196 where in noreorder sections even when not reordering.
197 (mips_no_prev_insn): Add preserver parameter. Change all
198 callers. Refer prev_nop_frag variables when appropriate.
199 (mips_emit_delays): Set up prev_nop_frag.
200 (s_mipsset): Clear prev_nop_frag if reordering.
204 * config/tc-mips.c (append_insn): Remove useless code which
205 handled swapping a mips16 jump with a mips16 instruction with a
208 * config/tc-mips.c (md_parse_option): When debugging, set
209 mips_optimize to 1, not 0.
211 * config/tc-mips.c (mips16_ip): Handle an extend operand.
213 * config/tc-mips.c (my_getExpression): In mips16 mode, if it looks
214 like the expression was based on `.', adjust the value of the
217 * config/tc-mips.c (append_insn): Warn about an attempt to put an
218 extended instruction in a delay slot when not reordering.
219 (md_convert_frag): Warn if an extended instruction appears in a
222 * config/tc-mips.c (mips_pseudo_table): Add "insn".
223 (s_insn): New static function.
224 * doc/c-mips.texi: Document .insn.
226 * config/tc-mips.c (md_begin): Add the general registers to the
228 (mips16_ip): First parse the expression, and then see whether it
229 came up with a register, rather than trying to first see whether
230 we are looking at a register.
235 * config/tc-tic80.c: Numerous changes and additions to flesh
236 out functions that were previously just stubs, and fix some
237 problems found using the new TIc80 testsuite cases.
242 * config/tc-mips.c (mips16_ip): Handle %gprel modifier.
243 (md_apply_fix): Handle BFD_RELOC_MIPS16_GPREL.
245 * config/tc-mips.c (append_insn): Output jump instruction as a
246 pair of 2 byte instructions, rather than as a single 4 byte
250 * config/tc-mips.c (md_begin): Correct test of mips_5900.
251 (mips_ip): Don't check INSN_ISA for a macro.
257 * config/tc-mips.c: added r5900.
262 * itbl-ops.c, itbl-lex.l, itbl-parse.y, itbl-ops.h,
263 config/itbl-mips.h: Add copyright message and fix indentation.
267 * itbl-ops.c: New file. Add support for dynamically read
268 instruction registers, opcodes and formats. Build internal table
269 for new instructions and provide callbacks for assembler and
271 * itbl-lex.l, itbl-parse.y: Lex and yacc parsers for instruction
273 * itbl-ops.h: New file. Header file for itbl support.
274 * config/itbl-mips.h: New file. Mips specific definitions for
279 * config/tc-mn10200.c (md_assemble): If a constant operand won't
280 fit into the constant field of a relaxable operand, then it does
285 * config/tc-mn10200.c (md_estimate_size_before_relax): Treat
286 a jsr target in a different section just like a jsr to
291 * config/tc-mips.c (mips_fix_adjustable): Don't adjust relocations
292 against any mips16 symbols, not just externally visible ones.
293 (md_apply_fix): Corresponding change.
297 * config/tc-mips.c (mips16_ip): Accept floating point registers in
298 the operand of the exit instruction.
302 * symbols.c (resolve_symbol_value): If we leave an equated symbol
303 as O_symbol, copy over the segment.
307 * config/tc-mips.c (md_apply_fix): If we aren't adjusting this
308 fixup to be against the section symbol, adjust the value
311 * symbols.c (resolve_symbol_value): Don't change X_add_number for
313 * write.c (write_relocs): Avoid looping on equated symbols.
314 Adjust fx_offset by X_add_number for each symbol.
315 * config/obj-coff.c (do_relocs_for): Avoid looping on equated
317 (fixup_segment): Add a loop to track down equated symbols and
318 adjust fx_offset appropriately.
322 * config/tc-mn10200.c (md_relax_table): Add entries to allow
324 (md_convert_frag): Handle jmp->bra relaxing.
325 (md_assemble): Handle jmp->bra relaxing.
326 (md_estimate_size_before_relax): Likewise.
330 * config/tc-i386.c (i386_align_code): Add comments explaining the
335 * config/tc-sparc.c (enforce_aligned_data): New static variable.
336 (sparc_cons_align): Don't do anything unless enforce_aligned_data
338 (md_longopts): Add "enforce-aligned-data".
339 (md_show_usage): Mention --enforce-aligned-data.
340 * doc/c-sparc.texi (Sparc-Aligned-Data): New node; document
341 enforce-aligned-data.
343 * config/tc-ppc.c (md_pseudo_table): If OBJ_XCOFF, add "long",
345 (ppc_xcoff_cons): New static function.
347 * write.c (relax_segment): Give an error if a .space symbol is
350 * read.c (read_a_source_file): Don't handle mri_pending_align if
351 the handler is s_globl or s_ignore.
355 * config/tc-d10v.c (find_opcode): Remove unused variable "numops".
357 * config/tc-tic80.c: Many additions to previous placeholder file.
358 * config/tc-tic80.h: Ditto.
363 * config/tc-i386.c (i386_align_code): Improve the nop patterns.
367 * config/tc-mips.c (mips_fix_adjustable): New function.
368 * config/tc-mips.h (tc_fix_adjustable): Call mips_fix_adjustable.
369 (mips_fix_adjustable): Declare.
372 * read.c (read_a_source_file): After doing an mri_pending_align,
373 adjust the line_label if there is one.
374 (s_space): Set mri_pending_align if an odd number of bytes were
379 * config/tc-d10v.h (md_do_align): Add this hook to call
380 d10v_cleanup() when a ".align" is detected. Fixes PR11487.
382 * config/tc-d10v.c (find_opcode): Correctly calculate
383 branch displacement when .aligns are present.
387 * config/tc-mn10200.c (md_relax_table): Define.
388 (md_convert_frag): Implement.
389 (md_assemble): Handle relaxable operands/instructions correctly.
390 (md_estimate_size_before_relax): Implement.
391 * config/tc-mn10200.h (TC_GENERIC_RELAX_TABLE): Define.
395 * config/tc-mips.c (append_insn): Give an error for jumps to a
397 (md_apply_fix): Make a branch to an odd address an error rather
400 * config/tc-mips.c (md_convert_frag): If the user explicitly
401 requested an extended opcode, pass warn as true to mips16_immed.
403 * config/tc-mips.c (mips16_ip): Handle a missing expression like
404 an explicit 0, so that explicitly extended instructions work
409 * ecoff.c (ecoff_build_symbols): Don't generate a local ECOFF
410 symbol for a common symbol.
414 Patch presumed to have been checked in awhile ago but wasn't.
416 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
417 (relax_segment, case rs_org): Move code inside braces. Move locals
418 target,after inside too.
419 (relax_segment, case rs_machine_dependent): Guts moved to ...
420 (relax_frag): New function.
421 Call md_prepare_relax_scan if defined.
425 * config/tc-m68k.c (m68k_ip): Reject pc-relative addresses for the
426 'p' operand specifier.
430 * config/tc-m68k.c (HAVE_LONG_BRANCH): New macro, returns true for
431 m68k family cpus which support long branch addressing modes.
432 (m68k_ip, md_convert_frag_1, md_estimate_size_before_relax,
433 md_create_long_jump): Use it.
437 * config/tc-mips.c (md_begin): Don't set SEC_ALLOC or SEC_LOAD for
438 the .reginfo or .MIPS.options section if configured for an
441 * config/tc-mips.c (md_begin): Don't set interlocks for
446 * read.c (read_a_source_file): Make sure the symbol ends with
447 whitespace before checking whether the next character is '='.
451 * config/tc-i386.c (sco_id): Moved from here...
452 * config/obj-elf.c (sco_id): ...to here. Adding the identifier
453 really is an SCO ELF specific thing, not just a SCO x86 specific
459 * configure.in (tic80-*-*): Don't require 'coff'.
460 * configure: Regenerate.
465 * read.c (emit_expr): Check for overflow of a negative value
467 * write.c (fixup_segment): Likewise.
468 * config/obj-coff.c (fixup_segment): Likewise.
470 * config/tc-m68k.c (struct label_line): Define.
471 (labels, current_label): New static variables.
472 (md_assemble): Mark current_label as text, and clear it.
473 (m68k_frob_label): New function.
474 (m68k_flush_pending_output): New function.
475 (m68k_frob_symbol): New function.
476 * config/tc-m68k.h (tc_frob_label): Define.
477 (md_flush_pending_output): Define.
478 (tc_frob_symbol): Don't warn, just call m68k_frob_symbol.
479 (tc_frob_coff_symbol): Likewise.
481 * read.c (read_a_source_file): When defining a macro in MRI mode,
482 don't add the symbol to the symbol table.
486 * config/tc-mn10300.c (tc_gen_reloc): Handle sym1-sym2 fixups
487 here since fixup_segment doesn't (linkrelax is set).
488 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
492 * config/tc-mn10200.c (md_assemble): Tweak fx_offset for pc-relative
497 * config/tc-hppa.c (struct hppa_fix_struct): Tweak fx_r_field's type
498 to avoid warnings with the native HP compiler.
499 (fix_new_hppa): Similarly for the r_type argument.
500 (pa_build_unwind_subspace, hppa_elf_mark_end_of_function): Enclose
501 in an #if OBJ_ELF to keep gcc -Wall quiet.
502 (md_apply_fix): Always initialize "result".
504 * config/tc-mn10200.c (md_assemble): Generate relocations.
508 * config/tc-m68k.c (s_even): Adjust the alignment of the current
513 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
514 pass on the new file hook.
516 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
517 !alpha_force_relocation, as local LITERALs can be adjusted to be
518 relative to the section.
522 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
524 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
525 aligned to odd boundary.
526 (tc_frob_coff_symbol): Define.
528 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
530 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
532 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
535 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
537 (md_assemble): If the size is 'B', set fx_signed.
538 (md_apply_fix_2): Use fx_signed when checking for overflow.
540 * write.h (struct fix): Add fx_signed field.
541 * write.c (fix_new_internal): Initialize fx_no_overflow and
543 (fixup_segment): Use fx_signed when checking for overflow.
544 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
545 fx_signed when checking for overflow.
549 * NOTES, NOTES.config: Removed. These are rarely, if ever,
550 updated, and all the useful information is in doc/internals.texi.
553 * read.c (read_a_source_file): Check for conditional operators
554 before doing an MRI pending alignment.
555 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
556 (tc_conditional_pseudop): Define.
557 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
558 * doc/internals.texi (CPU backend): Describe
559 tc_conditional_pseudoop.
562 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
564 (md_begin): Check m68k_rel32_from_cmdline before setting
566 (m68k_mri_mode_change): Likewise.
567 (md_longopts): Add --disp-size-default-16 and
568 --disp-size-default-32.
569 (md_parse_option): Handle new options.
570 (md_show_usage): Mention new options.
571 * doc/c-m68k.texi (M68K-Opts): Document new options.
574 * config/tc-m68k.c (m68k_index_width_default): New static
576 (m68k_ip): Use m68k_index_width_default to set the size of a base
577 register whose size was not given.
578 (md_longopts): Add --base-size-default-16 and
579 --base-size-default-32.
580 (md_parse_option): Handle new options.
581 (md_show_usage): Mention new options.
582 * doc/c-m68k.texi (M68K-Opts): Document new options.
584 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
586 * configure.in: Recognize mips-*-linux* target.
587 * configure: Rebuild.
589 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
590 work if valueT is only 32 bits.
592 * config/tc-mips.c: Throughout, check target_big_endian rather
594 (byte_order): Remove.
595 (mips_init_after_args): Remove.
596 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
597 (mips_init_after_args): Don't declare.
598 (tc_init_after_args): Don't define.
600 * config/tc-mips.h (tc_frob_after_relocs): Define if
602 (mips_elf_final_processing): Likewise.
603 (ELF_TC_SPECIAL_SECTIONS): Likewise.
608 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
609 sld.{h,w} and the operand is relocatable, adjust the adend by
610 shifting it right one bit.
615 * read.c (read_a_source_file): Check mri_pending_align after
616 checking for a macro. From Ronald F. Guilmette
619 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
621 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
622 add segment argument. If OBJ_ELF, treat a relocation against a
623 symbol in a linkonce section like a relocation against an external
625 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
629 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
633 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
634 includes config.h instead of host.h, tc.h instead of tp.h, and
635 targ-env.h instead of target-environment.h.
636 Also, obj-format.h includes targ-cpu.h instead of
639 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
640 * configure: Rebuild with autoconf.
641 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
642 (TARGET_FORMAT): Define to "coff-tic80".
643 * config/tc-tic80.c: New file for TIc80 support.
644 * config/tc-tic80.h: New file for TIc80 support.
649 * doc/as.texinfo (M): Mention explicitly that -M changes macro
654 * write.c (adjust_reloc_syms): If the fixup symbol has been
655 equated to an undefined symbol, convert the fixup to being against
656 the target symbol. Remove obsolete code handling a special case
661 * mpw-make.sed: Use NewFolderRecursive for installation.
665 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
669 * config/tc-d10v.c (md_assemble): Fix bug which caused
670 second instruction in a line to be case sensitize. PR11312
674 * config/tc-mn10200.c (mn10200_insert_operand): Don't
675 range check operands with MN10200_OPERAND_NOCHECK set.
676 (check_operand): Likewise.
680 * config/tc-mips.c: Undo part of last Friday's alignment changes.
681 (md_begin): Always align the text section to a four byte
683 (append_insn): Remove call to record_align.
685 * config/tc-mips.c (insn_label): Remove.
686 (struct insn_label_list): Define.
687 (insn_labels, free_insn_labels): New static variables.
688 (mips_clear_insn_labels): New static function.
689 (append_insn): Mark all mips16 text labels, and make them odd.
690 Handle all labels after emitting a nop, not just one. Call
691 mips_clear_insn_labels rather than just clearing insn_label.
692 (mips_emit_delays): Add insns parameter, and use it to decide
693 whether to mark mips16 labels. Handle all labels, not just one.
694 Force mips16 labels to be odd. Change all callers.
695 (mips16_immed): Don't check for an odd branch target.
696 (md_apply_fix): Don't check mips16 mode for a branch reloc.
697 (mips16_extended_frag): Ignore the low bit in a branch target.
698 (md_convert_frag): Likewise.
699 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
701 (mips_align, mips_flush_pending_output, s_cons): Likewise.
702 (s_float_cons, s_gpword): Likewise.
703 (s_align): Use insn_labels rather than insn_label.
704 (s_cons, s_float_cons, s_gpword): Likewise.
705 (mips_frob_file_after_relocs): New function.
706 (mips_define_label): Rewrite to add to insn_labels list.
707 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
708 * ecoff.c (ecoff_build_symbols): If the size of a function comes
709 out odd, increment it.
711 * config/tc-mips.c (append_insn): Only update prev_insn when not
712 reordering if place is NULL.
714 * config/tc-mips.c (mips16_ip): Check for a missing expression
715 when using the register indirect addressing mode.
719 * config/tc-mn10200.c (mn10200_insert_operand): Don't
720 check 24bit operands for overflow.
721 (check_operand): Likewise.
725 * doc/as.texinfo (Section): Document how to use the .section
726 pseudo-op for COFF and ELF.
730 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
734 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
735 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
736 store them. Adjust other RELAX_MIPS16 macros.
737 (RELAX_MIPS16_DSLOT): Define.
738 (RELAX_MIPS16_JAL_DSLOT): Define.
739 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
740 handling of whether previous instruction has a fixup. Set
741 prev_insn_reloc_type.
742 (mips_no_prev_insn): Clear prev_insn_reloc_type.
743 (mips16_extended_frag): Use the right base address for a PC
744 relative add or load.
745 (md_convert_frag): Likewise. If a PC relative add or load is
746 used, record the alignment for the section.
750 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
751 linkonce section into a reloc against the section symbol.
753 * config/tc-mips.c (mips16_macro): Remove nop instructions after
756 * config/tc-mips.c (md_begin): If configured for an embedded ELF
757 system, don't set the section alignment to 2**4.
758 (s_change_sec): Likewise.
759 (append_insn): Call record_alignment for the section.
760 (md_section_align): Don't align the section size for an embedded
765 * write.c (adjust_reloc_syms): Make sure that symbols are
766 resolved; expression symbols may have been skipped.
767 * config/obj-coff.c (fixup_segment): Likewise.
771 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
772 BFD_RELOC_24_PLT_PCREL relocation.
773 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
777 * config/tc-d10v.c (write_2_short): Remove code that called
778 parallel_ok() when the programmer specified parallel instructions.
782 * config/tc-mn10300.c (md_assemble): Update to handle endianness
785 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
786 * config/tc-mn10300.c (md_assemble): Likewise.
790 * config/tc-mips.c (append_insn): Make sure there is enough room
791 in a frag after a mips16 instruction to switch it with a jump
794 * config/tc-mips.c (mips16_extended_frag): Give an error for an
795 attempt to use a non absolute symbol in an extending frag.
799 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
800 * config/tc-mn10200.h: Likewise.
804 * app.c (do_scrub_chars): At the end of a C comment, pass space to
805 UNGET rather than PUT. Set old_state before setting state to -2.
807 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
808 when extending because the value is exactly maxtiny + 1.
810 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
811 arguments, and store them. Adjust other RELAX_MIPS16 macros.
812 (RELAX_MIPS16_USER_SMALL): Define.
813 (RELAX_MIPS16_USER_EXT): Define.
814 (mips16_small, mips16_ext): New static variables.
815 (append_insn): Pass mips16_small and mips16_ext to
817 (mips16_ip): Set mips16_small and mips16_ext.
818 (mips16_immed): Don't check mips16_autoextend.
819 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
821 * write.c (write_relocs): Print an error for an out of range
822 fixup, rather than calling abort.
824 * as.c (main): Unlink the output file if there are errors while
825 generating the fixups.
829 * config/tc-mips.c (mips16_extended_frag): Don't call
831 (md_convert_frag): Call resolve_symbol_value before calling
832 S_GET_VALUE, and don't add in the frag address.
834 * config/tc-mips.c (mips16_immed): Add file and line parameters,
835 and use them when reporting errors. Change all callers.
839 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
840 Remove '$' prefixing for registers.
844 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
845 is initialized before using it.
846 (d10v_cleanup): No longer uses its argument, so make it void.
848 * config/tc-d10v.h (d10v_cleanup): Change prototype.
852 * write.c (fixup_segment): Don't discard the symbol for a PC
853 relative fixup to an absolute symbol.
857 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
858 with multiple sections.
862 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
863 no-mips-16 to no-mips16.
864 (s_mipsset): Accept .set mips16 and .set nomips16.
868 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
869 argument, and check for +/- constant following the suffix, folding
870 it into the expression.
871 (ppc_elf_cons): Change ppc_elf_suffix calls.
872 (md_assemble): Ditto.
873 (shlib): Replace boolean mrelocatable with enumeration shlib.
874 (md_parse_option): Discriminate between PIC style shared libraries
876 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
881 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
882 change the comment characters.
884 * config/tc-ppc.c (comment_chars): Delete in favor of
886 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
888 (ppc_comment_chars): Select appropriate comment chars by default.
889 (msolaris): New flag for -m{,no-}solaris.
890 (md_parse_option): Recognize -K pic. Add support for
892 (md_show_usage): Update.
893 (md_begin): Do not set ELF flags if Solaris.
894 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
895 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
899 * as.c (main): Correct handling of flag_always_generate_output.
903 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
904 fx_offset, not fx_addnumber.
906 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
911 * config/tc-i386.c: Correct misspelling: balenced to balanced.
915 * config/tc-mips.c (md_section_align): Check for an alignment of
916 4, not an alignment of 16. Corrects August 7 patch.
920 * configure, conf.in: Rebuild with autoconf 2.12.
922 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
923 alignment is specified.
925 Add support for mips16 (16 bit MIPS implementation):
926 * config/tc-mips.c: Extensive additions for mips16 support, not
928 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
929 (insn_uses_reg): Change last parameter to an enum.
930 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
931 (md_relax_frag): Define as mips_relax_frag.
932 (mips_relax_frag): Declare.
933 (struct mips_cl_insn): Add use_extend and extend fields.
934 (tc_fix_adjustable): Define.
935 * config/obj-elf.h (S_GET_OTHER): Define.
936 (S_SET_OTHER): Define.
940 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
941 o and p operand specifiers.
945 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
946 (relax_segment, case rs_org): Move code inside braces. Move locals
947 target,after inside too.
948 (relax_segment, case rs_machine_dependent): Guts moved to ...
949 (relax_frag): New function.
950 Call md_prepare_relax_scan if defined.
951 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
956 * config/tc-mn10300.c (address_registers): Use '$' as register
957 prefix instead of '%'.
958 (data_registers, other_registers, md_assemble): Likewise.
960 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
961 (data_registers, other_registers, md_assemble): Likewise.
963 * config/tc-mn10300.c (md_assemble): Correctly determine the
964 correct location and type for each relocation.
965 (md_pcrel_from): Simplify.
969 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
970 converted into branch around branch.
974 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
975 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
977 * config/tc-mn10300.c (md_begin): Set linkrelax.
978 (md_assemble): Create fixups as needed.
979 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
983 * config/tc-d10v.c (parallel_ok): When automatically converting
984 serial ops to parallel, do not consider a branch as the first
989 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
993 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
995 (check_operand): Likewise.
999 * config/tc-d10v.c (parallel_ok): Branch and link instructions
1001 (write_2_short): Call parallel_ok to check whether two short
1002 instructions the user requested execute in parallel, can be
1007 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
1008 allow a branch and link in parallel with an exe instruction.
1012 * doc/c-d10v.texi: Add info on @word modifier.
1016 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
1017 operands are assumed to be 32bits. Use "bits" field to hold the
1018 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
1019 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
1022 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
1023 of a MN10300_OPERAND_SPLIT operand by operand->shift.
1025 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1026 MN10300_OPERAND_SPLIT.
1030 * config/tc-mn10300.c (md_assemble): Insert operands into
1031 the extension part of the instruction if necessary.
1032 (mn10300_insert_operand): Accept pointer to extension word
1033 argument. Make insn a pointer argument too. Return type
1034 is now void. All callers changed.
1038 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1039 repeated register operands.
1043 * doc/as.texinfo: Added section on reporting bugs.
1045 * config/tc-alpha.c: Change uses of void * to PTR. Change the
1046 alpha_macro emit field to expect a const argument, and change the
1047 arg field to be const. Fix some spacing to follow the GNU
1052 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
1053 (pca56) and 21264 (ev6) cpus.
1054 (md_apply_fix): Private relocation types are now negative.
1055 (alpha_force_relocation): Likewise.
1056 (tc_gen_reloc): Likewise.
1057 (emit_insn): Likewise.
1058 (emit_ldXu): Do the right thing when the hardware can do byte insns.
1059 (emit_stX): Likewise.
1060 (emit_sextX): Likewise.
1064 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
1065 a symbol found in a reloc.
1067 * symbols.c (resolve_symbol_value): Improve the error message if
1068 an undefined symbol is used in an expression.
1072 * doc/internals.texi: Rewrite, and add a lot of documentation.
1073 * doc/Makefile.in (internals.info): New target.
1078 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
1079 against weak symbols.
1083 * config/tc-v850.c (md_assemble): Don't lose for relaxable
1086 * config/tc-v850.c (md_convert_frag): Make sure we insert the
1087 fixup at the right address within the frag.
1089 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
1090 to an absolute value, instead increment it as needed.
1092 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
1093 * config/tc-v850.c: Fix some indention problems.
1094 (md_relax_table): Define for D9->D99 branch displacement
1096 (md_convert_frag): Do something useful instead of aborting.
1097 (md_estimate_size_before_relax): Likewise.
1098 (md_assemble): Note if the matching instruction has a relaxable
1099 operand. If it does, allocate frag with frag_var and don't
1105 * config/tc-d10v.h (md_cleanup): New function. This is needed to
1106 write out any buffered instructions when a ".end" is found.
1110 * read.c (read_a_source_file): New hook md_cleanup().
1114 * write.c (fix_new_exp): Use make_expr_symbol to build an
1115 expression symbol for a complex fixup.
1120 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
1121 hacks to improve parsing of complex hi, lo, zda, etc
1123 (md_assemble): Don't demand and eat a trailing ')' after finding
1124 a v850 relocation prefix. Sign extend the constant in a
1125 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
1127 (parse_cons_expression_v850): Don't eat a trailing ')' after
1128 finding a v850 relocation prefix.
1130 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
1131 (TC_CONS_FIX_NEW): Likewise.
1132 * config/tc-v850.c (parse_cons_expression_v850): New function.
1133 (cons_fix_new_v850): Likewise.
1135 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
1140 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
1141 for an unknown relocation type.
1146 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
1151 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
1152 differently for movea & sst/sld insns.
1157 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
1162 * obj.h (struct format_ops): Add frob_file_after_relocs field.
1163 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
1164 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
1165 frob_file_after_relocs field.
1166 * config/obj-elf.c (elf_format_ops): Likewise.
1167 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
1168 including obj-elf.h.
1172 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
1173 specially if not ELF.
1174 (md_apply_fix): Handle BFD_RELOC_64.
1175 (tc_gen_reloc): Handle BFD_RELOC_64.
1177 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
1178 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
1181 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
1182 Don't check S_GET_OTHER.
1186 * config/tc-mips.c (mips_ip): Accept an odd floating point
1187 register with l.s or s.s.
1189 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
1191 (obj_aout_type): New static function.
1195 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
1200 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
1201 and sdaoff expressions.
1203 * write.c (fixup_segment): Don't add symbol value to addend if
1204 TC_V850 and OBJ_ELF.
1205 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
1208 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
1209 (md_pcrel_from_section): Likewise.
1210 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
1214 * config/tc-v850.c (md_pcrel_from): Delete unused function.
1215 (md_pcrel_from_section): New function.
1216 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
1221 * config/tc-mips.c (load_register): Add cast to offsetT when using
1226 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
1227 (elf_frob_file_after_relocs): ... here. New function.
1228 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
1229 * write.c (write_object_file): Call *frob_after_relocs after the
1230 call to write_relocs.
1232 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
1234 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
1235 symbol, as this messes up linking. Instead, expand the recursive call
1236 inline and change up the appropriate bits to get the 0x8000 offset
1237 in the reloc addend.
1241 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
1242 two symbols in the same segment to be adjusted.
1244 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
1245 * configure: Rebuild.
1249 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
1251 (ppc_elf_validate_fix): Ditto.
1252 (md_assemble): Ditto.
1254 (ppc_pe_section): Ditto.
1255 (ppc_frob_symbol): Ditto.
1259 * config/mn10300.c (md_assemble): Pass an extra shift count
1260 to mn10300_insert_operand based on the opcode format.
1261 (mn10300_insert_operand): Accept and use extra shift count
1264 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
1265 formats rather than hard-coded constants.
1267 * config/tc-mn10300.c (md_assemble): Format D5 instructions
1268 are 7 bytes long. Write out instructions in big-endian format.
1272 * config/tc-mn10300.c (md_assemble): Tweak further so
1273 that all instructions are parsed correctly.
1277 * as.h: Include libiberty.h.
1278 (xmalloc, xrealloc): Don't declare.
1279 * as.c: Don't include libiberty.h.
1280 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
1281 * config/tc-mips.c: Likewise.
1282 * messages.c: Likewise.
1283 (xstrerror): Don't declare.
1284 * xmalloc.c: Remove.
1288 * config/tc-mn10300.h (pre_defined_registers) Remove.
1289 (system_registers, cc_names): Likewise.
1290 (address_registers, data_registers, other_registers): New register
1292 (register_name, system_register_name, cc_name): Remove.
1293 (mn10300_reloc_prefix): Likewise.
1294 (data_register_name): New function.
1295 (address_register_name, other_register_name): Likewise.
1296 (md_assemble): Rough cut at parsing operands. Remove lots of
1298 (md_apply_fix3): Disable for now.
1302 * config/tc-m68k.c (select_control_regs): New function, extracted
1303 out of m68k_init_after_args.
1304 (m68k_init_after_args): Use it.
1305 (mri_chip): Use it here as well to update set of allowed control
1310 * config/obj-elf.c (elf_begin): New function.
1311 (obj_elf_section): Add the section symbol to the symbol table.
1312 * config/obj-elf.h (obj_begin): Define.
1313 (elf_begin): Declare.
1314 * as.c (perform_an_assembly_pass): Call obj_begin if it is
1319 * config/obj-coff.c (fixup_segment): Subtract the section address
1320 from a PC relative reloc if TC_M68K.
1324 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
1325 .uaxword available even if not OBJ_ELF.
1326 (md_atof): Remove unused local variable wordP.
1330 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
1331 for Matsushita MN10x00 support.
1332 * configure.in: Recognize mn10x00-*-*
1333 * configure: Rebuilt.
1337 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
1338 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
1339 (load_expression): track clobbering of base reg before jmp/jsr.
1340 (s_alpha_file): pass case_hack flags and source filename via
1341 symbol table to bfd.
1342 * tc-alpha.h (TC_CONS_FIX_NEW): define
1346 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
1347 * configure: Rebuild.
1351 * read.c (s_macro): Warn if a macro has the same name as a
1353 (s_space): In m68k MRI mode, align to a word boundary.
1354 * macro.c (define_macro): Add namep parameter. Change all
1356 * macro.h (define_macro): Update declaration.
1358 * as.c (show_usage): Print bug report address.
1359 (parse_args): Change version printing to match current GNU
1361 * gasp.c (show_usage): Print bug report address.
1362 (main): Change version printing to match current GNU standards.
1364 * config/tc-m68k.c (init_table): Correct access control unit
1367 * config/tc-alpha.c: Add some static function prototypes.
1368 (alpha_macros): Move to top of file. Make static.
1369 (alpha_num_macros): Move to top of file.
1374 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
1380 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
1381 and just call sprintf_vma.
1385 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
1389 * read.c (float_cons): Call md_flush_pending_output if it is
1394 * config/tc-d10v.c (md_operand): Created. Allows operands to
1396 * config/tc-d10v.h (md_operand): Undefined.
1400 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
1401 (md_assemble): A fixup width of '3' means a 1 byte reloc.
1405 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
1406 reloc for the i960 for a reloc in the same section. This undoes
1407 one of the two changes made Aug 19.
1411 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
1412 symbols to the position of the debugging information.
1416 * expr.c (expr): Always use unsigned right shifts for >>.
1420 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
1421 slots when processing BL fixups.
1423 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
1424 on the first half of the instruction.
1428 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
1429 stabs expression, rather than giving an error.
1431 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
1436 * config/tc-mips.c (append_insn): Fill in the value for a constant
1437 jump, rather than creating a reloc.
1441 * config/tc-mips.c (append_insn): Don't swap an instruction which
1442 sets a condition code with an instruction which uses a condition
1444 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
1447 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
1448 (mips_ip): Give a better error message if the ISA level is wrong.
1449 (md_parse_option): Recognize -mcpu=[v][r]5000.
1453 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
1455 (load_register): Ensure hi32 bits are not lost during lo32bit
1456 processing. Fix shift offset that was overflowing into the next
1457 instruction field. Add code to generate shorter sequences for
1458 constants with a single contiguous seqeuence of ones.
1462 * config/tc-d10v.c (d10v_dot_word): New function to support
1463 "@word" with the word pseudo-op.
1464 (md_apply_fix3): Cleanup and changes to support correct sizes
1465 for 16 and 18-bit relocs.
1469 * configure.in (sparc-*-aout): Set `em'.
1470 * configure: Regenerated.
1471 * config/te-sparcaout.h: New file.
1472 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
1473 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
1474 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
1475 (SPECIAL_CASE_{SETSW,SETX}): Define.
1476 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
1477 (md_begin): Delete setting of `target_big_endian'.
1478 (output_insn): New function.
1479 (md_assemble): Rewrite. Add `setx' support.
1480 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
1482 (md_atof): Add little endian support.
1483 (md_number_to_chars): Likewise.
1484 (md_apply_fix): Likewise.
1485 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
1486 (md_parse_option): Likewise.
1487 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
1491 * ecoff.c (ecoff_new_file): New function.
1492 * ecoff.h (ecoff_new_file): Declare.
1493 * config/obj-ecoff.h (obj_app_file): Define.
1497 * config/tc-alpha.c (load_expression): Bias the .lit8 section
1498 symbol by 32k so that our 16-bit signed offset can address the
1503 * config/tc-mips.c (load_register): Remove unused variable tmp.
1507 * config/tc-mips.c (load_register): Remove unnecessary code that
1508 was causing the high 32bits of 64bit constants to be lost.
1512 * config/tc-d10v.c: Added changes to support function
1513 pointers and "@word" syntax.
1518 * config/tc-v850.c: Remove commented out and #if 0'd code.
1519 (v850_reloc_prefix): Provide prototype.
1520 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
1522 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
1523 (md_assemble): Add default to case statement.
1527 * config/tc-v850.c (md_assemble): Compute size of the instrction
1530 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
1535 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
1536 routines to fetch/store the updated instruction from/to memory.
1537 (v850_insert_operand): If the operand has a specialized insert
1542 * config/tc-v850.c (reg_name_search): Align calling convention to
1543 be like identical function found in tc-ppc.c.
1544 (get_reloc): Removed.
1545 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
1546 (md_assemble): emit fixups.
1547 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
1549 (md_apply_fix3): handle fixups/relocs.
1550 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
1556 * configure.in (sh-*-elf*): New target.
1557 * config/tc-sh.h (TARGET_ARCH): Define.
1558 (WORKING_DOT_WORD): Define.
1559 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
1560 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
1561 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
1562 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
1563 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
1564 (SUB_SEGMENT_ALIGN): Likewise.
1565 (RELOC_32): Don't define.
1566 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
1567 (target_big_endian): Declare if OBJ_ELF.
1568 (TARGET_FORMAT): Define if OBJ_ELF.
1569 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
1571 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
1572 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
1573 (struct sh_count_relocs): Define.
1574 (sh_count_relocs): New static function, broken out of
1575 sh_frob_file. Add BFD_ASSEMBLER code.
1576 (sh_frob_section): Likewise.
1577 (sh_frob_file): Call sh_frob_section.
1578 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
1579 call section_symbol rather than seg_info (seg)->dot.
1580 (md_section_align): Add OBJ_ELF version.
1581 (SWITCH_TABLE_CONS): Define.
1582 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
1583 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
1584 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
1586 (struct reloc_map): Define if not BFD_ASSEMBLER.
1587 (coff_reloc_map): Likewise.
1588 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
1589 (tc_gen_reloc): New function if BFD_ASSEMBLER.
1590 * write.c (write_relocs): Ifdef out fx_where test which triggers
1591 inappropriately for SH ELF.
1592 (write_object_file): Call tc_frob_file_before_adjust and
1593 obj_frob_file_before_adjust if they are defined.
1595 * write.c (write_object_file): Use BFD_RELOC_16, not
1596 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
1598 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
1603 * config/tc-d10v.c (find_opcode): Fix problem with calculating
1604 branch sizes in across sections.
1609 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
1614 * config/tc-v850.c (md_begin): Deal with end of opcode
1620 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
1621 the wrong opcode for cases like st2w where there are many forms
1622 of the same instruction.
1626 * expr.c (operand): If md_parse_name is defined, call it before
1627 calling symbol_find_or_make.
1628 * config/tc-ppc.h (md_parse_name): Define.
1629 (ppc_parse_name): Declare.
1630 * config/tc-ppc.c (reg_name_search): Add regs and regcount
1632 (register_name): Update call to reg_name_search.
1633 (cr_operand): New static variable.
1634 (cr_names): New static const array.
1635 (ppc_parse_name): New function.
1636 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
1637 cr_operand before calling expression.
1641 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
1642 hppa_gen_reloc_type call.
1646 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
1647 on addresses. Improved error messages.
1648 * doc/c-d10v.texi: Added docs for register pairs.
1652 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
1657 * ecoff.c (init_file): Initialize fMerge to 1.
1658 (add_file): Restore old file merging code, but only merge files if
1660 (ecoff_directive_loc): Clear fMerge field of current file.
1661 (ecoff_generate_asm_lineno): Likewise.
1665 * doc/c-d10v.texi: Fix typo.
1670 * config/tc-v850.c (md_assemble): Correct bit masking for
1671 hi and lo expressions.
1673 * config/tc-v850.c (md_assemble): Rough cut at demanding
1674 "ep" or "r30" in sst and sld instructions.
1675 (md_apply_fix3): Don't abort. Just warn that we don't
1678 * config/tc-v850.c (CC_NAME_CNT): Define.
1679 (cc_name): New function.
1680 (md_assemble): Handle V850_OPERAND_CC correctly.
1682 * config/tc-v850.c (md_assemble): Don't forget to initialize
1685 * config/tc-v850.c (reg_name_search): Generalize to search
1686 any given register table.
1687 (register_name): Pass appropriate table and size to reg_name_search.
1688 (system_register_name): New function.
1689 (SYSREG_NAME_CNT): Define.
1690 (md_assemble): Handle operands which are system registers.
1692 * config/tc-v850.c (md_assemble): If we find a register, but the
1693 opcode doesn't want a register, then we don't have a match.
1694 (md_assemble): Get size of the instruction from the opcode table.
1699 * configure.in: Set and substitute HLDENV.
1700 * configure: Rebuild.
1701 * Makefile.in (HLDENV): New variable.
1702 (as.new): Use $(HLDENV).
1704 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
1705 if an array dimension is not known.
1709 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
1710 * doc/c-d10v.texi: Cleanup.
1715 * config/tc-v850.c: New file.
1716 * config/tc-v850.h: New file.
1717 * configure (v850-*-elf): New target.
1718 * configure.in (v850-*-elf): New target.
1723 * doc/c-d10v.texi: New file.
1724 * doc/all.texi: Added D10V stuff.
1725 * doc/as.texinfo: Added D10V stuff.
1729 * config/tc-d10v.c: All references to defined symbols should
1730 now use the optimal instruction. .float and .double now work.
1734 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
1735 section address for the i960 as is done for the i386.
1739 * mpw-config.in: Add wildcards for config matching, add mips-*-*
1740 case, forward-include bfd/elf-bfd.h.
1744 * config/tc-d10v.c: Add additional information to the opcode
1745 table to help determinine which instructions can be done
1750 * config/tc-arm.c: Major changes to add Thumb support, with lots
1752 Reverted to INSN_SIZE macro, rather than insn_size variable.
1753 (insns): Added ARM "bx" instruction support.
1754 (tinsns): Added Thumb instruction definition structure.
1755 (arm_tops_hsh): Added hash structure for Thumb opcodes.
1756 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
1757 (opcode_select,s_arm,s_thumb,s_code): Added.
1758 (decode_shift): Allow upper-case RRX.
1759 (do_ldst): Simpler halfword support.
1760 (do_ldmstm): Improved.
1761 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
1762 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
1763 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
1764 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
1765 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
1767 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
1768 (md_parse_option): Add support for -mthumb.
1769 (md_show_usage): Updated to reflect new command line option.
1770 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
1771 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
1772 fragment information to be held.
1776 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
1777 non-null, but is a constant.
1778 (fix_new_arm): Call make_expr_symbol to make the expression symbol
1779 so that error reporting will work correctly.
1783 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
1784 against weak symbols.
1788 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
1789 (ppc_force_relocation): Declare if OBJ_XCOFF.
1790 * config/tc-ppc.c (ppc_force_relocation): New function if
1795 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
1799 * config/tc-d10v.c: Fix problem with relocs.
1803 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
1804 with nops if not in data_section or bss_section.
1808 Add support for openVMS/Alpha.
1809 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
1810 * config/obj-evax.c: New file.
1811 * config/obj-evax.h: New file.
1812 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
1814 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
1815 defined. Add case for bfd_target_evax_flavour.
1816 * config/vms-a-conf.h: New file.
1817 * conf-a-gas.com: New file.
1818 * configure.in: Add target alpha-*-*vms*.
1819 * configure: Rebuild.
1820 * makefile.vms: New file.
1821 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
1822 Don't call ffs on openVMS/Alpha.
1826 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
1827 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
1829 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
1830 (NEED_DECLARATION_SBRK): New macro.
1831 * configure, conf.in: Rebuild.
1832 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
1833 (strstr): Declare if NEED_DECLARATION_STRSTR.
1834 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
1838 * symbols.c (resolve_symbol_value): Handle addition or subtraction
1839 by a constant before entering the main switch. Reject attempts to
1840 apply an arithmetic function to non-absolute symbols, except for
1841 the special case of subtraction of two symbols in the same
1844 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
1845 not to more than a 16 byte boundary.
1847 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
1849 (tc_gen_reloc): If out of memory call as_fatal rather than
1850 assert. If no howto found, call as_bad_where rather than
1851 as_fatal. Change the error message slightly. Set howto to a
1852 non-NULL value in order to keep going.
1856 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
1860 * config/tc-h8300.c (get_specific): New operand "size" derived
1861 from ".b", ".w" and ".l" extensions. All callers changed. If
1862 the base instruction has no operands, then use the size to
1863 determine which specific instruction to use.
1867 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
1871 * config/tc-mips.c (md_section_align): Don't change addr if
1876 * config/tc-hppa.c: Revert yesterday's changes.
1880 * config/tc-d10v.c: Disable range checking on 16-bit values.
1884 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
1885 not just the ones that don't match.
1889 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
1893 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
1894 pre-cursor to adding Thumb support. Also added cpu_variant flag
1895 information to each of the asm_flg structures.
1896 (md_parse_option): Updated ARM7 parsing to allow 't' for
1897 thumb/halfword support, aswell as 'm' for long multiply.
1898 (md_show_usage): Updated help message.
1899 (md_assemble): Check that instruction flags are applicated to the
1900 current cpu variant.
1901 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
1902 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
1903 signextension instructions.
1904 (do_ldst): Generate halfword and signextension variants if
1905 mnemonic flags match.
1906 (ldst_extend): Do not allow shifts in the offset field of halfword
1907 or signextension instructions.
1908 (validate_offset_imm): Provide check on halfword and signextension
1910 (add_to_lit_pool): Merge identical literal pool values.
1914 * config/tc-hppa.c (selector_table): Add 'E' selector.
1915 (cons_fix_new_hppa): Don't coke on e_esel.
1916 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
1917 to help generate exception handling tables.
1918 (md_apply_fix): Don't try to apply fixups with an e_esel
1920 (hppa_fix_adjustable): Fixups with e_esel selectors
1925 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
1930 * config/tc-d10v.c: Added lots of error checking. Added hacks
1931 to support accumulator shifts.
1935 * symbols.c (S_SET_EXTERNAL): Let .weak override.
1936 (S_CLEAR_EXTERNAL): Likewise.
1937 (S_SET_WEAK): Remove error; just let .weak override.
1941 * config/tc-d10v.c (md_assemble): Now handles multiline
1946 * config/tc-d10v.c: Fix packaging bug. Added range checking.
1947 Added kludge for divs instruction. Fixed minor problem with
1948 multiple text sections.
1949 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1953 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
1954 addresses to be right-shifted by 2.
1958 * config/tc-d10v.c: Many changes to get relocs working.
1959 (register_name): No longer creates a symbol for register names.
1960 (pre_defined_registers): moved to opcodes/d10v-opc.c.
1961 (d10v_insert_operand): Now works correctly for either container.
1962 * config/tc-d10v.h (d10v_cleanup): Declare.
1966 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
1967 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
1971 * tc-alpha.c: Patches to track current minimum alignment to reduce
1972 the number of fragments created with frag_align.
1973 (alpha_current_align): New static variable.
1974 (s_alpha_text): Reset alignment to 0.
1975 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
1976 (s_alpha_stringer, s_alpha_space): New functions.
1977 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
1978 (alpha_cons_align): New function to replace both of them.
1979 (emit_insn): Only align if alpha_current_align is less than 2;
1980 reset alpha_current_align to 2.
1981 (s_alpha_gprel32): Likewise.
1982 (s_alpha_section): New function. Basically duplicate the other
1983 alpha section change hooks. Only define for ELF.
1984 (s_alpha_float_cons): Simplify alignment handling.
1985 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
1986 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
1987 Don't define the s_alpha_cons pseudo-ops. Do define
1988 s_alpha_stringer and s_alpha_space pseudo-ops.
1989 (alpha_align): Skip if less than current default alignment. Set
1991 * tc-alpha.h (md_flush_pending_output): Remove.
1992 (md_cons_align): Add.
1994 * tc-alpha.c: Add oodles of function description comments.
1995 (md_bignum_to_chars): Remove; there are no callers.
1996 (md_show_usage): Mention some more variants.
2001 * configure.in (sparc-*-sysv4*): New target.
2002 * configure: Rebuild.
2004 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
2005 uaxword to use s_uacons.
2006 (sparc_no_align_cons): New static variable.
2007 (s_uacons): New static function.
2008 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
2011 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
2013 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
2014 5 with regard to shared libraries.
2015 * configure: Rebuild.
2017 * config/tc-m68k.c (m68k_ip): Use the correct length when
2018 allocating space for the unsupported architecture error message.
2022 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
2026 * config/tc-d10v.c: New file.
2027 * config/tc-d10v.h: New file.
2028 * configure (d10v-*-elf): New target.
2029 * configure.in (d10v-*-elf): New target.
2033 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
2037 * config/tc-alpha.c (alpha_align): Change fill parameter
2038 to a pointer. Take NULL as 0 or nop depending on section. Change
2040 (s_alpha_align): Rename local variables.
2042 * doc/as.texinfo (.align): Document action of omitted
2047 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
2048 when an unsupported PC relative reloc is seen, rather than calling
2051 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
2052 Instead, check state when '#' comment is seen.
2056 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
2058 (tc_gen_reloc): If fixup was changed to be PC relative, change
2059 reloc type accordingly. Use name of reloc in error message.
2061 * as.h: Don't define const or volatile.
2062 * flonum.h: Don't define const.
2064 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
2065 if fx_pcrel is set. Correct setting the addend case in the
2066 OBJ_ELF case (from Andreas Schwab
2068 (md_show_usage): Correct -mfc5200 to -m5200.
2072 * doc/c-m68k.texi: Document -m5200 flag.
2073 * doc/as.texinfo: Likewise.
2075 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
2080 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
2081 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
2085 * Makefile.in (VERSION): Set to cygnus-2.7.1.
2087 * Released binutils 2.7.
2091 * config/tc-mips.c (mips_ip): Only perform range check when
2092 dealing with O_constant expressions.
2096 * m68k-parse.h (m68k_register): Add new coldfile control
2099 * config/tc-m68k.c (mcf5200_control_regs): New variable,
2100 array of control registers for the coldfire.
2101 (cpu_of_arch): Added mcf5200.
2102 (archs): Added mcf5200.
2103 (init_table): Add new control registers.
2104 (m68k_ip): Added support for new control registers.
2105 (m68k_init_after_args): Likewise.
2107 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
2111 * read.h (is_it_end_of_statement): Declare.
2112 * read.c (is_it_end_of_statement): Remove declaration.
2114 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
2116 (md_assemble): Fix handling of @l with an unsigned constant. Add
2117 default case to reloc switch.
2119 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
2122 * ecoff.c (last_lineno): New static variable.
2123 (add_procedure): Set last_lineno.
2124 (ecoff_directive_loc): Likewise.
2125 (ecoff_generate_asm_lineno): Likewise.
2126 (ecoff_fix_loc): New function.
2127 * ecoff.h (ecoff_fix_loc): Declare.
2128 * config/tc-mips.c (append_insn): When inserting nops, and using
2129 ECOFF debugging, call ecoff_fix_loc.
2133 * config/tc-h8300.c (build_bytes): If an operand type is
2134 marked as SRC_IN_DST retrieve it from the "destination" op.
2138 * configure.in (arm-*-riscix*): Set emulation to riscix.
2139 * configure: Rebuild.
2140 * config/te-riscix.h: New file to define TE_RISCIX.
2142 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
2146 * Makefile.in (config.status): Just run config.status as other
2151 * mpw-config.in (TARGET_OS): Add definition to conf.
2155 * config/tc-mips.c (append_insn): Parenthesize
2156 cop_interlocks expressions.
2160 * listing.c (listing_print): Close the listing file if it is not
2161 stdout. Close the other files opened for the listing.
2163 * config/tc-sparc.h (md_cons_align): Define.
2164 (sparc_cons_align): Declare.
2165 (HANDLE_ALIGN): Define.
2166 (sparc_handle_align): Declare.
2167 * config/tc-sparc.c (sparc_cons_align): New function.
2168 (sparc_handle_align): New function.
2169 * read.c (cons_worker): Call md_cons_align if it is defined.
2171 * as.h (struct frag): Add fr_file and fr_line fields.
2172 * frags.c (frag_new): Set fr_file and fr_line.
2173 (frag_var): Likewise.
2174 (frag_variant): Likewise.
2176 * as.h (struct frag): Remove unused align_mask and align_offset
2179 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
2184 * configure.in (mips-*-osf*): New target.
2185 * configure: Rebuild.
2187 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
2191 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
2192 between co-processor comparisons and branches for the VR4300.
2196 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
2197 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
2199 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
2200 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
2201 includedir): Use autoconf set values.
2206 * listing.c (listing_eject): Don't do anything if listing is 0.
2207 (listing_list): Likewise.
2208 (listing_source_line): Likewise.
2209 (listing_title): Don't save title if listing is 0.
2210 (listing_source_file): Check listing rather than listing_tail.
2212 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
2213 using shared libraries.
2214 * configure: Rebuild.
2218 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
2219 absolute expression if a relocation type was specified.
2223 * configure.in: Add support for *-*-rtems* configurations.
2224 * configure: Rebuild.
2228 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
2229 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
2230 opcodes library for alpha.
2231 * configure: Rebuild with autoconf 2.10.
2232 * config/tc-alpha.c: Substantial rewrite to add ELF support and
2233 use new opcode table.
2234 * config/tc-alpha.h (md_undefined_symbol): Don't define.
2235 (LOCAL_LABEL): Define differently if OBJ_ELF.
2236 (FAKE_LABEL_NAME): Define if OBJ_ELF.
2237 * config/alpha-opcode.h: Remove.
2238 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
2239 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
2240 include/opcode/alpha.h rather than config/alpha-opcode.h.
2244 * config/obj-aout.c (obj_emit_relocations): Give an error if the
2245 relocation symbol was not resolved.
2246 * config/obj-coff.c (do_relocs_for): Likewise.
2248 * write.c (adjust_reloc_syms): Refetch the symbol section after
2249 calling S_GET_VALUE, since it may have changed.
2251 * expr.c (struct expr_symbol_line): Define.
2252 (expr_symbol_lines): New static variable.
2253 (make_expr_symbol): Add entry to expr_symbol_lines.
2254 (expr_symbol_where): New function.
2255 * expr.h: Use extern on function declarations.
2256 (expr_symbol_where): Declare.
2257 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
2258 rather than printing the meaningless name of an expression
2263 * config/tc-i386.c (md_number_to_chars): Deleted.
2264 * config/tc-i386.h (md_number_to_chars): New macro.
2266 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
2268 (load_symbol_address, load_expression): Use build_mem.
2269 (build_operate): New function.
2270 (emit_addq_r): Use it.
2274 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
2278 * config/tc-vax.h (NOP_OPCODE): Define.
2282 * config/obj-vms.h (S_IS_COMMON): Define.
2283 (S_IS_LOCAL): Check for \002 as well as \001.
2284 (LONGWORD_ALIGNMENT): New macro.
2285 (SUB_SEGMENT_ALIGN): Use it.
2289 * config/vms-conf.h: Reconcile with conf.in.
2293 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
2296 * doc/as.texinfo: Document that any number of hex digits can
2299 * as.c (struct defsym_list): Define.
2300 (defsyms): New static variable.
2301 (parse_args): Just put --defsym arguments on defsyms list, rather
2303 (main): Define defsyms after output file is created.
2305 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
2308 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
2309 insn_error rather than calling check_absolute_expr.
2311 * as.c (emulation_name): Remove unused static variable.
2312 (default_emul_bfd_name): Add return NULL to avoid warning.
2313 * ecoff.c (ecoff_stab): Remove unused variables name and
2315 * frags.c (frag_new): Remove unused variable tmp.
2316 * hash.c (hash_grow): Parenthesize + within <<.
2317 (hash_print_statistics): Use %lu, not %d, to print unsigned
2319 * messages.c: Include "libiberty.h".
2320 (fprint_value): Add cast to avoid printf warning.
2321 (sprint_value): Likewise.
2322 * read.c: Include "ecoff.h".
2323 (emit_expr): Add casts to avoid printf warnings.
2324 * read.h: Use extern for function declarations.
2325 (pop_insert): Declare.
2326 * stabs.c: Include "ecoff.h".
2327 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
2328 former_last_fragP, and new_fragP.
2329 * subsegs.h (subsegs_print_statistics): Declare.
2330 * symbols.c (debug_verify_symchain): Change macro to discard
2332 * write.c (dump_section_relocs): Likewise.
2333 * write.h: Use extern for function declarations.
2334 (write_print_statistics): Declare.
2335 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
2337 * config/e-mipself.c (mipself_bfd_name): Likewise.
2338 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
2340 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
2341 emit relocations against external symbols.
2343 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
2344 message if bfd_reloc_type_lookup fails, rather than calling
2347 * config/tc-alpha.c (alpha_force_relocation): Add
2348 BFD_RELOC_12_PCREL to switch.
2352 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
2353 TE_PE (Lfoo, not .Lfoo).
2357 * read.c (s_fill): Don't warn about a zero repeat count.
2359 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
2360 coprocessor register.
2362 * config/tc-i386.c (md_assemble): When checking the size of a
2363 register to set the size of an instruction, do a bitwise and with
2364 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
2369 * config/tc-h8300.c (parse_reg): Tweak error messages.
2370 (build_bytes): Likewise.
2371 (skip_colonthing): Handle :32 suffix.
2372 (get_specific): Promote L_24 to L_32 if it makes a match.
2373 Don't always promote L_8 to L_16.
2374 (do_a_fix_imm): Clean up L_32 and L_24 handling.
2376 * config/tc-h8300.c (Smode): New variable.
2377 (h8300hmode): Turn off Hmode.
2378 (h8300smode): New function. Turn on Smode and Hmode.
2379 (md_pseudo_table): New ".h8300s" pseudo-op.
2380 (parse_reg): Handle "exr" register.
2381 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
2382 Handle "mach" and "machl" operands for ldmac.
2383 (get_specific): Handle "stm.l" and "ldm.l".
2384 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
2385 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
2390 * doc/as.texinfo: Reorder chapter of machine dependent options so
2391 that it is sorted by chip name.
2393 * doc/as.texinfo: Use consistant spelling of Vax.
2394 * doc/c-vax.texi: Likewise.
2398 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
2400 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
2401 (md_apply_fix): Likewise.
2402 (pa_try): New function.
2403 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
2407 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
2408 (pa_level): New function.
2412 * listing.c (listing_newline): Don't do anything if listing is 0.
2416 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
2417 symbol_create, not symbol_new.
2421 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
2422 absolute symbol unless TC_FORCE_RELOCATION returns true.
2424 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
2426 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
2427 the symbol to a pointer. Don't set previous_file_symbol.
2428 Simplify symbol list rearrangement.
2429 (coff_frob_symbol): Don't do anything with C_FILE symbols.
2430 (coff_adjust_symtab): Don't check previous_file_symbol.
2434 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
2436 (md_pseudo_table): Add ppc_elf_lcomm.
2440 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
2441 d0{24:1},d0' works without an immediate prefix on the bit numbers.
2442 (md_begin): Add digits to alt_notend_table.
2443 (md_parse_option): Make s a const pointer.
2445 * config/tc-sparc.c (md_pseudo_table): Add "empty".
2446 (s_empty): New static function.
2448 * config/obj-coff.c (struct filename_list): Only define if not
2450 (filename_list_head, filename_list_tail): Likewise.
2451 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
2452 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
2453 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
2454 have a section of N_ABS rather than N_DEBUG. If we do a merge,
2455 remove the new symbol from the list.
2456 (obj_coff_endef, both versions): Call tag_insert even if there is
2457 an old symbol with the same name, if the old symbol does not
2459 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
2460 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
2461 symbol that will be moved to the end of the symbol list.
2462 (coff_adjust_section_syms): Always call section_symbol for .text,
2464 (coff_frob_section): Likewise. Also, remove unused variable
2467 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
2468 manipulating frags directly.
2469 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
2472 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
2473 whether the next character is '*' before checking whether we are
2474 at the start of a line. Permit LINE_COMMENT_START to start a
2475 comment in state 1 (seen some whitespace) as well, to match the
2478 * gasp.c (do_align): Permit a fill value for .align.
2482 * read.c (next_char_of_string): Warn if a newline is seen in the
2483 middle of a string. Call bump_line_counters when appropriate.
2487 * symbols.c (colon): Use LOCAL_LABEL.
2491 * Makefile.in (install): Don't check to see if tooldir exists.
2492 Make $(tooldir) and $(tooldir)/bin.
2496 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
2501 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
2502 insn_error. Don't put the string "ERROR" in insn_error. Set
2503 insn_error rather than calling as_warn for an unsupported opcode.
2507 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
2508 before permitting -64.
2509 * output-file.c (output_file_create): Remove duplicate
2514 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
2515 section rather than a .reginfo section.
2516 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
2519 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
2520 bit value with the high bit set.
2524 * read.c (s_lcomm): Set section flags for .sbss section.
2526 * config/tc-mips.c (mips_64): New static variable.
2527 (mips_target_format): If mips_64, return elf64 targets rather than
2529 (md_longopts): Add "32" and "64".
2530 (md_parse_option): Handle -32 and -64.
2531 (md_show_usage): Mention -32 and -64.
2532 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
2537 * config/tc-ppc.c (comment_chars): Make '!' a comment character
2538 for Solaris compatibility.
2540 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
2541 .stabd with 4 arguments into a .stabn.
2545 * config/tc-mips.c (macro): When passing X_add_number to
2546 macro_build, cast it to int first.
2550 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
2551 created by emit_expr.
2553 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
2554 as_perror rather than assert.
2558 * config/tc-mips.c (mips_ip): Mark sections created to hold
2559 floating point information as read only.
2563 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
2567 * read.c (potable): Add .skip as a synonym for .space.
2569 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
2570 4 arguments, providing the 4th argument is 0, to allow
2571 compatibility with the Solaris assembler.
2575 * config/tc-sh.h (struct sh_segment_info_type): Define.
2576 (TC_SEGMENT_INFO_TYPE): Define.
2577 (sh_frob_label): Declare.
2578 (tc_frob_label): Define.
2579 (sh_flush_pending_output): Declare.
2580 (md_flush_pending_output): Define.
2581 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
2582 reloc before the instruction if necessary.
2583 (sh_frob_label): New function.
2584 (sh_flush_pending_output): New function.
2585 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
2586 when looking for the reloc for the target of .uses.
2587 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
2588 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
2590 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
2591 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
2592 the absolute symbol.
2594 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
2595 TC_SEGMENT_INFO_TYPE is defined.
2599 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
2600 matches the register size.
2604 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
2606 (do_relocs_for): Likewise.
2607 (fixup_segment): Don't just quit if linkrelax is set. Try to
2608 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
2613 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
2614 relative relocations with -mrelocatable. Also allow unfixed
2615 relocs in .ex_shared.
2619 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
2620 non-zero before assuming this is a long file name.
2621 (w_strings): Likewise.
2622 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
2624 * config/obj-coff.c (w_strings): Move declaration of i inside
2625 #ifdef block which uses it.
2629 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
2630 "relaxmode". Output relocs which identify various relaxing
2631 possibilities for mov.[bwl] instructions.
2632 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
2636 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
2637 (MD_APPLY_FIX3): Define.
2638 * config/tc-arm.c (my_get_expression): Only watch for bad segments
2640 (md_apply_fix3): Renamed from md_apply_fix.
2641 If pcrel reloc and symbol is in different section, undo effects
2646 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
2647 any reloc with an LR% or RR% field selector for SOM.
2651 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
2656 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
2658 (yank_symbols): Likewise.
2662 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
2663 BFD_ASSEMBLER, add name field.
2664 * config/obj-coff.c: Include "libiberty.h".
2665 (coff_header_append): Handle long section names.
2666 (crawl_symbols): Just use the name field for the symbol name,
2667 without worrying about null byte termination.
2668 (w_strings): Handle long section names.
2669 (write_object_file): Likewise. Also, use the name field, rather
2671 (obj_coff_add_segment): Permit long section names.
2672 (obj_coff_init_stab_section): Use the name field, rather than
2674 (adjust_stab_section): Likewise.
2675 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
2677 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
2678 (mem_fmt): Likewise.
2679 (md_apply_fix): Don't check fx_im_disp field.
2683 * configure.in: Add * after sparc*-*-vxworks.
2684 * configure: Rebuild.
2686 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
2687 define it to comment_chars. Use tc_comment_chars rather than
2689 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
2690 * config/tc-m68k.h (tc_comment_chars): Define.
2691 (m68k_comment_chars): Declare.
2692 * config/tc-m68k.c (m68k_comment_chars): Rename from
2693 comment_chars. Change into a pointer rather than an array.
2694 (md_longopts): Add "bitwise-or".
2695 (md_parse_option): Handle OPTION_BITWISE_OR.
2696 (md_show_usage): Mention --bitwise-or.
2697 * doc/c-m68k.texi: Document --bitwise-or.
2701 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
2704 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
2705 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
2706 correct handling of input line pointer, and call
2707 demand_empty_rest_of_line.
2711 * config/tc-sparc.c (in_bitfield_range): New static function.
2712 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
2713 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
2714 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
2715 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
2716 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
2720 * config/obj-coff.c: BFD_ASSEMBLER:
2721 (coff_last_bf): New static variable.
2722 (coff_frob_symbol): Set endndx of a .bf symbol.
2724 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
2725 (last_bfP): New static variable.
2726 (yank_symbols): Set endndx of a .bf symbol.
2730 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
2731 and -s options. Add -be for good measure.
2735 * read.c (s_space): Support non-constant fill value. Handle fill
2736 value correctly for a size other than 1.
2740 * config/tc-arm.c (my_get_float_expression): Update call to
2741 gen_to_words, X_PRECISION changed from 6 to 5.
2745 * config/tc-ppc.c (register_name,reg_name_search): Move register
2746 name lookup from PE specific code to all targets. Add support for
2747 -mregnames/-mno-regnames to control whether register names are
2749 (md_assemble): Call register_name for all platforms.
2750 (md_parse_option): Add support for -mregnames/-mno-regnames.
2752 * configure.in (powerpcle*-*-solaris): Add support.
2753 (powerpc*-*-linux): Ditto.
2754 * configure: Regenerate.
2756 * config/ppc-sol.mt: New config file for PowerPC Solaris.
2760 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
2761 be associated with a single %lo reloc.
2763 * config/tc-mips.c (load_address): Cast X_add_number to valueT
2764 before comparing against MAX_GPREL_OFFSET, so that negative
2765 numbers are handled correctly.
2770 * config/tc-sparc.c (last_insn): New static variable.
2771 (md_assemble): Warn about putting floating point branches in a
2772 delay slot. If architecture is less than v9, insert NOP
2773 instructions between floating point instructions and floating
2774 point branches. (The SunOS assembler does both these operations.)
2775 Save the last instruction opcode.
2776 (sparc_ip): Add pinsn parameter. Change caller.
2778 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
2779 for byte jump to next instruction to skip empty frags.
2783 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
2784 for addq with a 16 bit signed value, just emit a lda instruction
2789 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
2794 * configure.in: Permit --enable-shared to specify a list of
2796 * configure: Rebuild.
2800 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
2801 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
2805 * config/atof-ieee.c: Fix handling of denormalized extended
2806 precision numbers and overflow/underflow detection.
2807 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
2808 not include the 16 bit gap in the m68k extended precision format.
2812 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
2814 * configure: Rebuild.
2818 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
2819 be accessed using the $gp register.
2820 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
2824 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
2825 bfd_mach_sparc_sparclet if current_architecture is sparclet.
2829 * read.c (get_line_sb): Bump line counters based on
2830 input_line_pointer[-1], not *input_line_pointer. Don't bother to
2831 call LISTING_NEWLINE.
2832 (s_macro): Don't call demand_empty_rest_of_line.
2833 * app.c (do_scrub_chars): When handling C style comments, unget
2838 * read.h (enum linkonce_type): Define.
2839 (s_linkonce): Declare.
2840 * read.c (potable): Add "linkonce".
2841 (s_linkonce): New function.
2842 * subsegs.h (segment_info_type): Add linkonce field to
2843 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
2844 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
2845 (obj_coff_pe_handle_link_once): Declare if TE_PE.
2846 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
2848 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
2849 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
2850 entry based on the linkonce field in segment_info.
2851 * doc/as.texinfo: Document .linkonce.
2855 * doc/as.1: Changed to be recognized by catman -w on Solaris.
2859 * stabs.c (s_stab_generic): Call the listing functions before
2860 doing the rest of the processing, which may involve freeing the
2861 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
2862 SEPARATE_STAB_SECTIONS case.
2864 * config/tc-hppa.c: Remove nested comment.
2865 (tc_gen_reloc): Move label done inside the ifdef in which it is
2867 (md_apply_fix): Pass pointers to correct types to libhppa.h
2868 functions. Always return a value.
2870 * config/tc-mips.h (tc_frob_file): Define.
2871 (mips_frob_file): Declare.
2872 * config/tc-mips.c (struct mips_hi_fixup): Define.
2873 (mips_hi_fixup_list): New static variable.
2874 (imm_unmatched_hi): New static variable.
2875 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
2876 offset_reloc. Pass imm_unmatched_hi to append_insn.
2877 (append_insn): Add unmatched_hi parameter. If it is set, add the
2878 new fixup to mips_hi_fixup_list. Change all callers.
2879 (mips_ip): Set imm_unmatched_hi when appropriate.
2880 (mips_frob_file): New function.
2884 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
2885 * configure: Regenerated.
2889 * as.c (main): Call bfd_set_error_program_name.
2893 * as.h (strdup): Don't declare.
2894 * stabs.c: Include libiberty.h
2895 (get_stab_string_offset): Use xstrdup rather than strdup.
2896 (s_stab_generic): Likewise.
2897 * as.c (parse_args): Likewise.
2898 * read.c (s_mri_sect): Likewise.
2900 * gasp.c (change_base): Recognize \(...) construct documented to
2901 pass through enclosed characters literally through to the output.
2902 (process_assigns): Likewise. Also, be more careful to avoid
2903 looking past the end of the buffer.
2907 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
2912 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
2916 * config/alpha-opcode.h: Added cvtst instruction.
2920 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
2921 string. Extract the addend from the result of expression.
2925 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
2926 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
2927 mode, change the state to 3 rather than 1.
2931 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
2932 correspond to 11 March change.
2936 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
2937 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
2941 * configure: Rebuild with autoconf 2.8.
2945 * config/atof-ieee.c (gen_to_words): Improve handling of
2946 X_PRECISION numbers. Based on patches from Andreas Schwab
2951 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
2953 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
2954 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
2955 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
2956 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
2961 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
2965 * config/tc-hppa.c (SUBSEG_MILLI): Define.
2966 (pa_def_subspaces): Add $MILLICODE$.
2967 (pa_spaces_begin): Set section flags for $MILLICODE$.
2971 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
2972 not SHT_NOBITS. Don't tamper with flags based on type if a
2973 special section was found (revert Feb 29 change).
2975 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
2976 if aligning to a longword boundary or greater.
2980 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
2984 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
2985 * configure: Regenerated.
2989 * config/tc-hppa.c: Add default definitions for R_N0SEL and
2990 R_N1SEL since they're not defined for old versions of hpux.
2992 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
2993 Set "sym_ptr_ptr" and "addend" fields to dummy values for
2994 R_N0SEL and R_N1SEL.
2998 * flonum-konst.c: Add two more constants for 1e+-2048 and
2999 1e+-4096, and correct the other constants.
3001 * symbols.c (resolve_symbol_value): Handle O_logical_not.
3005 * config/obj-elf.c (obj_elf_section): Allow predefined section
3006 types to set the nobits type. Avoid a shadowed declaration.
3010 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
3011 reduce relocs using e_nlrsel field selectors.
3013 * write.c (fix_new_exp): Don't use #elif. Some compilers
3016 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
3018 (pa_chk_field_selector): Handle new field selectors for SOM.
3022 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
3023 * configure: Rebuild.
3024 * config/te-linux.h (LOCAL_LABELS_FB): Define.
3025 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
3027 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
3032 * mpw-make.sed: Update to handle shared library support.
3036 * config/tc-sparc.c (sparc_ip): Print all architectures that support
3037 the insn on mismatch.
3041 * configure.in: Add support for a29-coff.
3042 * configure: Rebuild.
3046 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
3047 of the section when looking for the R_SH_USES fixup, because the
3048 frag addresses have not yet been adjusted.
3050 * gdbinit.in: Set a breakpoint on as_warn_where.
3052 * config/tc-mips.c (macro): Add missing arguments to macro_build
3057 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
3058 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
3062 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
3066 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
3067 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
3068 Recognize [uU] format args as sparclet cpregs.
3072 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
3077 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
3078 well as S_IS_EXTERNAL.
3079 (tc_fix_adjustable): Likewise.
3080 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
3081 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
3083 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
3084 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
3088 * config/tc-sparc.c (max_architecture): Change to sparclite for
3090 (default_compatible): Delete.
3091 (sparc_ffs): New function.
3092 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
3093 (sparc_ip): Rewrite architecture match and bump logic.
3097 * configure.in: Check for 'do not mix' from native linker before
3098 trying to use -rpath.
3099 * configure: Rebuild.
3103 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3104 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3105 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3106 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
3108 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
3109 to the symbol lnnoptr field, to get the correct file offset.
3113 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
3114 symbols that are not function, file, or section symbols to be
3119 * configure: Set and substitute RPATH_ENVVAR.
3120 * configure: Rebuild.
3121 * Makefile.in (RPATH_ENVVAR): New variable.
3122 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
3125 doesn't need opcodes. If configuring shared, opcodes needs bfd.
3126 * configure: Rebuild.
3130 * read.c (s_mri_sect): Don't return '\0' in type. Set all
3131 appropriate flags in BFD section.
3133 * configure.in (m68k-*-psos*): New target.
3134 * configure: Rebuild.
3135 * config/te-psos.h: New file.
3136 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
3142 * configure.in: Remove duplicate setting of cpu_type. Check
3143 whether opcodes library is required for on all targets, not just
3145 * configure: Rebuild.
3147 * config/tc-mips.c (mips_big_got): New static variable.
3148 (s_extern): Don't declare.
3149 (reg_needs_delay): New static function.
3150 (macro_build): Permit GOT/CALL_HI/LO relocs.
3151 (macro_build_lui): If place is not NULL, use the number in the
3153 (load_address): Handle mips_big_got case.
3154 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
3156 (OPTION_XGOT): Define.
3157 (md_longopts): Add "xgot" if OBJ_ELF.
3158 (md_parse_option): Handle -xgot.
3159 (md_show_usage): Mention -xgot.
3160 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
3161 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
3165 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
3166 the byte relocation, point it to the low byte of the word.
3170 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
3171 * configure: Rebuild.
3175 * configure.in: Recognize any sparc* cpu.
3176 * configure: Regenerated.
3180 * read.c (potable): Add "mri" and ".mri".
3181 (s_mri): New function.
3182 * read.h (s_mri): Declare.
3183 * app.c (scrub_m68k_mri): New static variable.
3184 (mri_pseudo): New static variable.
3185 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
3186 flag_m68k_mri. Initialize scrub_m68k_mri.
3187 (mri_state, mri_last_ch): New static variables.
3188 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
3190 (app_push): Save new fields.
3191 (app_pop): Restore new fields.
3192 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
3193 If TC_M68K, use a trivial state machine to look for occurrences of
3194 the .mri pseudo-op, and change the mode appropriately.
3195 * as.h (do_scrub_begin): Update prototype.
3196 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
3198 * config/tc-m68k.c (reg_prefix_optional_seen): New static
3200 (m68k_mri_mode_change): New function.
3201 (md_parse_option): Set reg_prefix_optional_seen.
3202 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
3203 (MRI_MODE_CHANGE): Define.
3204 * doc/as.texinfo: Document .mri pseudo-op.
3206 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
3211 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
3212 * config/m68k-parse.h (enum pic_relocation): Define.
3213 (struct m68k_exp): Add pic_reloc field.
3214 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
3215 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
3216 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
3218 * config/tc-m68k.c: Delete definition of NO_RELOC.
3219 (struct m68k_it): Add pic_reloc field.
3220 (add_fix): Copy over pic_reloc field.
3221 (md_pseudo_table): Interpret .align parameter as byte count.
3222 (mote_pseudo_table): Likewise.
3223 (tc_m68k_fix_adjustable): New function.
3224 (get_reloc_code): New function.
3225 (md_assemble): Use it as last argument to fix_new_exp.
3226 (md_apply_fix_2): For a relocation against a symbol don't put the
3227 addend into the data.
3228 (tc_gen_reloc): Different addend computation for OBJ_ELF.
3229 (m68k_ip): Don't relax an operand that requires pic relocation.
3230 (md_begin): Align .text, .data and .bss on 4 byte boundary by
3232 * write.c (fixup_segment): Don't add symbol value to addend if
3233 TC_M68K and OBJ_ELF.
3234 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
3235 (motorola_operand): Add rule for `(zapc, EXPR)'.
3239 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
3240 *really* is permissible before the comma.
3244 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
3245 inserting nop instructions.
3249 * config/te-aux.h: Change include of aux.h to aux-coff.h.
3253 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
3254 class setting for a CALLNAME symbol in COFF.
3256 * read.c (potable): Pass negative numbers for new .balign[wl] and
3257 .p2align[wl] pseudo-ops.
3258 (s_align_bytes): Treat a negative argument as specifying the fill
3260 (s_align_ptwo): Likewise.
3264 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
3265 (do_align): Take new len parameter. Change all callers. Pass it
3267 (s_align_bytes): Arg now indicates the length of the fill pattern.
3268 (s_align_ptwo): Likewise.
3269 * config/obj-coff.c (write_object_file): Pass length to
3271 * config/tc-i386.h (md_do_align): Take new len parameter.
3272 * config/tc-m88k.h (md_do_align): Likewise.
3273 * config/tc-m88k.c (m88k_do_align): Likewise.
3274 * config/tc-sh.h (md_do_align): Likewise.
3275 * config/tc-sh.c (sh_do_align): Likewise.
3276 * doc/as.texinfo: Document new pseudo-ops.
3278 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
3279 done in cvt_frag_to_fill.
3281 * config/tc-sh.h (sh_do_align): Declare.
3282 (md_do_align): Define.
3283 * config/tc-sh.c (sh_do_align): New function.
3285 * ecoff.c (ecoff_build_lineno): Don't try to store the address
3286 difference if the next address is before the current one.
3288 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
3289 (archs): Initialize new field.
3290 (m68k_ip): Don't list alias names when listing CPUs which support
3293 * as.c (main): Call parse_args before read_begin.
3294 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
3295 front of generated pseudo-ops.
3296 * read.c (potable): Ignore "name".
3297 (s_app_file): Permit a single quote after the string, since one
3298 may appear in m68k MRI mode.
3300 * configure.in: Check for --enable-shared. If linking against
3301 shared BFD and opcodes, fix library name on SunOS, and try to set
3303 * configure: Rebuild.
3307 * as.h (flag_m68k_mri): Declare.
3308 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
3309 * Many files: For MRI syntax that is specific to the m68k MRI
3310 assembler, check flag_m68k_mri rather than flag_mri or
3311 MRI_MODE_NEEDS_PSEUDO_DOT.
3315 * config/tc-i960.c (ARCH_HX): Define.
3317 (targ_has_sfr): Handle ARCH_HX.
3318 (targ_has_iclass): Handle ARCH_HX.
3319 (tc_coff_fix2rtype): Add return 0 to avoid warning.
3320 (tc_headers_hook): If the architecture was specified explicitly,
3321 use it when setting the flags. Set the extern variable coff_flags
3322 rather than headers->filehdr.f_flags, since the latter is set
3323 unconditionally in obj-coff.c.
3324 (i960_handle_align): Remove unused variable fixp.
3326 Support for building bfd and opcodes as shared libraries, based on
3328 * configure.in: Set OPCODES and BFD to search directories.
3329 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
3330 * configure: Rebuild.
3331 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
3332 (LIBDEPS): New variable.
3333 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
3334 $(HLDFLAGS) in link.
3335 (check): Set LD_LIBRARY_PATH in the environment.
3339 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
3340 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
3341 read-only, not read/write.
3345 * Makefile.in (INSTALL_XFORM): Remove -e.
3349 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
3350 than comparing S_GET_SEGMENT to undefined_section.
3351 (write_object_file): Skip symbols which were equated to an
3352 undefined or common symbol.
3353 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
3354 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3356 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
3357 or S_IS_COMMON call S_GET_VALUE.
3358 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3359 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
3360 to an undefined or common symbol, convert the reloc to be against
3362 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3363 undefined or common symbol.
3364 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3365 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
3366 to an undefined or common symbol, convert the reloc to be against
3368 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3369 undefined or common symbol.
3370 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
3371 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3373 (yank_symbols): Skip symbols which were equated to an undefined or
3378 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
3379 * config/obj-bout.h (S_IS_LOCAL): Likewise.
3381 * configure.in: Make sure we only add m68k-parse.o to
3382 ${extra_objects} once, no matter how many m68k targets have been
3384 * configure: Rebuild.
3388 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
3389 * configure: Rebuild.
3393 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
3394 * doc/c-m68k.texi: Document .extend and .ldouble.
3396 * configure.in (m68*-apple-aux*): New target.
3397 * config/te-aux.h: New file.
3398 * config/obj-coff.c (compare_external_relocs): New static function
3400 (do_relocs_for): Sort relocs if TE_AUX.
3401 (fixup_segment): If TE_AUX, store common symbol value in segment.
3402 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
3406 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
3408 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
3410 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
3411 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
3413 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
3414 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
3415 an undefined symbol and the size of a function symbol.
3416 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
3417 for all common symbols.
3421 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
3424 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
3429 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
3432 * ecoff.c (localsym_t): Add addend field.
3433 (add_ecoff_symbol): Add addend argument. Change all callers.
3434 (coff_sym_value): Make static.
3435 (coff_sym_addend): New static variable.
3436 (ecoff_directive_def): Initialize coff_sym_addend.
3437 (ecoff_directive_val): Accept symbol + constant.
3438 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
3439 (ecoff_build_symbols): Include the addend in the symbol value.
3443 * config/tc-ppc.c (md_assemble): Ignore overflow on
3444 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
3448 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
3449 bytes in size, so offset appropriately in big endian mode when
3450 writing the bottom 2 bytes.
3454 * config/tc-sparc.c (default_compatible): New static local.
3455 (md_begin): Initialize it. Rewrite warn_on_bump handling.
3456 (sparc_ip): If no architecture or -bump specified, don't mark as
3457 mismatched those in default_compatible.
3462 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
3463 * configure: Rebuild.
3464 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
3465 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
3466 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
3467 (sco_id): Declare if SCO_ELF.
3471 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
3472 ({max,warn_after}_architecture): New static locals.
3473 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
3474 If both architecture and -bump requested, set max_architecture to max.
3475 (sparc_md_end): Simplify.
3476 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
3477 Rewrite code to bump architecture and check for conflicts.
3478 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
3479 with Solaris assembler.
3480 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
3481 (md_show_usage): Update.
3485 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
3489 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
3490 #ifdef sparcv9 when choosing value.
3493 (sparc_md_end): Declare.
3494 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
3495 (initial_architecture): New static local.
3496 (can_bump_v9_p): Likewise.
3497 (NO_V9): Delete all occurrences.
3498 (sparc_md_end): New function.
3499 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
3500 Don't bump architecture to v9 unless can_bump_v9_p set.
3501 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
3502 * configure.in (sparc64 target cpu): Don't set obj_format here.
3503 (SPARC_V9): Renamed from sparcv9.
3504 (sparc64-*-elf*): Define SPARC_ARCH64.
3505 * configure: Regenerated.
3506 * acconfig.h (SPARC_V9): Renamed from sparcv9.
3507 (SPARC_ARCH64): Add.
3508 * conf.in: Regenerated.
3509 * config/vmsconf.h: Update.
3513 * config/tc-mips.c (load_register): Optimise "dli" loads.
3514 (md_show_usage): add "-mcpu=vr4100" to help text.
3518 * symbols.c (resolve_symbol_value): If a symbol is equated to an
3519 undefined symbol, preserve the X_op of O_symbol.
3520 (S_GET_VALUE): Fix check to permit this case.
3521 * write.c (write_relocs): If a reloc is against an undefined
3522 symbol equated to another symbol, change the reloc to be against
3524 * config/obj-coff.c (do_relocs_for): Likewise.
3526 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
3531 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
3532 computer is PowerPC.
3534 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
3536 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
3537 mentioned in the eabi.
3541 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
3542 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
3546 * config/tc-ppc.c (mapping): Add more relocation suffixes.
3550 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
3551 as a section it is ok to have unadorned -mrelocatable pointers in.
3555 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
3560 * config/obj-elf.c (obj_elf_section): Add hooks so machine
3561 dependent section attributes can be handled.
3563 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
3564 macros to add support for exclude section flag and ordered section
3567 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
3568 functions to add support for exclude section flag and ordered
3573 * subsegs.c (section_symbol): Don't try to look up the section
3574 symbol in the hash table. It should be possible to have a symbol
3575 with the same name as a section, but no connection to it.
3577 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
3580 * expr.c (operand): Skip whitespace after a close parenthesis.
3585 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
3586 as_warn_where for relocation overflow.
3587 (parse_reg): Accept register name only if next character is
3592 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
3593 addresses for static variables.
3597 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
3598 and xgot give the old toc16 relocation.
3599 (md_apply_fix3): Support all GOT relocations.
3603 * config/tc-mips.c (load_address): Correctly handle a constant in
3608 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
3609 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
3610 (build_Mbytes): Handle REG_B.
3614 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
3618 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
3622 * config/obj-coff.c (write_object_file): Set the s_align field to
3623 the number of bytes, rather than to the power of 2.
3627 * Makefile.in (DISTCLEAN_HERE): New variable.
3628 (distclean): Use it.
3629 (maintainer-clean): Depend upon clean-here rather than clean,
3630 distclean, and clean-info. Run make maintainer-clean in doc.
3631 Remove files listed in DISTCLEAN_HERE.
3632 * doc/Makefile.in (maintainer-clean realclean): Split out from
3633 distclean. Depend upon clean-info and distclean.
3637 * mac-as.r: Fix copyright and version strings.
3638 (cfrg): Use PROG_NAME instead of literal name.
3642 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
3644 * config/tc-a29k.h (tc_unrecognized_line): Define.
3645 * config/tc-a29k.c (a29k_unrecognized_line): New function.
3646 (md_operand): Handle a29k style local dollar labels.
3650 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
3654 * read.c (s_fill): If md_flush_pending_output is defined, call
3659 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
3660 Treat rs_align_code like rs_align.
3664 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
3665 converting an absolute address into a PC-relative one.
3669 * mpw-config.in: Don't always use te-generic.h for emulation.
3670 (powerpc-apple-macos): Use emulation te-macos.h.
3672 (mips-gm-magic): New configuration.
3674 * mpw-make.sed (install, install-only): Edit in Mac-specific
3679 * configure.in: Improve message about unsupported ELF targets.
3680 * configure: Rebuild.
3682 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
3683 pattern is zero. From Manfred Hollstein.
3687 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
3688 the various idata sections, we check the name on the .section pseudo.
3692 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
3693 add section's paddr to add_number; compatibility to native as and
3698 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
3700 * hash.c (struct hash_entry): Moved here...
3701 * hash.h (struct hash_entry): ...from here.
3703 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
3704 if it's already known to be null.
3708 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
3709 bit relocs converted to PC relative relocs. This turned out to
3710 add the offset from the beginning of .text twice.
3714 * stabs.c (s_stab_generic): In 's' case, free string from
3717 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
3719 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
3722 * subsegs.h (segment_info_type): Make bitfields unsigned.
3724 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
3725 and move them together. On most systems this will reduce the
3726 structure size by one word.
3727 (operatorT): Define O_max.
3728 * expr.c (expr_begin): Verify that X_op is wide enough to hold
3731 * read.c (pop_insert): Print error returned by hash table
3734 * as.c (dump_statistics): Split out from main; dump some hash
3735 table stats and target-specific stats.
3736 (start_time): No longer automatic to main.
3737 (main): Set file-level start_time and call dump_statistics at
3738 exit. Exit by calling xexit.
3739 (show_usage): Make --statistics description less specific.
3740 * subsegs.c (subsegs_print_statistics): New function.
3741 * write.c (write_print_statistics): New function.
3742 (n_fixups): New static variable.
3743 (fix_new_internal): Increment it.
3744 * read.c (read_print_statistics): New function.
3745 * read.h (read_print_statistics): Declare.
3746 * symbol.c (symbol_print_statistics): New function.
3747 * symbol.h (symbol_print_statistics): Declare.
3748 * hash.c (hash_print_statistics): New function.
3749 * hash.h (hash_print_statistics): Declare.
3750 * config/tc-i386.c (i386_print_statistics): New function.
3751 * config/tc-i386.h (i386_print_statistics): Declare.
3752 (tc_print_statistics): New macro.
3753 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
3756 * hash.c (DELETED): Rewrite to use a valid but unique address.
3757 (START_POWER): Reduce to 10.
3758 (enum stat_enum): New enumerator, replacing STAT_* index macros.
3759 Add new values for counting strcmp calls.
3760 (GROW_FACTOR): New macro.
3761 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
3763 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
3764 (hash_new): Use FULL_VALUE.
3765 (struct hash_control): Definition moved here.
3766 (hash_code): Don't mask to low bits.
3767 (hash_ask): Mask returned hash code here. Check hash value before
3768 calling strcmp; count strcmp calls.
3769 * hash.h (struct hash_control): Declare, don't define, here.
3770 (HASH_STATLENGTH): Deleted.
3771 (struct hash_entry): Add field for hash code.
3772 (hash_say, hash_apply): Don't declare.
3774 * hash.c (destroy): Return void.
3775 (applicatee): Ditto.
3776 (main): Fix declarations.
3777 (hash_apply): Return void. Argument `function' returns void. Put
3778 inside "#ifdef TEST".
3779 (hash_say): Define only if TEST is defined.
3780 * hash.h (hash_apply, hash_say): Declarations deleted.
3784 * configure: Rebuild with autoconf 2.7.
3788 * aclocal.m4 (AC_PROG_CC): Remove local definition.
3789 * configure: Rebuild with autoconf 2.6.
3793 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
3794 (ppc_stabx): Don't increment ppc_debug_name_section_size.
3796 (ppc_frob_file): Remove.
3797 * config/tc-ppc.h (tc_frob_file): Don't define.
3798 (ppc_frob_file): Don't declare.
3802 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
3803 * config/alpha-opcode.h: Include one-operand variants of jmp and
3806 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
3807 with te-delta.h in 8.3 file systems.
3808 * configure.in: Adjusted.
3812 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
3814 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
3816 * expr.c (operand): Handle MRI suffixes after unadorned 0.
3820 Version 2.6 released.
3821 * Makefile.in (VERSION): Updated to 2.6.
3823 * config/obj-coff.c (write_object_file): Change use of md_do_align
3824 to pass a pointer rather than a fill value, to match other uses.
3828 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
3830 * as.c (main): Move md_end call to just after call to
3831 perform_an_assembly_pass. Delete cpu-specific code here.
3832 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
3833 * config/tc-arm.c (md_end): Unused function deleted.
3834 * config/tc-ns32k.c (md_end): Ditto.
3836 * config/tc-i386.c (i386_align_code): New function, moved here
3837 from HANDLE_ALIGN macro.
3838 * config/tc-i386.h (HANDLE_ALIGN): Call it.
3842 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
3847 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
3848 multi-byte nops to avoid changing gcc. The necessary gcc change
3849 might break old assemblers.
3853 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
3854 generate optimal multi-byte nop instructions for ".align n"
3855 ".align n,0x90", and aligns requiring more than 15 bytes of
3856 padding still generate multiple 0x90's as before.
3860 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
3864 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
3865 * configure: Update.
3866 * mpw-make.sed: Reorder commands to make sed happier.
3867 * config/te-macos.h: New file.
3868 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
3872 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
3873 conditional branches.
3874 (md_apply_fix): Likewise.
3878 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
3879 generate a reloc which the linker will not handle correctly. Fix
3880 overflow checking--R_IREL is 18 bits, not 17.
3884 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
3885 from a PC relative reloc if TC_A29K.
3887 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
3888 code depends upon it.
3890 * config/tc-a29k.c (md_operand): Handle $float, $double, and
3891 $extend. Based on code from Eric Freudenthal
3893 * config/tc-a29k.h (LEX_DOLLAR): Define.
3894 * read.c (LEX_DOLLAR): Define if not defined.
3895 (lex_type): Use LEX_DOLLAR.
3899 * configure.in (a29k-nyu-sym1): New target, just like other a29k
3904 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
3908 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
3909 used only by "dlltool.c".
3913 * config/tc-alpha.c: Undefine inline if not __GNUC__.
3914 (md_pseudo_table): Don't define "extern".
3918 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
3920 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
3924 * macro.c (macro_expand_body): Don't warn about == with a
3925 nonexistent parameter, in case it is in a comment field.
3927 * as.c (main): On TC_A29K, call macro_init with macro_alternate
3929 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
3930 set, even if macro_alternate is set.
3931 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
3932 (sub_actual): If macro_strip_at, and kind is '@', don't look up
3933 the token unless it ended in '@'.
3934 * config/tc-a29k.c (line_separator_chars): Remove '@'.
3935 * doc/c-a29k.texi: Document macro usage on A29K.
3939 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
3941 (install_operand): Likewise.
3943 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
3947 * configure.in (m88k-motorola-sysv*): New target.
3948 * configure: Rebuild.
3949 * config/te-delta88.h: New file.
3950 * config/obj-coff.c (write_object_file): Use md_do_align if it is
3952 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
3953 (md_do_align): Define.
3954 * config/tc-m88k.c: Include "subsegs.h".
3955 (m88k_do_align): New function.
3957 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
3958 (COFF_NOLOAD_PROBLEM): Define.
3959 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
3963 * config/tc-i386.c (md_assemble): For a jump instruction with
3964 non-constant target, require 7 available bytes in the current
3969 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
3974 * configure.in (alpha-*-linux*): Use ecoff.
3975 * configure: Rebuild.
3976 * ecoff.c (ecoff_directive_extern): New function.
3977 (ecoff_directive_weakext): New function.
3978 (ecoff_build_symbols): Handle weak symbols.
3979 (ecoff_setup_ext): Likewise.
3980 (ecoff_frob_symbol): Warn about weak common symbols.
3981 * ecoff.h (ecoff_directive_extern): Declare.
3982 (ecoff_directive_weakext): Declare.
3983 * symbols.c (S_IS_WEAK): New function.
3984 * symbols.h (S_IS_WEAK): Declare.
3985 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
3987 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
3992 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
3994 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
3995 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
3996 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
3997 * config/obj-coff.c (coff_frob_symbol): Don't call
3998 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
3999 one. If OBJ_XCOFF, try to figure out whether the symbol is going
4002 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
4003 (ppc_stab_symbol): New static variable.
4004 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
4005 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
4006 sy_tc.real_name to the stab string.
4007 (ppc_bc, ppc_ec): New static functions.
4008 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
4010 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
4012 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
4015 * input-scrub.c (input_scrub_push): Reset sb_index.
4019 * listing.c (listing_newline): Don't create a frag in the absolute
4024 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
4025 (ppc_csect): Move most of the code to ppc_change_csect, and call
4027 (ppc_change_csect): New static function, taken from ppc_csect.
4028 (ppc_section): New static function.
4029 (ppc_saw_abs): New static varable.
4030 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
4031 a symbol has no csect.
4032 (ppc_adjust_symtab): New function.
4033 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
4034 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
4036 * write.c (write_object_file): If tc_adjust_symtab is defined,
4037 call it just before the call to obj_adjust_symtab.
4039 * symbols.c (symbol_find_or_make): Change name to be const.
4040 * symbols.h (symbol_find_or_make): Update declaration.
4044 * doc/as.texinfo (Align): Mention SH.
4045 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
4047 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
4048 (Z8000 Directives, even): Don't give numeric align value, instead
4053 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
4058 * Makefile.in (diststuff): Also make info.
4059 (maintainer-clean realclean): Also make clean-info.
4063 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
4064 (pa_subspace): For ".nsubspa", always create a new subspace
4065 with the given attributes, even if one already exists with the
4070 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
4071 TC_FORCE_RELOCATION, taking an additional section argument. If
4072 the section of the target symbol is not the same as the current
4073 section, always force the relocation to be used.
4074 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
4076 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
4077 md_pcrel_from function, taking an additional section argument.
4078 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
4080 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
4081 older TC_FORCE_RELOCATION if not defined.
4082 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
4083 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
4084 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
4085 TC_FORCE_RELOCATION.
4089 * input-scrub.c (as_where): Set name to null pointer if we don't
4091 * messages.c (identify): Only print filename if non-null.
4092 (as_show_where): Ditto, for line number too.
4093 (as_warn_internal, as_bad_internal): Ditto.
4095 * input-file.c (input_file_open): If the input file can't be
4096 opened, consider it an error.
4100 * config/tc-mips.c: Added mips_4100 control, and support for
4101 accepting the 4100 as a MIPS architecture variant (md_begin,
4102 macro_build, mips_ip, md_parse_option). Adding suitable
4103 command-line OPTIONs, and updating the help text (md_show_usage).
4107 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
4108 BFD_ASSEMBLER is defined.
4110 * Use one active frag and one obstack per frag chain:
4111 * frags.c (frags): Variable deleted.
4112 (frag_alloc): New function.
4113 (frag_grow, frag_more, frag_variant, frag_now_fix,
4114 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
4116 (frag_new): Ditto. Verify that frch_last and frag_now match on
4117 entry and exit, and that old frag_now has non-zero type. Replace
4118 "know" uses with "assert". Use frag_alloc instead of mucking with
4120 * frags.h (frags): Declaration deleted.
4121 * subsegs.h (struct frchain): Add new field frch_frag_now.
4122 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
4124 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
4125 give it any stricter alignment than frchainS structures need. Do
4126 not initialize frags obstack. Set frag_now to point to
4127 dummy_obstack. Initialize absolute_frchain.
4128 (subseg_set_rest): Save and restore frag_now in frch_frag_now
4129 field of frchainS. Don't create new frags on section switch, and
4130 use frag_alloc when creating a new frag chain. For absolute
4131 section, set frchain_now to absolute_frchain. Verify that
4132 frch_last and frag_now match on entry and exit. Initialize
4133 per-chain obstack, and under gcc, set required alignment to that
4134 needed by fragS structure.
4136 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
4138 * stabs.c (get_stab_string_offset): Only copy input string if a
4139 fresh copy is needed, not if the section already exists.
4140 (s_stab_generic): Cache stab section name to bypass lookups, since
4141 usually it will match. Could be made faster still by changing the
4142 memory allocation rules.
4143 (s_xstab): Cache section name to bypass repeated string
4146 * frags.c (frag_new): Deleted register declarations.
4148 * listing.c (frag_now): Don't declare.
4150 * as.c (chunksize): New variable.
4151 (debug_memory): New variable.
4152 (main): If debug_memory is set, reduce chunksize and
4154 * as.h (chunksize): Declare it.
4155 * read.c (read_begin): Use it.
4157 * config/tc-alpha.c (md_shortopts): Include 'g'.
4158 (md_parse_option): Ignore it.
4160 * Makefile.in (distclean): Remove Makefile and config.status from
4161 testsuite directory.
4162 (clean-here): Don't delete testsuite. Instead, delete only the
4163 files within it that would be generated by running tests.
4165 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
4166 frag_now_fix instead of accessing obstack info directly.
4167 * config/tc-arm.c (s_ltorg): Ditto.
4168 (md_assemble): Ditto.
4170 * config/tc-i386.c (md_assemble): Call frag_grow instead of
4175 * stabs.c (aout_process_stab): Insert debug symbol into symbol
4176 chain after parsing value expression, if any, to avoid separating
4177 continued .stabs lines.
4181 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
4182 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
4186 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
4187 -mcom support and PPC_OPCODE_ANY for -many.
4188 (md_parse_option): Ditto.
4190 (md_begin): For duplicate instructions, print all duplicates
4195 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
4196 common mode operation.
4197 (md_show_usage): Add -mcom to usage message.
4201 * expr.c (op_rank): Add O_symbol_rva.
4202 * expr.h (operatorT): Add O_symbol_rva.
4203 * read.c (cons_worker): Set O_symbol_rva when necessary.
4204 * write.c (fix_new_exp): Understand O_symbol_rva.
4208 * config/tc-mips.c: Correct s_cons arguments. From Michael
4213 * config/tc-ppc.c (ppc_macro): Make count unsigned.
4214 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
4215 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
4217 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
4218 make non-static. Change all users.
4219 * config/obj-coff.h (coff_n_line_nos): Declare.
4225 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
4227 * config/obj-elf.c (obj_elf_common): Convert specified byte
4228 alignment to power of two. Set size of local bss symbol.
4230 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
4234 * sb.c, macro.c: Decide whether to include <string.h> or
4235 <strings.h> just as as.h does.
4239 * Makefile.in (site.exp): Fix setting of $srcdir.
4241 * config/tc-arm.c (md_atof): Fix little-endian output.
4242 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
4247 * doc/as.texinfo: Split out the NS32k family documentation,
4248 despite its being commented out for now.
4249 * doc/c-ns32k.texi: New file.
4251 * sb.c, macro.c: Include string.h.
4253 * Makefile.in (comparison): Only check *.o; we don't care if
4254 timestamps inserted by the native linker differ.
4256 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
4257 if alignment stricter than 4 bytes is requested; in that case,
4258 align to a 4-byte boundary first.
4262 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
4263 uninitialized" warning for `Max_Source_Offset'.
4267 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
4268 qualifiers and detect errors if present.
4269 (md_assemble): Add call to parse_toc_entry. Also added some support
4270 for the [tocv] qualifier.
4271 (ppc_pe_tocd): New function to support data in the toc section.
4275 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
4280 * config/tc-m68k.c (isword): Accept all values from -65536 to
4281 +65535, so ~VAL will not be rejected.
4283 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
4284 skip characters after the pseudo-op.
4286 * read.c (get_line_sb): Don't look past buffer_limit.
4287 (s_include): In MRI mode, skip characters after the file name.
4291 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
4292 periods in register names.
4294 For older changes see ChangeLog.1.