3 * config/tc-mips.c (md_begin): Add the general registers to the
5 (mips16_ip): First parse the expression, and then see whether it
6 came up with a register, rather than trying to first see whether
7 we are looking at a register.
12 * config/tc-tic80.c: Numerous changes and additions to flesh
13 out functions that were previously just stubs, and fix some
14 problems found using the new TIc80 testsuite cases.
19 * config/tc-mips.c (mips16_ip): Handle %gprel modifier.
20 (md_apply_fix): Handle BFD_RELOC_MIPS16_GPREL.
22 * config/tc-mips.c (append_insn): Output jump instruction as a
23 pair of 2 byte instructions, rather than as a single 4 byte
27 * config/tc-mips.c (md_begin): Correct test of mips_5900.
28 (mips_ip): Don't check INSN_ISA for a macro.
34 * config/tc-mips.c: added r5900.
39 * itbl-ops.c, itbl-lex.l, itbl-parse.y, itbl-ops.h,
40 config/itbl-mips.h: Add copyright message and fix indentation.
44 * itbl-ops.c: New file. Add support for dynamically read
45 instruction registers, opcodes and formats. Build internal table
46 for new instructions and provide callbacks for assembler and
48 * itbl-lex.l, itbl-parse.y: Lex and yacc parsers for instruction
50 * itbl-ops.h: New file. Header file for itbl support.
51 * config/itbl-mips.h: New file. Mips specific definitions for
56 * config/tc-mn10200.c (md_assemble): If a constant operand won't
57 fit into the constant field of a relaxable operand, then it does
62 * config/tc-mn10200.c (md_estimate_size_before_relax): Treat
63 a jsr target in a different section just like a jsr to
68 * config/tc-mips.c (mips_fix_adjustable): Don't adjust relocations
69 against any mips16 symbols, not just externally visible ones.
70 (md_apply_fix): Corresponding change.
74 * config/tc-mips.c (mips16_ip): Accept floating point registers in
75 the operand of the exit instruction.
79 * symbols.c (resolve_symbol_value): If we leave an equated symbol
80 as O_symbol, copy over the segment.
84 * config/tc-mips.c (md_apply_fix): If we aren't adjusting this
85 fixup to be against the section symbol, adjust the value
88 * symbols.c (resolve_symbol_value): Don't change X_add_number for
90 * write.c (write_relocs): Avoid looping on equated symbols.
91 Adjust fx_offset by X_add_number for each symbol.
92 * config/obj-coff.c (do_relocs_for): Avoid looping on equated
94 (fixup_segment): Add a loop to track down equated symbols and
95 adjust fx_offset appropriately.
99 * config/tc-mn10200.c (md_relax_table): Add entries to allow
101 (md_convert_frag): Handle jmp->bra relaxing.
102 (md_assemble): Handle jmp->bra relaxing.
103 (md_estimate_size_before_relax): Likewise.
107 * config/tc-i386.c (i386_align_code): Add comments explaining the
112 * config/tc-sparc.c (enforce_aligned_data): New static variable.
113 (sparc_cons_align): Don't do anything unless enforce_aligned_data
115 (md_longopts): Add "enforce-aligned-data".
116 (md_show_usage): Mention --enforce-aligned-data.
117 * doc/c-sparc.texi (Sparc-Aligned-Data): New node; document
118 enforce-aligned-data.
120 * config/tc-ppc.c (md_pseudo_table): If OBJ_XCOFF, add "long",
122 (ppc_xcoff_cons): New static function.
124 * write.c (relax_segment): Give an error if a .space symbol is
127 * read.c (read_a_source_file): Don't handle mri_pending_align if
128 the handler is s_globl or s_ignore.
132 * config/tc-d10v.c (find_opcode): Remove unused variable "numops".
134 * config/tc-tic80.c: Many additions to previous placeholder file.
135 * config/tc-tic80.h: Ditto.
140 * config/tc-i386.c (i386_align_code): Improve the nop patterns.
144 * config/tc-mips.c (mips_fix_adjustable): New function.
145 * config/tc-mips.h (tc_fix_adjustable): Call mips_fix_adjustable.
146 (mips_fix_adjustable): Declare.
149 * read.c (read_a_source_file): After doing an mri_pending_align,
150 adjust the line_label if there is one.
151 (s_space): Set mri_pending_align if an odd number of bytes were
157 * config/tc-d10v.h (md_do_align): Add this hook to call
158 d10v_cleanup() when a ".align" is detected. Fixes PR11487.
160 * config/tc-d10v.c (find_opcode): Correctly calculate
161 branch displacement when .aligns are present.
166 * config/tc-mn10200.c (md_relax_table): Define.
167 (md_convert_frag): Implement.
168 (md_assemble): Handle relaxable operands/instructions correctly.
169 (md_estimate_size_before_relax): Implement.
170 * config/tc-mn10200.h (TC_GENERIC_RELAX_TABLE): Define.
174 * config/tc-mips.c (append_insn): Give an error for jumps to a
176 (md_apply_fix): Make a branch to an odd address an error rather
179 * config/tc-mips.c (md_convert_frag): If the user explicitly
180 requested an extended opcode, pass warn as true to mips16_immed.
182 * config/tc-mips.c (mips16_ip): Handle a missing expression like
183 an explicit 0, so that explicitly extended instructions work
188 * ecoff.c (ecoff_build_symbols): Don't generate a local ECOFF
189 symbol for a common symbol.
193 Patch presumed to have been checked in awhile ago but wasn't.
195 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
196 (relax_segment, case rs_org): Move code inside braces. Move locals
197 target,after inside too.
198 (relax_segment, case rs_machine_dependent): Guts moved to ...
199 (relax_frag): New function.
200 Call md_prepare_relax_scan if defined.
204 * config/tc-m68k.c (m68k_ip): Reject pc-relative addresses for the
205 'p' operand specifier.
209 * config/tc-m68k.c (HAVE_LONG_BRANCH): New macro, returns true for
210 m68k family cpus which support long branch addressing modes.
211 (m68k_ip, md_convert_frag_1, md_estimate_size_before_relax,
212 md_create_long_jump): Use it.
216 * config/tc-mips.c (md_begin): Don't set SEC_ALLOC or SEC_LOAD for
217 the .reginfo or .MIPS.options section if configured for an
220 * config/tc-mips.c (md_begin): Don't set interlocks for
225 * read.c (read_a_source_file): Make sure the symbol ends with
226 whitespace before checking whether the next character is '='.
230 * config/tc-i386.c (sco_id): Moved from here...
231 * config/obj-elf.c (sco_id): ...to here. Adding the identifier
232 really is an SCO ELF specific thing, not just a SCO x86 specific
238 * configure.in (tic80-*-*): Don't require 'coff'.
239 * configure: Regenerate.
244 * read.c (emit_expr): Check for overflow of a negative value
246 * write.c (fixup_segment): Likewise.
247 * config/obj-coff.c (fixup_segment): Likewise.
249 * config/tc-m68k.c (struct label_line): Define.
250 (labels, current_label): New static variables.
251 (md_assemble): Mark current_label as text, and clear it.
252 (m68k_frob_label): New function.
253 (m68k_flush_pending_output): New function.
254 (m68k_frob_symbol): New function.
255 * config/tc-m68k.h (tc_frob_label): Define.
256 (md_flush_pending_output): Define.
257 (tc_frob_symbol): Don't warn, just call m68k_frob_symbol.
258 (tc_frob_coff_symbol): Likewise.
260 * read.c (read_a_source_file): When defining a macro in MRI mode,
261 don't add the symbol to the symbol table.
265 * config/tc-mn10300.c (tc_gen_reloc): Handle sym1-sym2 fixups
266 here since fixup_segment doesn't (linkrelax is set).
267 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
271 * config/tc-mn10200.c (md_assemble): Tweak fx_offset for pc-relative
276 * config/tc-hppa.c (struct hppa_fix_struct): Tweak fx_r_field's type
277 to avoid warnings with the native HP compiler.
278 (fix_new_hppa): Similarly for the r_type argument.
279 (pa_build_unwind_subspace, hppa_elf_mark_end_of_function): Enclose
280 in an #if OBJ_ELF to keep gcc -Wall quiet.
281 (md_apply_fix): Always initialize "result".
283 * config/tc-mn10200.c (md_assemble): Generate relocations.
287 * config/tc-m68k.c (s_even): Adjust the alignment of the current
292 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
293 pass on the new file hook.
295 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
296 !alpha_force_relocation, as local LITERALs can be adjusted to be
297 relative to the section.
301 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
303 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
304 aligned to odd boundary.
305 (tc_frob_coff_symbol): Define.
307 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
309 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
311 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
314 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
316 (md_assemble): If the size is 'B', set fx_signed.
317 (md_apply_fix_2): Use fx_signed when checking for overflow.
319 * write.h (struct fix): Add fx_signed field.
320 * write.c (fix_new_internal): Initialize fx_no_overflow and
322 (fixup_segment): Use fx_signed when checking for overflow.
323 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
324 fx_signed when checking for overflow.
328 * NOTES, NOTES.config: Removed. These are rarely, if ever,
329 updated, and all the useful information is in doc/internals.texi.
332 * read.c (read_a_source_file): Check for conditional operators
333 before doing an MRI pending alignment.
334 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
335 (tc_conditional_pseudop): Define.
336 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
337 * doc/internals.texi (CPU backend): Describe
338 tc_conditional_pseudoop.
341 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
343 (md_begin): Check m68k_rel32_from_cmdline before setting
345 (m68k_mri_mode_change): Likewise.
346 (md_longopts): Add --disp-size-default-16 and
347 --disp-size-default-32.
348 (md_parse_option): Handle new options.
349 (md_show_usage): Mention new options.
350 * doc/c-m68k.texi (M68K-Opts): Document new options.
353 * config/tc-m68k.c (m68k_index_width_default): New static
355 (m68k_ip): Use m68k_index_width_default to set the size of a base
356 register whose size was not given.
357 (md_longopts): Add --base-size-default-16 and
358 --base-size-default-32.
359 (md_parse_option): Handle new options.
360 (md_show_usage): Mention new options.
361 * doc/c-m68k.texi (M68K-Opts): Document new options.
363 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
365 * configure.in: Recognize mips-*-linux* target.
366 * configure: Rebuild.
368 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
369 work if valueT is only 32 bits.
371 * config/tc-mips.c: Throughout, check target_big_endian rather
373 (byte_order): Remove.
374 (mips_init_after_args): Remove.
375 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
376 (mips_init_after_args): Don't declare.
377 (tc_init_after_args): Don't define.
379 * config/tc-mips.h (tc_frob_after_relocs): Define if
381 (mips_elf_final_processing): Likewise.
382 (ELF_TC_SPECIAL_SECTIONS): Likewise.
387 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
388 sld.{h,w} and the operand is relocatable, adjust the adend by
389 shifting it right one bit.
394 * read.c (read_a_source_file): Check mri_pending_align after
395 checking for a macro. From Ronald F. Guilmette
398 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
400 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
401 add segment argument. If OBJ_ELF, treat a relocation against a
402 symbol in a linkonce section like a relocation against an external
404 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
408 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
412 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
413 includes config.h instead of host.h, tc.h instead of tp.h, and
414 targ-env.h instead of target-environment.h.
415 Also, obj-format.h includes targ-cpu.h instead of
418 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
419 * configure: Rebuild with autoconf.
420 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
421 (TARGET_FORMAT): Define to "coff-tic80".
422 * config/tc-tic80.c: New file for TIc80 support.
423 * config/tc-tic80.h: New file for TIc80 support.
428 * doc/as.texinfo (M): Mention explicitly that -M changes macro
433 * write.c (adjust_reloc_syms): If the fixup symbol has been
434 equated to an undefined symbol, convert the fixup to being against
435 the target symbol. Remove obsolete code handling a special case
440 * mpw-make.sed: Use NewFolderRecursive for installation.
445 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
449 * config/tc-d10v.c (md_assemble): Fix bug which caused
450 second instruction in a line to be case sensitize. PR11312
455 * config/tc-mn10200.c (mn10200_insert_operand): Don't
456 range check operands with MN10200_OPERAND_NOCHECK set.
457 (check_operand): Likewise.
461 * config/tc-mips.c: Undo part of last Friday's alignment changes.
462 (md_begin): Always align the text section to a four byte
464 (append_insn): Remove call to record_align.
466 * config/tc-mips.c (insn_label): Remove.
467 (struct insn_label_list): Define.
468 (insn_labels, free_insn_labels): New static variables.
469 (mips_clear_insn_labels): New static function.
470 (append_insn): Mark all mips16 text labels, and make them odd.
471 Handle all labels after emitting a nop, not just one. Call
472 mips_clear_insn_labels rather than just clearing insn_label.
473 (mips_emit_delays): Add insns parameter, and use it to decide
474 whether to mark mips16 labels. Handle all labels, not just one.
475 Force mips16 labels to be odd. Change all callers.
476 (mips16_immed): Don't check for an odd branch target.
477 (md_apply_fix): Don't check mips16 mode for a branch reloc.
478 (mips16_extended_frag): Ignore the low bit in a branch target.
479 (md_convert_frag): Likewise.
480 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
482 (mips_align, mips_flush_pending_output, s_cons): Likewise.
483 (s_float_cons, s_gpword): Likewise.
484 (s_align): Use insn_labels rather than insn_label.
485 (s_cons, s_float_cons, s_gpword): Likewise.
486 (mips_frob_file_after_relocs): New function.
487 (mips_define_label): Rewrite to add to insn_labels list.
488 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
489 * ecoff.c (ecoff_build_symbols): If the size of a function comes
490 out odd, increment it.
492 * config/tc-mips.c (append_insn): Only update prev_insn when not
493 reordering if place is NULL.
495 * config/tc-mips.c (mips16_ip): Check for a missing expression
496 when using the register indirect addressing mode.
500 * config/tc-mn10200.c (mn10200_insert_operand): Don't
501 check 24bit operands for overflow.
502 (check_operand): Likewise.
506 * doc/as.texinfo (Section): Document how to use the .section
507 pseudo-op for COFF and ELF.
511 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
515 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
516 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
517 store them. Adjust other RELAX_MIPS16 macros.
518 (RELAX_MIPS16_DSLOT): Define.
519 (RELAX_MIPS16_JAL_DSLOT): Define.
520 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
521 handling of whether previous instruction has a fixup. Set
522 prev_insn_reloc_type.
523 (mips_no_prev_insn): Clear prev_insn_reloc_type.
524 (mips16_extended_frag): Use the right base address for a PC
525 relative add or load.
526 (md_convert_frag): Likewise. If a PC relative add or load is
527 used, record the alignment for the section.
531 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
532 linkonce section into a reloc against the section symbol.
534 * config/tc-mips.c (mips16_macro): Remove nop instructions after
537 * config/tc-mips.c (md_begin): If configured for an embedded ELF
538 system, don't set the section alignment to 2**4.
539 (s_change_sec): Likewise.
540 (append_insn): Call record_alignment for the section.
541 (md_section_align): Don't align the section size for an embedded
546 * write.c (adjust_reloc_syms): Make sure that symbols are
547 resolved; expression symbols may have been skipped.
548 * config/obj-coff.c (fixup_segment): Likewise.
552 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
553 BFD_RELOC_24_PLT_PCREL relocation.
554 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
559 * config/tc-d10v.c (write_2_short): Remove code that called
560 parallel_ok() when the programmer specified parallel instructions.
565 * config/tc-mn10300.c (md_assemble): Update to handle endianness
568 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
569 * config/tc-mn10300.c (md_assemble): Likewise.
573 * config/tc-mips.c (append_insn): Make sure there is enough room
574 in a frag after a mips16 instruction to switch it with a jump
577 * config/tc-mips.c (mips16_extended_frag): Give an error for an
578 attempt to use a non absolute symbol in an extending frag.
582 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
583 * config/tc-mn10200.h: Likewise.
587 * app.c (do_scrub_chars): At the end of a C comment, pass space to
588 UNGET rather than PUT. Set old_state before setting state to -2.
590 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
591 when extending because the value is exactly maxtiny + 1.
593 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
594 arguments, and store them. Adjust other RELAX_MIPS16 macros.
595 (RELAX_MIPS16_USER_SMALL): Define.
596 (RELAX_MIPS16_USER_EXT): Define.
597 (mips16_small, mips16_ext): New static variables.
598 (append_insn): Pass mips16_small and mips16_ext to
600 (mips16_ip): Set mips16_small and mips16_ext.
601 (mips16_immed): Don't check mips16_autoextend.
602 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
604 * write.c (write_relocs): Print an error for an out of range
605 fixup, rather than calling abort.
607 * as.c (main): Unlink the output file if there are errors while
608 generating the fixups.
612 * config/tc-mips.c (mips16_extended_frag): Don't call
614 (md_convert_frag): Call resolve_symbol_value before calling
615 S_GET_VALUE, and don't add in the frag address.
617 * config/tc-mips.c (mips16_immed): Add file and line parameters,
618 and use them when reporting errors. Change all callers.
622 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
623 Remove '$' prefixing for registers.
628 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
629 is initialized before using it.
630 (d10v_cleanup): No longer uses its argument, so make it void.
632 * config/tc-d10v.h (d10v_cleanup): Change prototype.
637 * write.c (fixup_segment): Don't discard the symbol for a PC
638 relative fixup to an absolute symbol.
643 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
644 with multiple sections.
649 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
650 no-mips-16 to no-mips16.
651 (s_mipsset): Accept .set mips16 and .set nomips16.
655 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
656 argument, and check for +/- constant following the suffix, folding
657 it into the expression.
658 (ppc_elf_cons): Change ppc_elf_suffix calls.
659 (md_assemble): Ditto.
660 (shlib): Replace boolean mrelocatable with enumeration shlib.
661 (md_parse_option): Discriminate between PIC style shared libraries
663 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
668 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
669 change the comment characters.
671 * config/tc-ppc.c (comment_chars): Delete in favor of
673 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
675 (ppc_comment_chars): Select appropriate comment chars by default.
676 (msolaris): New flag for -m{,no-}solaris.
677 (md_parse_option): Recognize -K pic. Add support for
679 (md_show_usage): Update.
680 (md_begin): Do not set ELF flags if Solaris.
681 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
682 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
686 * as.c (main): Correct handling of flag_always_generate_output.
690 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
691 fx_offset, not fx_addnumber.
693 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
698 * config/tc-i386.c: Correct misspelling: balenced to balanced.
702 * config/tc-mips.c (md_section_align): Check for an alignment of
703 4, not an alignment of 16. Corrects August 7 patch.
707 * configure, conf.in: Rebuild with autoconf 2.12.
709 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
710 alignment is specified.
712 Add support for mips16 (16 bit MIPS implementation):
713 * config/tc-mips.c: Extensive additions for mips16 support, not
715 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
716 (insn_uses_reg): Change last parameter to an enum.
717 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
718 (md_relax_frag): Define as mips_relax_frag.
719 (mips_relax_frag): Declare.
720 (struct mips_cl_insn): Add use_extend and extend fields.
721 (tc_fix_adjustable): Define.
722 * config/obj-elf.h (S_GET_OTHER): Define.
723 (S_SET_OTHER): Define.
727 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
728 o and p operand specifiers.
732 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
733 (relax_segment, case rs_org): Move code inside braces. Move locals
734 target,after inside too.
735 (relax_segment, case rs_machine_dependent): Guts moved to ...
736 (relax_frag): New function.
737 Call md_prepare_relax_scan if defined.
738 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
743 * config/tc-mn10300.c (address_registers): Use '$' as register
744 prefix instead of '%'.
745 (data_registers, other_registers, md_assemble): Likewise.
747 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
748 (data_registers, other_registers, md_assemble): Likewise.
750 * config/tc-mn10300.c (md_assemble): Correctly determine the
751 correct location and type for each relocation.
752 (md_pcrel_from): Simplify.
756 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
757 converted into branch around branch.
761 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
762 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
764 * config/tc-mn10300.c (md_begin): Set linkrelax.
765 (md_assemble): Create fixups as needed.
766 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
771 * config/tc-d10v.c (parallel_ok): When automatically converting
772 serial ops to parallel, do not consider a branch as the first
778 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
782 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
784 (check_operand): Likewise.
789 * config/tc-d10v.c (parallel_ok): Branch and link instructions
791 (write_2_short): Call parallel_ok to check whether two short
792 instructions the user requested execute in parallel, can be
797 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
798 allow a branch and link in parallel with an exe instruction.
802 * doc/c-d10v.texi: Add info on @word modifier.
807 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
808 operands are assumed to be 32bits. Use "bits" field to hold the
809 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
810 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
813 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
814 of a MN10300_OPERAND_SPLIT operand by operand->shift.
816 * config/tc-mn10300.c (mn10300_insert_operand): Handle
817 MN10300_OPERAND_SPLIT.
821 * config/tc-mn10300.c (md_assemble): Insert operands into
822 the extension part of the instruction if necessary.
823 (mn10300_insert_operand): Accept pointer to extension word
824 argument. Make insn a pointer argument too. Return type
825 is now void. All callers changed.
829 * config/tc-mn10300.c (mn10300_insert_operand): Handle
830 repeated register operands.
834 * doc/as.texinfo: Added section on reporting bugs.
836 * config/tc-alpha.c: Change uses of void * to PTR. Change the
837 alpha_macro emit field to expect a const argument, and change the
838 arg field to be const. Fix some spacing to follow the GNU
843 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
844 (pca56) and 21264 (ev6) cpus.
845 (md_apply_fix): Private relocation types are now negative.
846 (alpha_force_relocation): Likewise.
847 (tc_gen_reloc): Likewise.
848 (emit_insn): Likewise.
849 (emit_ldXu): Do the right thing when the hardware can do byte insns.
850 (emit_stX): Likewise.
851 (emit_sextX): Likewise.
855 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
856 a symbol found in a reloc.
858 * symbols.c (resolve_symbol_value): Improve the error message if
859 an undefined symbol is used in an expression.
863 * doc/internals.texi: Rewrite, and add a lot of documentation.
864 * doc/Makefile.in (internals.info): New target.
869 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
870 against weak symbols.
874 * config/tc-v850.c (md_assemble): Don't lose for relaxable
877 * config/tc-v850.c (md_convert_frag): Make sure we insert the
878 fixup at the right address within the frag.
880 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
881 to an absolute value, instead increment it as needed.
883 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
884 * config/tc-v850.c: Fix some indention problems.
885 (md_relax_table): Define for D9->D99 branch displacement
887 (md_convert_frag): Do something useful instead of aborting.
888 (md_estimate_size_before_relax): Likewise.
889 (md_assemble): Note if the matching instruction has a relaxable
890 operand. If it does, allocate frag with frag_var and don't
897 * config/tc-d10v.h (md_cleanup): New function. This is needed to
898 write out any buffered instructions when a ".end" is found.
903 * read.c (read_a_source_file): New hook md_cleanup().
907 * write.c (fix_new_exp): Use make_expr_symbol to build an
908 expression symbol for a complex fixup.
913 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
914 hacks to improve parsing of complex hi, lo, zda, etc
916 (md_assemble): Don't demand and eat a trailing ')' after finding
917 a v850 relocation prefix. Sign extend the constant in a
918 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
920 (parse_cons_expression_v850): Don't eat a trailing ')' after
921 finding a v850 relocation prefix.
923 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
924 (TC_CONS_FIX_NEW): Likewise.
925 * config/tc-v850.c (parse_cons_expression_v850): New function.
926 (cons_fix_new_v850): Likewise.
928 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
933 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
934 for an unknown relocation type.
939 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
944 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
945 differently for movea & sst/sld insns.
951 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
957 * obj.h (struct format_ops): Add frob_file_after_relocs field.
958 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
959 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
960 frob_file_after_relocs field.
961 * config/obj-elf.c (elf_format_ops): Likewise.
962 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
967 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
968 specially if not ELF.
969 (md_apply_fix): Handle BFD_RELOC_64.
970 (tc_gen_reloc): Handle BFD_RELOC_64.
972 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
973 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
976 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
977 Don't check S_GET_OTHER.
981 * config/tc-mips.c (mips_ip): Accept an odd floating point
982 register with l.s or s.s.
984 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
986 (obj_aout_type): New static function.
990 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
995 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
996 and sdaoff expressions.
998 * write.c (fixup_segment): Don't add symbol value to addend if
1000 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
1003 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
1004 (md_pcrel_from_section): Likewise.
1005 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
1009 * config/tc-v850.c (md_pcrel_from): Delete unused function.
1010 (md_pcrel_from_section): New function.
1011 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
1016 * config/tc-mips.c (load_register): Add cast to offsetT when using
1021 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
1022 (elf_frob_file_after_relocs): ... here. New function.
1023 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
1024 * write.c (write_object_file): Call *frob_after_relocs after the
1025 call to write_relocs.
1027 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
1029 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
1030 symbol, as this messes up linking. Instead, expand the recursive call
1031 inline and change up the appropriate bits to get the 0x8000 offset
1032 in the reloc addend.
1036 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
1037 two symbols in the same segment to be adjusted.
1039 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
1040 * configure: Rebuild.
1044 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
1046 (ppc_elf_validate_fix): Ditto.
1047 (md_assemble): Ditto.
1049 (ppc_pe_section): Ditto.
1050 (ppc_frob_symbol): Ditto.
1054 * config/mn10300.c (md_assemble): Pass an extra shift count
1055 to mn10300_insert_operand based on the opcode format.
1056 (mn10300_insert_operand): Accept and use extra shift count
1059 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
1060 formats rather than hard-coded constants.
1062 * config/tc-mn10300.c (md_assemble): Format D5 instructions
1063 are 7 bytes long. Write out instructions in big-endian format.
1067 * config/tc-mn10300.c (md_assemble): Tweak further so
1068 that all instructions are parsed correctly.
1072 * as.h: Include libiberty.h.
1073 (xmalloc, xrealloc): Don't declare.
1074 * as.c: Don't include libiberty.h.
1075 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
1076 * config/tc-mips.c: Likewise.
1077 * messages.c: Likewise.
1078 (xstrerror): Don't declare.
1079 * xmalloc.c: Remove.
1083 * config/tc-mn10300.h (pre_defined_registers) Remove.
1084 (system_registers, cc_names): Likewise.
1085 (address_registers, data_registers, other_registers): New register
1087 (register_name, system_register_name, cc_name): Remove.
1088 (mn10300_reloc_prefix): Likewise.
1089 (data_register_name): New function.
1090 (address_register_name, other_register_name): Likewise.
1091 (md_assemble): Rough cut at parsing operands. Remove lots of
1093 (md_apply_fix3): Disable for now.
1097 * config/tc-m68k.c (select_control_regs): New function, extracted
1098 out of m68k_init_after_args.
1099 (m68k_init_after_args): Use it.
1100 (mri_chip): Use it here as well to update set of allowed control
1105 * config/obj-elf.c (elf_begin): New function.
1106 (obj_elf_section): Add the section symbol to the symbol table.
1107 * config/obj-elf.h (obj_begin): Define.
1108 (elf_begin): Declare.
1109 * as.c (perform_an_assembly_pass): Call obj_begin if it is
1114 * config/obj-coff.c (fixup_segment): Subtract the section address
1115 from a PC relative reloc if TC_M68K.
1119 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
1120 .uaxword available even if not OBJ_ELF.
1121 (md_atof): Remove unused local variable wordP.
1125 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
1126 for Matsushita MN10x00 support.
1127 * configure.in: Recognize mn10x00-*-*
1128 * configure: Rebuilt.
1132 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
1133 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
1134 (load_expression): track clobbering of base reg before jmp/jsr.
1135 (s_alpha_file): pass case_hack flags and source filename via
1136 symbol table to bfd.
1137 * tc-alpha.h (TC_CONS_FIX_NEW): define
1141 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
1142 * configure: Rebuild.
1146 * read.c (s_macro): Warn if a macro has the same name as a
1148 (s_space): In m68k MRI mode, align to a word boundary.
1149 * macro.c (define_macro): Add namep parameter. Change all
1151 * macro.h (define_macro): Update declaration.
1153 * as.c (show_usage): Print bug report address.
1154 (parse_args): Change version printing to match current GNU
1156 * gasp.c (show_usage): Print bug report address.
1157 (main): Change version printing to match current GNU standards.
1159 * config/tc-m68k.c (init_table): Correct access control unit
1162 * config/tc-alpha.c: Add some static function prototypes.
1163 (alpha_macros): Move to top of file. Make static.
1164 (alpha_num_macros): Move to top of file.
1169 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
1175 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
1176 and just call sprintf_vma.
1180 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
1184 * read.c (float_cons): Call md_flush_pending_output if it is
1190 * config/tc-d10v.c (md_operand): Created. Allows operands to
1192 * config/tc-d10v.h (md_operand): Undefined.
1197 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
1198 (md_assemble): A fixup width of '3' means a 1 byte reloc.
1202 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
1203 reloc for the i960 for a reloc in the same section. This undoes
1204 one of the two changes made Aug 19.
1208 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
1209 symbols to the position of the debugging information.
1213 * expr.c (expr): Always use unsigned right shifts for >>.
1217 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
1218 slots when processing BL fixups.
1220 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
1221 on the first half of the instruction.
1225 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
1226 stabs expression, rather than giving an error.
1228 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
1233 * config/tc-mips.c (append_insn): Fill in the value for a constant
1234 jump, rather than creating a reloc.
1238 * config/tc-mips.c (append_insn): Don't swap an instruction which
1239 sets a condition code with an instruction which uses a condition
1241 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
1244 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
1245 (mips_ip): Give a better error message if the ISA level is wrong.
1246 (md_parse_option): Recognize -mcpu=[v][r]5000.
1250 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
1252 (load_register): Ensure hi32 bits are not lost during lo32bit
1253 processing. Fix shift offset that was overflowing into the next
1254 instruction field. Add code to generate shorter sequences for
1255 constants with a single contiguous seqeuence of ones.
1260 * config/tc-d10v.c (d10v_dot_word): New function to support
1261 "@word" with the word pseudo-op.
1262 (md_apply_fix3): Cleanup and changes to support correct sizes
1263 for 16 and 18-bit relocs.
1268 * configure.in (sparc-*-aout): Set `em'.
1269 * configure: Regenerated.
1270 * config/te-sparcaout.h: New file.
1271 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
1272 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
1273 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
1274 (SPECIAL_CASE_{SETSW,SETX}): Define.
1275 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
1276 (md_begin): Delete setting of `target_big_endian'.
1277 (output_insn): New function.
1278 (md_assemble): Rewrite. Add `setx' support.
1279 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
1281 (md_atof): Add little endian support.
1282 (md_number_to_chars): Likewise.
1283 (md_apply_fix): Likewise.
1284 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
1285 (md_parse_option): Likewise.
1286 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
1290 * ecoff.c (ecoff_new_file): New function.
1291 * ecoff.h (ecoff_new_file): Declare.
1292 * config/obj-ecoff.h (obj_app_file): Define.
1296 * config/tc-alpha.c (load_expression): Bias the .lit8 section
1297 symbol by 32k so that our 16-bit signed offset can address the
1302 * config/tc-mips.c (load_register): Remove unused variable tmp.
1306 * config/tc-mips.c (load_register): Remove unnecessary code that
1307 was causing the high 32bits of 64bit constants to be lost.
1312 * config/tc-d10v.c: Added changes to support function
1313 pointers and "@word" syntax.
1319 * config/tc-v850.c: Remove commented out and #if 0'd code.
1320 (v850_reloc_prefix): Provide prototype.
1321 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
1323 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
1324 (md_assemble): Add default to case statement.
1328 * config/tc-v850.c (md_assemble): Compute size of the instrction
1331 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
1336 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
1337 routines to fetch/store the updated instruction from/to memory.
1338 (v850_insert_operand): If the operand has a specialized insert
1343 * config/tc-v850.c (reg_name_search): Align calling convention to
1344 be like identical function found in tc-ppc.c.
1345 (get_reloc): Removed.
1346 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
1347 (md_assemble): emit fixups.
1348 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
1350 (md_apply_fix3): handle fixups/relocs.
1351 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
1357 * configure.in (sh-*-elf*): New target.
1358 * config/tc-sh.h (TARGET_ARCH): Define.
1359 (WORKING_DOT_WORD): Define.
1360 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
1361 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
1362 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
1363 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
1364 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
1365 (SUB_SEGMENT_ALIGN): Likewise.
1366 (RELOC_32): Don't define.
1367 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
1368 (target_big_endian): Declare if OBJ_ELF.
1369 (TARGET_FORMAT): Define if OBJ_ELF.
1370 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
1372 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
1373 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
1374 (struct sh_count_relocs): Define.
1375 (sh_count_relocs): New static function, broken out of
1376 sh_frob_file. Add BFD_ASSEMBLER code.
1377 (sh_frob_section): Likewise.
1378 (sh_frob_file): Call sh_frob_section.
1379 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
1380 call section_symbol rather than seg_info (seg)->dot.
1381 (md_section_align): Add OBJ_ELF version.
1382 (SWITCH_TABLE_CONS): Define.
1383 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
1384 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
1385 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
1387 (struct reloc_map): Define if not BFD_ASSEMBLER.
1388 (coff_reloc_map): Likewise.
1389 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
1390 (tc_gen_reloc): New function if BFD_ASSEMBLER.
1391 * write.c (write_relocs): Ifdef out fx_where test which triggers
1392 inappropriately for SH ELF.
1393 (write_object_file): Call tc_frob_file_before_adjust and
1394 obj_frob_file_before_adjust if they are defined.
1396 * write.c (write_object_file): Use BFD_RELOC_16, not
1397 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
1399 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
1405 * config/tc-d10v.c (find_opcode): Fix problem with calculating
1406 branch sizes in across sections.
1412 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
1417 * config/tc-v850.c (md_begin): Deal with end of opcode
1424 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
1425 the wrong opcode for cases like st2w where there are many forms
1426 of the same instruction.
1431 * expr.c (operand): If md_parse_name is defined, call it before
1432 calling symbol_find_or_make.
1433 * config/tc-ppc.h (md_parse_name): Define.
1434 (ppc_parse_name): Declare.
1435 * config/tc-ppc.c (reg_name_search): Add regs and regcount
1437 (register_name): Update call to reg_name_search.
1438 (cr_operand): New static variable.
1439 (cr_names): New static const array.
1440 (ppc_parse_name): New function.
1441 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
1442 cr_operand before calling expression.
1446 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
1447 hppa_gen_reloc_type call.
1452 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
1453 on addresses. Improved error messages.
1454 * doc/c-d10v.texi: Added docs for register pairs.
1458 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
1464 * ecoff.c (init_file): Initialize fMerge to 1.
1465 (add_file): Restore old file merging code, but only merge files if
1467 (ecoff_directive_loc): Clear fMerge field of current file.
1468 (ecoff_generate_asm_lineno): Likewise.
1473 * doc/c-d10v.texi: Fix typo.
1479 * config/tc-v850.c (md_assemble): Correct bit masking for
1480 hi and lo expressions.
1482 * config/tc-v850.c (md_assemble): Rough cut at demanding
1483 "ep" or "r30" in sst and sld instructions.
1484 (md_apply_fix3): Don't abort. Just warn that we don't
1487 * config/tc-v850.c (CC_NAME_CNT): Define.
1488 (cc_name): New function.
1489 (md_assemble): Handle V850_OPERAND_CC correctly.
1491 * config/tc-v850.c (md_assemble): Don't forget to initialize
1494 * config/tc-v850.c (reg_name_search): Generalize to search
1495 any given register table.
1496 (register_name): Pass appropriate table and size to reg_name_search.
1497 (system_register_name): New function.
1498 (SYSREG_NAME_CNT): Define.
1499 (md_assemble): Handle operands which are system registers.
1501 * config/tc-v850.c (md_assemble): If we find a register, but the
1502 opcode doesn't want a register, then we don't have a match.
1503 (md_assemble): Get size of the instruction from the opcode table.
1508 * configure.in: Set and substitute HLDENV.
1509 * configure: Rebuild.
1510 * Makefile.in (HLDENV): New variable.
1511 (as.new): Use $(HLDENV).
1513 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
1514 if an array dimension is not known.
1519 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
1520 * doc/c-d10v.texi: Cleanup.
1526 * config/tc-v850.c: New file.
1527 * config/tc-v850.h: New file.
1528 * configure (v850-*-elf): New target.
1529 * configure.in (v850-*-elf): New target.
1535 * doc/c-d10v.texi: New file.
1536 * doc/all.texi: Added D10V stuff.
1537 * doc/as.texinfo: Added D10V stuff.
1541 * config/tc-d10v.c: All references to defined symbols should
1542 now use the optimal instruction. .float and .double now work.
1547 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
1548 section address for the i960 as is done for the i386.
1552 * mpw-config.in: Add wildcards for config matching, add mips-*-*
1553 case, forward-include bfd/elf-bfd.h.
1558 * config/tc-d10v.c: Add additional information to the opcode
1559 table to help determinine which instructions can be done
1565 * config/tc-arm.c: Major changes to add Thumb support, with lots
1567 Reverted to INSN_SIZE macro, rather than insn_size variable.
1568 (insns): Added ARM "bx" instruction support.
1569 (tinsns): Added Thumb instruction definition structure.
1570 (arm_tops_hsh): Added hash structure for Thumb opcodes.
1571 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
1572 (opcode_select,s_arm,s_thumb,s_code): Added.
1573 (decode_shift): Allow upper-case RRX.
1574 (do_ldst): Simpler halfword support.
1575 (do_ldmstm): Improved.
1576 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
1577 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
1578 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
1579 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
1580 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
1582 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
1583 (md_parse_option): Add support for -mthumb.
1584 (md_show_usage): Updated to reflect new command line option.
1585 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
1586 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
1587 fragment information to be held.
1591 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
1592 non-null, but is a constant.
1593 (fix_new_arm): Call make_expr_symbol to make the expression symbol
1594 so that error reporting will work correctly.
1598 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
1599 against weak symbols.
1603 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
1604 (ppc_force_relocation): Declare if OBJ_XCOFF.
1605 * config/tc-ppc.c (ppc_force_relocation): New function if
1610 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
1615 * config/tc-d10v.c: Fix problem with relocs.
1620 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
1621 with nops if not in data_section or bss_section.
1625 Add support for openVMS/Alpha.
1626 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
1627 * config/obj-evax.c: New file.
1628 * config/obj-evax.h: New file.
1629 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
1631 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
1632 defined. Add case for bfd_target_evax_flavour.
1633 * config/vms-a-conf.h: New file.
1634 * conf-a-gas.com: New file.
1635 * configure.in: Add target alpha-*-*vms*.
1636 * configure: Rebuild.
1637 * makefile.vms: New file.
1638 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
1639 Don't call ffs on openVMS/Alpha.
1643 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
1644 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
1646 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
1647 (NEED_DECLARATION_SBRK): New macro.
1648 * configure, conf.in: Rebuild.
1649 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
1650 (strstr): Declare if NEED_DECLARATION_STRSTR.
1651 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
1655 * symbols.c (resolve_symbol_value): Handle addition or subtraction
1656 by a constant before entering the main switch. Reject attempts to
1657 apply an arithmetic function to non-absolute symbols, except for
1658 the special case of subtraction of two symbols in the same
1661 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
1662 not to more than a 16 byte boundary.
1664 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
1666 (tc_gen_reloc): If out of memory call as_fatal rather than
1667 assert. If no howto found, call as_bad_where rather than
1668 as_fatal. Change the error message slightly. Set howto to a
1669 non-NULL value in order to keep going.
1674 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
1679 * config/tc-h8300.c (get_specific): New operand "size" derived
1680 from ".b", ".w" and ".l" extensions. All callers changed. If
1681 the base instruction has no operands, then use the size to
1682 determine which specific instruction to use.
1686 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
1690 * config/tc-mips.c (md_section_align): Don't change addr if
1695 * config/tc-hppa.c: Revert yesterday's changes.
1700 * config/tc-d10v.c: Disable range checking on 16-bit values.
1705 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
1706 not just the ones that don't match.
1711 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
1716 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
1717 pre-cursor to adding Thumb support. Also added cpu_variant flag
1718 information to each of the asm_flg structures.
1719 (md_parse_option): Updated ARM7 parsing to allow 't' for
1720 thumb/halfword support, aswell as 'm' for long multiply.
1721 (md_show_usage): Updated help message.
1722 (md_assemble): Check that instruction flags are applicated to the
1723 current cpu variant.
1724 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
1725 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
1726 signextension instructions.
1727 (do_ldst): Generate halfword and signextension variants if
1728 mnemonic flags match.
1729 (ldst_extend): Do not allow shifts in the offset field of halfword
1730 or signextension instructions.
1731 (validate_offset_imm): Provide check on halfword and signextension
1733 (add_to_lit_pool): Merge identical literal pool values.
1737 * config/tc-hppa.c (selector_table): Add 'E' selector.
1738 (cons_fix_new_hppa): Don't coke on e_esel.
1739 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
1740 to help generate exception handling tables.
1741 (md_apply_fix): Don't try to apply fixups with an e_esel
1743 (hppa_fix_adjustable): Fixups with e_esel selectors
1748 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
1754 * config/tc-d10v.c: Added lots of error checking. Added hacks
1755 to support accumulator shifts.
1760 * symbols.c (S_SET_EXTERNAL): Let .weak override.
1761 (S_CLEAR_EXTERNAL): Likewise.
1762 (S_SET_WEAK): Remove error; just let .weak override.
1767 * config/tc-d10v.c (md_assemble): Now handles multiline
1772 * config/tc-d10v.c: Fix packaging bug. Added range checking.
1773 Added kludge for divs instruction. Fixed minor problem with
1774 multiple text sections.
1775 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1779 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
1780 addresses to be right-shifted by 2.
1784 * config/tc-d10v.c: Many changes to get relocs working.
1785 (register_name): No longer creates a symbol for register names.
1786 (pre_defined_registers): moved to opcodes/d10v-opc.c.
1787 (d10v_insert_operand): Now works correctly for either container.
1788 * config/tc-d10v.h (d10v_cleanup): Declare.
1793 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
1794 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
1798 * tc-alpha.c: Patches to track current minimum alignment to reduce
1799 the number of fragments created with frag_align.
1800 (alpha_current_align): New static variable.
1801 (s_alpha_text): Reset alignment to 0.
1802 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
1803 (s_alpha_stringer, s_alpha_space): New functions.
1804 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
1805 (alpha_cons_align): New function to replace both of them.
1806 (emit_insn): Only align if alpha_current_align is less than 2;
1807 reset alpha_current_align to 2.
1808 (s_alpha_gprel32): Likewise.
1809 (s_alpha_section): New function. Basically duplicate the other
1810 alpha section change hooks. Only define for ELF.
1811 (s_alpha_float_cons): Simplify alignment handling.
1812 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
1813 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
1814 Don't define the s_alpha_cons pseudo-ops. Do define
1815 s_alpha_stringer and s_alpha_space pseudo-ops.
1816 (alpha_align): Skip if less than current default alignment. Set
1818 * tc-alpha.h (md_flush_pending_output): Remove.
1819 (md_cons_align): Add.
1821 * tc-alpha.c: Add oodles of function description comments.
1822 (md_bignum_to_chars): Remove; there are no callers.
1823 (md_show_usage): Mention some more variants.
1828 * configure.in (sparc-*-sysv4*): New target.
1829 * configure: Rebuild.
1831 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
1832 uaxword to use s_uacons.
1833 (sparc_no_align_cons): New static variable.
1834 (s_uacons): New static function.
1835 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
1838 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
1840 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
1841 5 with regard to shared libraries.
1842 * configure: Rebuild.
1844 * config/tc-m68k.c (m68k_ip): Use the correct length when
1845 allocating space for the unsupported architecture error message.
1850 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
1854 * config/tc-d10v.c: New file.
1855 * config/tc-d10v.h: New file.
1856 * configure (d10v-*-elf): New target.
1857 * configure.in (d10v-*-elf): New target.
1862 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
1866 * config/tc-alpha.c (alpha_align): Change fill parameter
1867 to a pointer. Take NULL as 0 or nop depending on section. Change
1869 (s_alpha_align): Rename local variables.
1871 * doc/as.texinfo (.align): Document action of omitted
1876 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
1877 when an unsupported PC relative reloc is seen, rather than calling
1880 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
1881 Instead, check state when '#' comment is seen.
1885 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
1887 (tc_gen_reloc): If fixup was changed to be PC relative, change
1888 reloc type accordingly. Use name of reloc in error message.
1890 * as.h: Don't define const or volatile.
1891 * flonum.h: Don't define const.
1893 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
1894 if fx_pcrel is set. Correct setting the addend case in the
1895 OBJ_ELF case (from Andreas Schwab
1897 (md_show_usage): Correct -mfc5200 to -m5200.
1901 * doc/c-m68k.texi: Document -m5200 flag.
1902 * doc/as.texinfo: Likewise.
1904 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
1909 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
1910 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
1914 * Makefile.in (VERSION): Set to cygnus-2.7.1.
1916 * Released binutils 2.7.
1920 * config/tc-mips.c (mips_ip): Only perform range check when
1921 dealing with O_constant expressions.
1925 * m68k-parse.h (m68k_register): Add new coldfile control
1928 * config/tc-m68k.c (mcf5200_control_regs): New variable,
1929 array of control registers for the coldfire.
1930 (cpu_of_arch): Added mcf5200.
1931 (archs): Added mcf5200.
1932 (init_table): Add new control registers.
1933 (m68k_ip): Added support for new control registers.
1934 (m68k_init_after_args): Likewise.
1936 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
1940 * read.h (is_it_end_of_statement): Declare.
1941 * read.c (is_it_end_of_statement): Remove declaration.
1943 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
1945 (md_assemble): Fix handling of @l with an unsigned constant. Add
1946 default case to reloc switch.
1948 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
1951 * ecoff.c (last_lineno): New static variable.
1952 (add_procedure): Set last_lineno.
1953 (ecoff_directive_loc): Likewise.
1954 (ecoff_generate_asm_lineno): Likewise.
1955 (ecoff_fix_loc): New function.
1956 * ecoff.h (ecoff_fix_loc): Declare.
1957 * config/tc-mips.c (append_insn): When inserting nops, and using
1958 ECOFF debugging, call ecoff_fix_loc.
1962 * config/tc-h8300.c (build_bytes): If an operand type is
1963 marked as SRC_IN_DST retrieve it from the "destination" op.
1967 * configure.in (arm-*-riscix*): Set emulation to riscix.
1968 * configure: Rebuild.
1969 * config/te-riscix.h: New file to define TE_RISCIX.
1971 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
1975 * Makefile.in (config.status): Just run config.status as other
1980 * mpw-config.in (TARGET_OS): Add definition to conf.
1984 * config/tc-mips.c (append_insn): Parenthesize
1985 cop_interlocks expressions.
1989 * listing.c (listing_print): Close the listing file if it is not
1990 stdout. Close the other files opened for the listing.
1992 * config/tc-sparc.h (md_cons_align): Define.
1993 (sparc_cons_align): Declare.
1994 (HANDLE_ALIGN): Define.
1995 (sparc_handle_align): Declare.
1996 * config/tc-sparc.c (sparc_cons_align): New function.
1997 (sparc_handle_align): New function.
1998 * read.c (cons_worker): Call md_cons_align if it is defined.
2000 * as.h (struct frag): Add fr_file and fr_line fields.
2001 * frags.c (frag_new): Set fr_file and fr_line.
2002 (frag_var): Likewise.
2003 (frag_variant): Likewise.
2005 * as.h (struct frag): Remove unused align_mask and align_offset
2008 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
2013 * configure.in (mips-*-osf*): New target.
2014 * configure: Rebuild.
2016 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
2020 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
2021 between co-processor comparisons and branches for the VR4300.
2025 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
2026 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
2028 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
2029 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
2030 includedir): Use autoconf set values.
2035 * listing.c (listing_eject): Don't do anything if listing is 0.
2036 (listing_list): Likewise.
2037 (listing_source_line): Likewise.
2038 (listing_title): Don't save title if listing is 0.
2039 (listing_source_file): Check listing rather than listing_tail.
2041 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
2042 using shared libraries.
2043 * configure: Rebuild.
2047 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
2048 absolute expression if a relocation type was specified.
2052 * configure.in: Add support for *-*-rtems* configurations.
2053 * configure: Rebuild.
2057 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
2058 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
2059 opcodes library for alpha.
2060 * configure: Rebuild with autoconf 2.10.
2061 * config/tc-alpha.c: Substantial rewrite to add ELF support and
2062 use new opcode table.
2063 * config/tc-alpha.h (md_undefined_symbol): Don't define.
2064 (LOCAL_LABEL): Define differently if OBJ_ELF.
2065 (FAKE_LABEL_NAME): Define if OBJ_ELF.
2066 * config/alpha-opcode.h: Remove.
2067 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
2068 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
2069 include/opcode/alpha.h rather than config/alpha-opcode.h.
2073 * config/obj-aout.c (obj_emit_relocations): Give an error if the
2074 relocation symbol was not resolved.
2075 * config/obj-coff.c (do_relocs_for): Likewise.
2077 * write.c (adjust_reloc_syms): Refetch the symbol section after
2078 calling S_GET_VALUE, since it may have changed.
2080 * expr.c (struct expr_symbol_line): Define.
2081 (expr_symbol_lines): New static variable.
2082 (make_expr_symbol): Add entry to expr_symbol_lines.
2083 (expr_symbol_where): New function.
2084 * expr.h: Use extern on function declarations.
2085 (expr_symbol_where): Declare.
2086 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
2087 rather than printing the meaningless name of an expression
2092 * config/tc-i386.c (md_number_to_chars): Deleted.
2093 * config/tc-i386.h (md_number_to_chars): New macro.
2095 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
2097 (load_symbol_address, load_expression): Use build_mem.
2098 (build_operate): New function.
2099 (emit_addq_r): Use it.
2103 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
2107 * config/tc-vax.h (NOP_OPCODE): Define.
2111 * config/obj-vms.h (S_IS_COMMON): Define.
2112 (S_IS_LOCAL): Check for \002 as well as \001.
2113 (LONGWORD_ALIGNMENT): New macro.
2114 (SUB_SEGMENT_ALIGN): Use it.
2118 * config/vms-conf.h: Reconcile with conf.in.
2122 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
2125 * doc/as.texinfo: Document that any number of hex digits can
2128 * as.c (struct defsym_list): Define.
2129 (defsyms): New static variable.
2130 (parse_args): Just put --defsym arguments on defsyms list, rather
2132 (main): Define defsyms after output file is created.
2134 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
2137 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
2138 insn_error rather than calling check_absolute_expr.
2140 * as.c (emulation_name): Remove unused static variable.
2141 (default_emul_bfd_name): Add return NULL to avoid warning.
2142 * ecoff.c (ecoff_stab): Remove unused variables name and
2144 * frags.c (frag_new): Remove unused variable tmp.
2145 * hash.c (hash_grow): Parenthesize + within <<.
2146 (hash_print_statistics): Use %lu, not %d, to print unsigned
2148 * messages.c: Include "libiberty.h".
2149 (fprint_value): Add cast to avoid printf warning.
2150 (sprint_value): Likewise.
2151 * read.c: Include "ecoff.h".
2152 (emit_expr): Add casts to avoid printf warnings.
2153 * read.h: Use extern for function declarations.
2154 (pop_insert): Declare.
2155 * stabs.c: Include "ecoff.h".
2156 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
2157 former_last_fragP, and new_fragP.
2158 * subsegs.h (subsegs_print_statistics): Declare.
2159 * symbols.c (debug_verify_symchain): Change macro to discard
2161 * write.c (dump_section_relocs): Likewise.
2162 * write.h: Use extern for function declarations.
2163 (write_print_statistics): Declare.
2164 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
2166 * config/e-mipself.c (mipself_bfd_name): Likewise.
2167 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
2169 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
2170 emit relocations against external symbols.
2172 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
2173 message if bfd_reloc_type_lookup fails, rather than calling
2176 * config/tc-alpha.c (alpha_force_relocation): Add
2177 BFD_RELOC_12_PCREL to switch.
2181 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
2182 TE_PE (Lfoo, not .Lfoo).
2186 * read.c (s_fill): Don't warn about a zero repeat count.
2188 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
2189 coprocessor register.
2191 * config/tc-i386.c (md_assemble): When checking the size of a
2192 register to set the size of an instruction, do a bitwise and with
2193 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
2198 * config/tc-h8300.c (parse_reg): Tweak error messages.
2199 (build_bytes): Likewise.
2200 (skip_colonthing): Handle :32 suffix.
2201 (get_specific): Promote L_24 to L_32 if it makes a match.
2202 Don't always promote L_8 to L_16.
2203 (do_a_fix_imm): Clean up L_32 and L_24 handling.
2205 * config/tc-h8300.c (Smode): New variable.
2206 (h8300hmode): Turn off Hmode.
2207 (h8300smode): New function. Turn on Smode and Hmode.
2208 (md_pseudo_table): New ".h8300s" pseudo-op.
2209 (parse_reg): Handle "exr" register.
2210 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
2211 Handle "mach" and "machl" operands for ldmac.
2212 (get_specific): Handle "stm.l" and "ldm.l".
2213 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
2214 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
2219 * doc/as.texinfo: Reorder chapter of machine dependent options so
2220 that it is sorted by chip name.
2222 * doc/as.texinfo: Use consistant spelling of Vax.
2223 * doc/c-vax.texi: Likewise.
2227 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
2229 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
2230 (md_apply_fix): Likewise.
2231 (pa_try): New function.
2232 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
2236 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
2237 (pa_level): New function.
2241 * listing.c (listing_newline): Don't do anything if listing is 0.
2245 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
2246 symbol_create, not symbol_new.
2250 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
2251 absolute symbol unless TC_FORCE_RELOCATION returns true.
2253 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
2255 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
2256 the symbol to a pointer. Don't set previous_file_symbol.
2257 Simplify symbol list rearrangement.
2258 (coff_frob_symbol): Don't do anything with C_FILE symbols.
2259 (coff_adjust_symtab): Don't check previous_file_symbol.
2263 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
2265 (md_pseudo_table): Add ppc_elf_lcomm.
2269 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
2270 d0{24:1},d0' works without an immediate prefix on the bit numbers.
2271 (md_begin): Add digits to alt_notend_table.
2272 (md_parse_option): Make s a const pointer.
2274 * config/tc-sparc.c (md_pseudo_table): Add "empty".
2275 (s_empty): New static function.
2277 * config/obj-coff.c (struct filename_list): Only define if not
2279 (filename_list_head, filename_list_tail): Likewise.
2280 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
2281 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
2282 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
2283 have a section of N_ABS rather than N_DEBUG. If we do a merge,
2284 remove the new symbol from the list.
2285 (obj_coff_endef, both versions): Call tag_insert even if there is
2286 an old symbol with the same name, if the old symbol does not
2288 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
2289 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
2290 symbol that will be moved to the end of the symbol list.
2291 (coff_adjust_section_syms): Always call section_symbol for .text,
2293 (coff_frob_section): Likewise. Also, remove unused variable
2296 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
2297 manipulating frags directly.
2298 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
2301 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
2302 whether the next character is '*' before checking whether we are
2303 at the start of a line. Permit LINE_COMMENT_START to start a
2304 comment in state 1 (seen some whitespace) as well, to match the
2307 * gasp.c (do_align): Permit a fill value for .align.
2311 * read.c (next_char_of_string): Warn if a newline is seen in the
2312 middle of a string. Call bump_line_counters when appropriate.
2316 * symbols.c (colon): Use LOCAL_LABEL.
2320 * Makefile.in (install): Don't check to see if tooldir exists.
2321 Make $(tooldir) and $(tooldir)/bin.
2325 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
2330 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
2331 insn_error. Don't put the string "ERROR" in insn_error. Set
2332 insn_error rather than calling as_warn for an unsupported opcode.
2336 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
2337 before permitting -64.
2338 * output-file.c (output_file_create): Remove duplicate
2343 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
2344 section rather than a .reginfo section.
2345 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
2348 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
2349 bit value with the high bit set.
2353 * read.c (s_lcomm): Set section flags for .sbss section.
2355 * config/tc-mips.c (mips_64): New static variable.
2356 (mips_target_format): If mips_64, return elf64 targets rather than
2358 (md_longopts): Add "32" and "64".
2359 (md_parse_option): Handle -32 and -64.
2360 (md_show_usage): Mention -32 and -64.
2361 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
2366 * config/tc-ppc.c (comment_chars): Make '!' a comment character
2367 for Solaris compatibility.
2369 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
2370 .stabd with 4 arguments into a .stabn.
2374 * config/tc-mips.c (macro): When passing X_add_number to
2375 macro_build, cast it to int first.
2379 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
2380 created by emit_expr.
2382 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
2383 as_perror rather than assert.
2387 * config/tc-mips.c (mips_ip): Mark sections created to hold
2388 floating point information as read only.
2392 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
2396 * read.c (potable): Add .skip as a synonym for .space.
2398 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
2399 4 arguments, providing the 4th argument is 0, to allow
2400 compatibility with the Solaris assembler.
2404 * config/tc-sh.h (struct sh_segment_info_type): Define.
2405 (TC_SEGMENT_INFO_TYPE): Define.
2406 (sh_frob_label): Declare.
2407 (tc_frob_label): Define.
2408 (sh_flush_pending_output): Declare.
2409 (md_flush_pending_output): Define.
2410 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
2411 reloc before the instruction if necessary.
2412 (sh_frob_label): New function.
2413 (sh_flush_pending_output): New function.
2414 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
2415 when looking for the reloc for the target of .uses.
2416 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
2417 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
2419 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
2420 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
2421 the absolute symbol.
2423 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
2424 TC_SEGMENT_INFO_TYPE is defined.
2428 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
2429 matches the register size.
2433 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
2435 (do_relocs_for): Likewise.
2436 (fixup_segment): Don't just quit if linkrelax is set. Try to
2437 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
2442 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
2443 relative relocations with -mrelocatable. Also allow unfixed
2444 relocs in .ex_shared.
2448 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
2449 non-zero before assuming this is a long file name.
2450 (w_strings): Likewise.
2451 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
2453 * config/obj-coff.c (w_strings): Move declaration of i inside
2454 #ifdef block which uses it.
2458 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
2459 "relaxmode". Output relocs which identify various relaxing
2460 possibilities for mov.[bwl] instructions.
2461 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
2465 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
2466 (MD_APPLY_FIX3): Define.
2467 * config/tc-arm.c (my_get_expression): Only watch for bad segments
2469 (md_apply_fix3): Renamed from md_apply_fix.
2470 If pcrel reloc and symbol is in different section, undo effects
2475 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
2476 any reloc with an LR% or RR% field selector for SOM.
2480 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
2485 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
2487 (yank_symbols): Likewise.
2491 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
2492 BFD_ASSEMBLER, add name field.
2493 * config/obj-coff.c: Include "libiberty.h".
2494 (coff_header_append): Handle long section names.
2495 (crawl_symbols): Just use the name field for the symbol name,
2496 without worrying about null byte termination.
2497 (w_strings): Handle long section names.
2498 (write_object_file): Likewise. Also, use the name field, rather
2500 (obj_coff_add_segment): Permit long section names.
2501 (obj_coff_init_stab_section): Use the name field, rather than
2503 (adjust_stab_section): Likewise.
2504 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
2506 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
2507 (mem_fmt): Likewise.
2508 (md_apply_fix): Don't check fx_im_disp field.
2512 * configure.in: Add * after sparc*-*-vxworks.
2513 * configure: Rebuild.
2515 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
2516 define it to comment_chars. Use tc_comment_chars rather than
2518 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
2519 * config/tc-m68k.h (tc_comment_chars): Define.
2520 (m68k_comment_chars): Declare.
2521 * config/tc-m68k.c (m68k_comment_chars): Rename from
2522 comment_chars. Change into a pointer rather than an array.
2523 (md_longopts): Add "bitwise-or".
2524 (md_parse_option): Handle OPTION_BITWISE_OR.
2525 (md_show_usage): Mention --bitwise-or.
2526 * doc/c-m68k.texi: Document --bitwise-or.
2530 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
2533 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
2534 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
2535 correct handling of input line pointer, and call
2536 demand_empty_rest_of_line.
2540 * config/tc-sparc.c (in_bitfield_range): New static function.
2541 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
2542 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
2543 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
2544 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
2545 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
2549 * config/obj-coff.c: BFD_ASSEMBLER:
2550 (coff_last_bf): New static variable.
2551 (coff_frob_symbol): Set endndx of a .bf symbol.
2553 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
2554 (last_bfP): New static variable.
2555 (yank_symbols): Set endndx of a .bf symbol.
2559 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
2560 and -s options. Add -be for good measure.
2564 * read.c (s_space): Support non-constant fill value. Handle fill
2565 value correctly for a size other than 1.
2569 * config/tc-arm.c (my_get_float_expression): Update call to
2570 gen_to_words, X_PRECISION changed from 6 to 5.
2574 * config/tc-ppc.c (register_name,reg_name_search): Move register
2575 name lookup from PE specific code to all targets. Add support for
2576 -mregnames/-mno-regnames to control whether register names are
2578 (md_assemble): Call register_name for all platforms.
2579 (md_parse_option): Add support for -mregnames/-mno-regnames.
2581 * configure.in (powerpcle*-*-solaris): Add support.
2582 (powerpc*-*-linux): Ditto.
2583 * configure: Regenerate.
2585 * config/ppc-sol.mt: New config file for PowerPC Solaris.
2589 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
2590 be associated with a single %lo reloc.
2592 * config/tc-mips.c (load_address): Cast X_add_number to valueT
2593 before comparing against MAX_GPREL_OFFSET, so that negative
2594 numbers are handled correctly.
2599 * config/tc-sparc.c (last_insn): New static variable.
2600 (md_assemble): Warn about putting floating point branches in a
2601 delay slot. If architecture is less than v9, insert NOP
2602 instructions between floating point instructions and floating
2603 point branches. (The SunOS assembler does both these operations.)
2604 Save the last instruction opcode.
2605 (sparc_ip): Add pinsn parameter. Change caller.
2607 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
2608 for byte jump to next instruction to skip empty frags.
2612 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
2613 for addq with a 16 bit signed value, just emit a lda instruction
2618 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
2623 * configure.in: Permit --enable-shared to specify a list of
2625 * configure: Rebuild.
2629 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
2630 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
2634 * config/atof-ieee.c: Fix handling of denormalized extended
2635 precision numbers and overflow/underflow detection.
2636 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
2637 not include the 16 bit gap in the m68k extended precision format.
2641 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
2643 * configure: Rebuild.
2647 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
2648 be accessed using the $gp register.
2649 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
2653 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
2654 bfd_mach_sparc_sparclet if current_architecture is sparclet.
2658 * read.c (get_line_sb): Bump line counters based on
2659 input_line_pointer[-1], not *input_line_pointer. Don't bother to
2660 call LISTING_NEWLINE.
2661 (s_macro): Don't call demand_empty_rest_of_line.
2662 * app.c (do_scrub_chars): When handling C style comments, unget
2667 * read.h (enum linkonce_type): Define.
2668 (s_linkonce): Declare.
2669 * read.c (potable): Add "linkonce".
2670 (s_linkonce): New function.
2671 * subsegs.h (segment_info_type): Add linkonce field to
2672 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
2673 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
2674 (obj_coff_pe_handle_link_once): Declare if TE_PE.
2675 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
2677 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
2678 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
2679 entry based on the linkonce field in segment_info.
2680 * doc/as.texinfo: Document .linkonce.
2684 * doc/as.1: Changed to be recognized by catman -w on Solaris.
2688 * stabs.c (s_stab_generic): Call the listing functions before
2689 doing the rest of the processing, which may involve freeing the
2690 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
2691 SEPARATE_STAB_SECTIONS case.
2693 * config/tc-hppa.c: Remove nested comment.
2694 (tc_gen_reloc): Move label done inside the ifdef in which it is
2696 (md_apply_fix): Pass pointers to correct types to libhppa.h
2697 functions. Always return a value.
2699 * config/tc-mips.h (tc_frob_file): Define.
2700 (mips_frob_file): Declare.
2701 * config/tc-mips.c (struct mips_hi_fixup): Define.
2702 (mips_hi_fixup_list): New static variable.
2703 (imm_unmatched_hi): New static variable.
2704 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
2705 offset_reloc. Pass imm_unmatched_hi to append_insn.
2706 (append_insn): Add unmatched_hi parameter. If it is set, add the
2707 new fixup to mips_hi_fixup_list. Change all callers.
2708 (mips_ip): Set imm_unmatched_hi when appropriate.
2709 (mips_frob_file): New function.
2713 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
2714 * configure: Regenerated.
2718 * as.c (main): Call bfd_set_error_program_name.
2722 * as.h (strdup): Don't declare.
2723 * stabs.c: Include libiberty.h
2724 (get_stab_string_offset): Use xstrdup rather than strdup.
2725 (s_stab_generic): Likewise.
2726 * as.c (parse_args): Likewise.
2727 * read.c (s_mri_sect): Likewise.
2729 * gasp.c (change_base): Recognize \(...) construct documented to
2730 pass through enclosed characters literally through to the output.
2731 (process_assigns): Likewise. Also, be more careful to avoid
2732 looking past the end of the buffer.
2736 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
2741 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
2745 * config/alpha-opcode.h: Added cvtst instruction.
2749 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
2750 string. Extract the addend from the result of expression.
2754 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
2755 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
2756 mode, change the state to 3 rather than 1.
2760 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
2761 correspond to 11 March change.
2765 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
2766 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
2770 * configure: Rebuild with autoconf 2.8.
2774 * config/atof-ieee.c (gen_to_words): Improve handling of
2775 X_PRECISION numbers. Based on patches from Andreas Schwab
2780 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
2782 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
2783 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
2784 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
2785 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
2790 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
2794 * config/tc-hppa.c (SUBSEG_MILLI): Define.
2795 (pa_def_subspaces): Add $MILLICODE$.
2796 (pa_spaces_begin): Set section flags for $MILLICODE$.
2800 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
2801 not SHT_NOBITS. Don't tamper with flags based on type if a
2802 special section was found (revert Feb 29 change).
2804 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
2805 if aligning to a longword boundary or greater.
2809 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
2813 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
2814 * configure: Regenerated.
2818 * config/tc-hppa.c: Add default definitions for R_N0SEL and
2819 R_N1SEL since they're not defined for old versions of hpux.
2821 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
2822 Set "sym_ptr_ptr" and "addend" fields to dummy values for
2823 R_N0SEL and R_N1SEL.
2827 * flonum-konst.c: Add two more constants for 1e+-2048 and
2828 1e+-4096, and correct the other constants.
2830 * symbols.c (resolve_symbol_value): Handle O_logical_not.
2834 * config/obj-elf.c (obj_elf_section): Allow predefined section
2835 types to set the nobits type. Avoid a shadowed declaration.
2839 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
2840 reduce relocs using e_nlrsel field selectors.
2842 * write.c (fix_new_exp): Don't use #elif. Some compilers
2845 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
2847 (pa_chk_field_selector): Handle new field selectors for SOM.
2851 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
2852 * configure: Rebuild.
2853 * config/te-linux.h (LOCAL_LABELS_FB): Define.
2854 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
2856 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
2861 * mpw-make.sed: Update to handle shared library support.
2865 * config/tc-sparc.c (sparc_ip): Print all architectures that support
2866 the insn on mismatch.
2870 * configure.in: Add support for a29-coff.
2871 * configure: Rebuild.
2875 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
2876 of the section when looking for the R_SH_USES fixup, because the
2877 frag addresses have not yet been adjusted.
2879 * gdbinit.in: Set a breakpoint on as_warn_where.
2881 * config/tc-mips.c (macro): Add missing arguments to macro_build
2886 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
2887 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
2891 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
2895 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
2896 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
2897 Recognize [uU] format args as sparclet cpregs.
2901 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
2906 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
2907 well as S_IS_EXTERNAL.
2908 (tc_fix_adjustable): Likewise.
2909 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
2910 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
2912 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
2913 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
2917 * config/tc-sparc.c (max_architecture): Change to sparclite for
2919 (default_compatible): Delete.
2920 (sparc_ffs): New function.
2921 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
2922 (sparc_ip): Rewrite architecture match and bump logic.
2926 * configure.in: Check for 'do not mix' from native linker before
2927 trying to use -rpath.
2928 * configure: Rebuild.
2932 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2933 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2934 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2935 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
2937 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
2938 to the symbol lnnoptr field, to get the correct file offset.
2942 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
2943 symbols that are not function, file, or section symbols to be
2948 * configure: Set and substitute RPATH_ENVVAR.
2949 * configure: Rebuild.
2950 * Makefile.in (RPATH_ENVVAR): New variable.
2951 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
2954 doesn't need opcodes. If configuring shared, opcodes needs bfd.
2955 * configure: Rebuild.
2959 * read.c (s_mri_sect): Don't return '\0' in type. Set all
2960 appropriate flags in BFD section.
2962 * configure.in (m68k-*-psos*): New target.
2963 * configure: Rebuild.
2964 * config/te-psos.h: New file.
2965 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
2971 * configure.in: Remove duplicate setting of cpu_type. Check
2972 whether opcodes library is required for on all targets, not just
2974 * configure: Rebuild.
2976 * config/tc-mips.c (mips_big_got): New static variable.
2977 (s_extern): Don't declare.
2978 (reg_needs_delay): New static function.
2979 (macro_build): Permit GOT/CALL_HI/LO relocs.
2980 (macro_build_lui): If place is not NULL, use the number in the
2982 (load_address): Handle mips_big_got case.
2983 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
2985 (OPTION_XGOT): Define.
2986 (md_longopts): Add "xgot" if OBJ_ELF.
2987 (md_parse_option): Handle -xgot.
2988 (md_show_usage): Mention -xgot.
2989 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
2990 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
2994 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
2995 the byte relocation, point it to the low byte of the word.
2999 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
3000 * configure: Rebuild.
3004 * configure.in: Recognize any sparc* cpu.
3005 * configure: Regenerated.
3009 * read.c (potable): Add "mri" and ".mri".
3010 (s_mri): New function.
3011 * read.h (s_mri): Declare.
3012 * app.c (scrub_m68k_mri): New static variable.
3013 (mri_pseudo): New static variable.
3014 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
3015 flag_m68k_mri. Initialize scrub_m68k_mri.
3016 (mri_state, mri_last_ch): New static variables.
3017 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
3019 (app_push): Save new fields.
3020 (app_pop): Restore new fields.
3021 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
3022 If TC_M68K, use a trivial state machine to look for occurrences of
3023 the .mri pseudo-op, and change the mode appropriately.
3024 * as.h (do_scrub_begin): Update prototype.
3025 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
3027 * config/tc-m68k.c (reg_prefix_optional_seen): New static
3029 (m68k_mri_mode_change): New function.
3030 (md_parse_option): Set reg_prefix_optional_seen.
3031 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
3032 (MRI_MODE_CHANGE): Define.
3033 * doc/as.texinfo: Document .mri pseudo-op.
3035 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
3040 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
3041 * config/m68k-parse.h (enum pic_relocation): Define.
3042 (struct m68k_exp): Add pic_reloc field.
3043 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
3044 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
3045 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
3047 * config/tc-m68k.c: Delete definition of NO_RELOC.
3048 (struct m68k_it): Add pic_reloc field.
3049 (add_fix): Copy over pic_reloc field.
3050 (md_pseudo_table): Interpret .align parameter as byte count.
3051 (mote_pseudo_table): Likewise.
3052 (tc_m68k_fix_adjustable): New function.
3053 (get_reloc_code): New function.
3054 (md_assemble): Use it as last argument to fix_new_exp.
3055 (md_apply_fix_2): For a relocation against a symbol don't put the
3056 addend into the data.
3057 (tc_gen_reloc): Different addend computation for OBJ_ELF.
3058 (m68k_ip): Don't relax an operand that requires pic relocation.
3059 (md_begin): Align .text, .data and .bss on 4 byte boundary by
3061 * write.c (fixup_segment): Don't add symbol value to addend if
3062 TC_M68K and OBJ_ELF.
3063 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
3064 (motorola_operand): Add rule for `(zapc, EXPR)'.
3068 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
3069 *really* is permissible before the comma.
3073 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
3074 inserting nop instructions.
3078 * config/te-aux.h: Change include of aux.h to aux-coff.h.
3082 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
3083 class setting for a CALLNAME symbol in COFF.
3085 * read.c (potable): Pass negative numbers for new .balign[wl] and
3086 .p2align[wl] pseudo-ops.
3087 (s_align_bytes): Treat a negative argument as specifying the fill
3089 (s_align_ptwo): Likewise.
3093 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
3094 (do_align): Take new len parameter. Change all callers. Pass it
3096 (s_align_bytes): Arg now indicates the length of the fill pattern.
3097 (s_align_ptwo): Likewise.
3098 * config/obj-coff.c (write_object_file): Pass length to
3100 * config/tc-i386.h (md_do_align): Take new len parameter.
3101 * config/tc-m88k.h (md_do_align): Likewise.
3102 * config/tc-m88k.c (m88k_do_align): Likewise.
3103 * config/tc-sh.h (md_do_align): Likewise.
3104 * config/tc-sh.c (sh_do_align): Likewise.
3105 * doc/as.texinfo: Document new pseudo-ops.
3107 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
3108 done in cvt_frag_to_fill.
3110 * config/tc-sh.h (sh_do_align): Declare.
3111 (md_do_align): Define.
3112 * config/tc-sh.c (sh_do_align): New function.
3114 * ecoff.c (ecoff_build_lineno): Don't try to store the address
3115 difference if the next address is before the current one.
3117 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
3118 (archs): Initialize new field.
3119 (m68k_ip): Don't list alias names when listing CPUs which support
3122 * as.c (main): Call parse_args before read_begin.
3123 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
3124 front of generated pseudo-ops.
3125 * read.c (potable): Ignore "name".
3126 (s_app_file): Permit a single quote after the string, since one
3127 may appear in m68k MRI mode.
3129 * configure.in: Check for --enable-shared. If linking against
3130 shared BFD and opcodes, fix library name on SunOS, and try to set
3132 * configure: Rebuild.
3136 * as.h (flag_m68k_mri): Declare.
3137 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
3138 * Many files: For MRI syntax that is specific to the m68k MRI
3139 assembler, check flag_m68k_mri rather than flag_mri or
3140 MRI_MODE_NEEDS_PSEUDO_DOT.
3144 * config/tc-i960.c (ARCH_HX): Define.
3146 (targ_has_sfr): Handle ARCH_HX.
3147 (targ_has_iclass): Handle ARCH_HX.
3148 (tc_coff_fix2rtype): Add return 0 to avoid warning.
3149 (tc_headers_hook): If the architecture was specified explicitly,
3150 use it when setting the flags. Set the extern variable coff_flags
3151 rather than headers->filehdr.f_flags, since the latter is set
3152 unconditionally in obj-coff.c.
3153 (i960_handle_align): Remove unused variable fixp.
3155 Support for building bfd and opcodes as shared libraries, based on
3157 * configure.in: Set OPCODES and BFD to search directories.
3158 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
3159 * configure: Rebuild.
3160 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
3161 (LIBDEPS): New variable.
3162 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
3163 $(HLDFLAGS) in link.
3164 (check): Set LD_LIBRARY_PATH in the environment.
3168 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
3169 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
3170 read-only, not read/write.
3174 * Makefile.in (INSTALL_XFORM): Remove -e.
3178 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
3179 than comparing S_GET_SEGMENT to undefined_section.
3180 (write_object_file): Skip symbols which were equated to an
3181 undefined or common symbol.
3182 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
3183 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3185 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
3186 or S_IS_COMMON call S_GET_VALUE.
3187 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3188 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
3189 to an undefined or common symbol, convert the reloc to be against
3191 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3192 undefined or common symbol.
3193 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3194 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
3195 to an undefined or common symbol, convert the reloc to be against
3197 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3198 undefined or common symbol.
3199 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
3200 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3202 (yank_symbols): Skip symbols which were equated to an undefined or
3207 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
3208 * config/obj-bout.h (S_IS_LOCAL): Likewise.
3210 * configure.in: Make sure we only add m68k-parse.o to
3211 ${extra_objects} once, no matter how many m68k targets have been
3213 * configure: Rebuild.
3217 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
3218 * configure: Rebuild.
3222 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
3223 * doc/c-m68k.texi: Document .extend and .ldouble.
3225 * configure.in (m68*-apple-aux*): New target.
3226 * config/te-aux.h: New file.
3227 * config/obj-coff.c (compare_external_relocs): New static function
3229 (do_relocs_for): Sort relocs if TE_AUX.
3230 (fixup_segment): If TE_AUX, store common symbol value in segment.
3231 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
3235 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
3237 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
3239 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
3240 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
3242 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
3243 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
3244 an undefined symbol and the size of a function symbol.
3245 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
3246 for all common symbols.
3250 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
3253 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
3258 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
3261 * ecoff.c (localsym_t): Add addend field.
3262 (add_ecoff_symbol): Add addend argument. Change all callers.
3263 (coff_sym_value): Make static.
3264 (coff_sym_addend): New static variable.
3265 (ecoff_directive_def): Initialize coff_sym_addend.
3266 (ecoff_directive_val): Accept symbol + constant.
3267 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
3268 (ecoff_build_symbols): Include the addend in the symbol value.
3272 * config/tc-ppc.c (md_assemble): Ignore overflow on
3273 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
3277 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
3278 bytes in size, so offset appropriately in big endian mode when
3279 writing the bottom 2 bytes.
3283 * config/tc-sparc.c (default_compatible): New static local.
3284 (md_begin): Initialize it. Rewrite warn_on_bump handling.
3285 (sparc_ip): If no architecture or -bump specified, don't mark as
3286 mismatched those in default_compatible.
3291 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
3292 * configure: Rebuild.
3293 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
3294 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
3295 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
3296 (sco_id): Declare if SCO_ELF.
3300 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
3301 ({max,warn_after}_architecture): New static locals.
3302 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
3303 If both architecture and -bump requested, set max_architecture to max.
3304 (sparc_md_end): Simplify.
3305 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
3306 Rewrite code to bump architecture and check for conflicts.
3307 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
3308 with Solaris assembler.
3309 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
3310 (md_show_usage): Update.
3314 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
3318 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
3319 #ifdef sparcv9 when choosing value.
3322 (sparc_md_end): Declare.
3323 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
3324 (initial_architecture): New static local.
3325 (can_bump_v9_p): Likewise.
3326 (NO_V9): Delete all occurrences.
3327 (sparc_md_end): New function.
3328 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
3329 Don't bump architecture to v9 unless can_bump_v9_p set.
3330 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
3331 * configure.in (sparc64 target cpu): Don't set obj_format here.
3332 (SPARC_V9): Renamed from sparcv9.
3333 (sparc64-*-elf*): Define SPARC_ARCH64.
3334 * configure: Regenerated.
3335 * acconfig.h (SPARC_V9): Renamed from sparcv9.
3336 (SPARC_ARCH64): Add.
3337 * conf.in: Regenerated.
3338 * config/vmsconf.h: Update.
3342 * config/tc-mips.c (load_register): Optimise "dli" loads.
3343 (md_show_usage): add "-mcpu=vr4100" to help text.
3347 * symbols.c (resolve_symbol_value): If a symbol is equated to an
3348 undefined symbol, preserve the X_op of O_symbol.
3349 (S_GET_VALUE): Fix check to permit this case.
3350 * write.c (write_relocs): If a reloc is against an undefined
3351 symbol equated to another symbol, change the reloc to be against
3353 * config/obj-coff.c (do_relocs_for): Likewise.
3355 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
3360 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
3361 computer is PowerPC.
3363 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
3365 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
3366 mentioned in the eabi.
3370 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
3371 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
3375 * config/tc-ppc.c (mapping): Add more relocation suffixes.
3379 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
3380 as a section it is ok to have unadorned -mrelocatable pointers in.
3384 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
3389 * config/obj-elf.c (obj_elf_section): Add hooks so machine
3390 dependent section attributes can be handled.
3392 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
3393 macros to add support for exclude section flag and ordered section
3396 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
3397 functions to add support for exclude section flag and ordered
3402 * subsegs.c (section_symbol): Don't try to look up the section
3403 symbol in the hash table. It should be possible to have a symbol
3404 with the same name as a section, but no connection to it.
3406 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
3409 * expr.c (operand): Skip whitespace after a close parenthesis.
3414 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
3415 as_warn_where for relocation overflow.
3416 (parse_reg): Accept register name only if next character is
3421 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
3422 addresses for static variables.
3426 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
3427 and xgot give the old toc16 relocation.
3428 (md_apply_fix3): Support all GOT relocations.
3432 * config/tc-mips.c (load_address): Correctly handle a constant in
3437 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
3438 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
3439 (build_Mbytes): Handle REG_B.
3443 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
3447 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
3451 * config/obj-coff.c (write_object_file): Set the s_align field to
3452 the number of bytes, rather than to the power of 2.
3456 * Makefile.in (DISTCLEAN_HERE): New variable.
3457 (distclean): Use it.
3458 (maintainer-clean): Depend upon clean-here rather than clean,
3459 distclean, and clean-info. Run make maintainer-clean in doc.
3460 Remove files listed in DISTCLEAN_HERE.
3461 * doc/Makefile.in (maintainer-clean realclean): Split out from
3462 distclean. Depend upon clean-info and distclean.
3466 * mac-as.r: Fix copyright and version strings.
3467 (cfrg): Use PROG_NAME instead of literal name.
3471 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
3473 * config/tc-a29k.h (tc_unrecognized_line): Define.
3474 * config/tc-a29k.c (a29k_unrecognized_line): New function.
3475 (md_operand): Handle a29k style local dollar labels.
3479 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
3483 * read.c (s_fill): If md_flush_pending_output is defined, call
3488 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
3489 Treat rs_align_code like rs_align.
3493 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
3494 converting an absolute address into a PC-relative one.
3498 * mpw-config.in: Don't always use te-generic.h for emulation.
3499 (powerpc-apple-macos): Use emulation te-macos.h.
3501 (mips-gm-magic): New configuration.
3503 * mpw-make.sed (install, install-only): Edit in Mac-specific
3508 * configure.in: Improve message about unsupported ELF targets.
3509 * configure: Rebuild.
3511 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
3512 pattern is zero. From Manfred Hollstein.
3516 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
3517 the various idata sections, we check the name on the .section pseudo.
3521 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
3522 add section's paddr to add_number; compatibility to native as and
3527 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
3529 * hash.c (struct hash_entry): Moved here...
3530 * hash.h (struct hash_entry): ...from here.
3532 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
3533 if it's already known to be null.
3537 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
3538 bit relocs converted to PC relative relocs. This turned out to
3539 add the offset from the beginning of .text twice.
3543 * stabs.c (s_stab_generic): In 's' case, free string from
3546 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
3548 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
3551 * subsegs.h (segment_info_type): Make bitfields unsigned.
3553 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
3554 and move them together. On most systems this will reduce the
3555 structure size by one word.
3556 (operatorT): Define O_max.
3557 * expr.c (expr_begin): Verify that X_op is wide enough to hold
3560 * read.c (pop_insert): Print error returned by hash table
3563 * as.c (dump_statistics): Split out from main; dump some hash
3564 table stats and target-specific stats.
3565 (start_time): No longer automatic to main.
3566 (main): Set file-level start_time and call dump_statistics at
3567 exit. Exit by calling xexit.
3568 (show_usage): Make --statistics description less specific.
3569 * subsegs.c (subsegs_print_statistics): New function.
3570 * write.c (write_print_statistics): New function.
3571 (n_fixups): New static variable.
3572 (fix_new_internal): Increment it.
3573 * read.c (read_print_statistics): New function.
3574 * read.h (read_print_statistics): Declare.
3575 * symbol.c (symbol_print_statistics): New function.
3576 * symbol.h (symbol_print_statistics): Declare.
3577 * hash.c (hash_print_statistics): New function.
3578 * hash.h (hash_print_statistics): Declare.
3579 * config/tc-i386.c (i386_print_statistics): New function.
3580 * config/tc-i386.h (i386_print_statistics): Declare.
3581 (tc_print_statistics): New macro.
3582 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
3585 * hash.c (DELETED): Rewrite to use a valid but unique address.
3586 (START_POWER): Reduce to 10.
3587 (enum stat_enum): New enumerator, replacing STAT_* index macros.
3588 Add new values for counting strcmp calls.
3589 (GROW_FACTOR): New macro.
3590 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
3592 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
3593 (hash_new): Use FULL_VALUE.
3594 (struct hash_control): Definition moved here.
3595 (hash_code): Don't mask to low bits.
3596 (hash_ask): Mask returned hash code here. Check hash value before
3597 calling strcmp; count strcmp calls.
3598 * hash.h (struct hash_control): Declare, don't define, here.
3599 (HASH_STATLENGTH): Deleted.
3600 (struct hash_entry): Add field for hash code.
3601 (hash_say, hash_apply): Don't declare.
3603 * hash.c (destroy): Return void.
3604 (applicatee): Ditto.
3605 (main): Fix declarations.
3606 (hash_apply): Return void. Argument `function' returns void. Put
3607 inside "#ifdef TEST".
3608 (hash_say): Define only if TEST is defined.
3609 * hash.h (hash_apply, hash_say): Declarations deleted.
3613 * configure: Rebuild with autoconf 2.7.
3617 * aclocal.m4 (AC_PROG_CC): Remove local definition.
3618 * configure: Rebuild with autoconf 2.6.
3622 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
3623 (ppc_stabx): Don't increment ppc_debug_name_section_size.
3625 (ppc_frob_file): Remove.
3626 * config/tc-ppc.h (tc_frob_file): Don't define.
3627 (ppc_frob_file): Don't declare.
3631 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
3632 * config/alpha-opcode.h: Include one-operand variants of jmp and
3635 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
3636 with te-delta.h in 8.3 file systems.
3637 * configure.in: Adjusted.
3641 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
3643 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
3645 * expr.c (operand): Handle MRI suffixes after unadorned 0.
3649 Version 2.6 released.
3650 * Makefile.in (VERSION): Updated to 2.6.
3652 * config/obj-coff.c (write_object_file): Change use of md_do_align
3653 to pass a pointer rather than a fill value, to match other uses.
3657 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
3659 * as.c (main): Move md_end call to just after call to
3660 perform_an_assembly_pass. Delete cpu-specific code here.
3661 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
3662 * config/tc-arm.c (md_end): Unused function deleted.
3663 * config/tc-ns32k.c (md_end): Ditto.
3665 * config/tc-i386.c (i386_align_code): New function, moved here
3666 from HANDLE_ALIGN macro.
3667 * config/tc-i386.h (HANDLE_ALIGN): Call it.
3671 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
3676 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
3677 multi-byte nops to avoid changing gcc. The necessary gcc change
3678 might break old assemblers.
3682 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
3683 generate optimal multi-byte nop instructions for ".align n"
3684 ".align n,0x90", and aligns requiring more than 15 bytes of
3685 padding still generate multiple 0x90's as before.
3689 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
3693 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
3694 * configure: Update.
3695 * mpw-make.sed: Reorder commands to make sed happier.
3696 * config/te-macos.h: New file.
3697 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
3701 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
3702 conditional branches.
3703 (md_apply_fix): Likewise.
3707 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
3708 generate a reloc which the linker will not handle correctly. Fix
3709 overflow checking--R_IREL is 18 bits, not 17.
3713 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
3714 from a PC relative reloc if TC_A29K.
3716 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
3717 code depends upon it.
3719 * config/tc-a29k.c (md_operand): Handle $float, $double, and
3720 $extend. Based on code from Eric Freudenthal
3722 * config/tc-a29k.h (LEX_DOLLAR): Define.
3723 * read.c (LEX_DOLLAR): Define if not defined.
3724 (lex_type): Use LEX_DOLLAR.
3728 * configure.in (a29k-nyu-sym1): New target, just like other a29k
3733 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
3737 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
3738 used only by "dlltool.c".
3742 * config/tc-alpha.c: Undefine inline if not __GNUC__.
3743 (md_pseudo_table): Don't define "extern".
3747 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
3749 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
3753 * macro.c (macro_expand_body): Don't warn about == with a
3754 nonexistent parameter, in case it is in a comment field.
3756 * as.c (main): On TC_A29K, call macro_init with macro_alternate
3758 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
3759 set, even if macro_alternate is set.
3760 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
3761 (sub_actual): If macro_strip_at, and kind is '@', don't look up
3762 the token unless it ended in '@'.
3763 * config/tc-a29k.c (line_separator_chars): Remove '@'.
3764 * doc/c-a29k.texi: Document macro usage on A29K.
3768 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
3770 (install_operand): Likewise.
3772 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
3776 * configure.in (m88k-motorola-sysv*): New target.
3777 * configure: Rebuild.
3778 * config/te-delta88.h: New file.
3779 * config/obj-coff.c (write_object_file): Use md_do_align if it is
3781 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
3782 (md_do_align): Define.
3783 * config/tc-m88k.c: Include "subsegs.h".
3784 (m88k_do_align): New function.
3786 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
3787 (COFF_NOLOAD_PROBLEM): Define.
3788 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
3792 * config/tc-i386.c (md_assemble): For a jump instruction with
3793 non-constant target, require 7 available bytes in the current
3798 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
3803 * configure.in (alpha-*-linux*): Use ecoff.
3804 * configure: Rebuild.
3805 * ecoff.c (ecoff_directive_extern): New function.
3806 (ecoff_directive_weakext): New function.
3807 (ecoff_build_symbols): Handle weak symbols.
3808 (ecoff_setup_ext): Likewise.
3809 (ecoff_frob_symbol): Warn about weak common symbols.
3810 * ecoff.h (ecoff_directive_extern): Declare.
3811 (ecoff_directive_weakext): Declare.
3812 * symbols.c (S_IS_WEAK): New function.
3813 * symbols.h (S_IS_WEAK): Declare.
3814 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
3816 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
3821 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
3823 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
3824 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
3825 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
3826 * config/obj-coff.c (coff_frob_symbol): Don't call
3827 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
3828 one. If OBJ_XCOFF, try to figure out whether the symbol is going
3831 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
3832 (ppc_stab_symbol): New static variable.
3833 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
3834 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
3835 sy_tc.real_name to the stab string.
3836 (ppc_bc, ppc_ec): New static functions.
3837 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
3839 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
3841 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
3844 * input-scrub.c (input_scrub_push): Reset sb_index.
3848 * listing.c (listing_newline): Don't create a frag in the absolute
3853 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
3854 (ppc_csect): Move most of the code to ppc_change_csect, and call
3856 (ppc_change_csect): New static function, taken from ppc_csect.
3857 (ppc_section): New static function.
3858 (ppc_saw_abs): New static varable.
3859 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
3860 a symbol has no csect.
3861 (ppc_adjust_symtab): New function.
3862 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
3863 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
3865 * write.c (write_object_file): If tc_adjust_symtab is defined,
3866 call it just before the call to obj_adjust_symtab.
3868 * symbols.c (symbol_find_or_make): Change name to be const.
3869 * symbols.h (symbol_find_or_make): Update declaration.
3873 * doc/as.texinfo (Align): Mention SH.
3874 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
3876 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
3877 (Z8000 Directives, even): Don't give numeric align value, instead
3882 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
3887 * Makefile.in (diststuff): Also make info.
3888 (maintainer-clean realclean): Also make clean-info.
3892 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
3893 (pa_subspace): For ".nsubspa", always create a new subspace
3894 with the given attributes, even if one already exists with the
3899 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
3900 TC_FORCE_RELOCATION, taking an additional section argument. If
3901 the section of the target symbol is not the same as the current
3902 section, always force the relocation to be used.
3903 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
3905 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
3906 md_pcrel_from function, taking an additional section argument.
3907 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
3909 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
3910 older TC_FORCE_RELOCATION if not defined.
3911 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
3912 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
3913 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
3914 TC_FORCE_RELOCATION.
3918 * input-scrub.c (as_where): Set name to null pointer if we don't
3920 * messages.c (identify): Only print filename if non-null.
3921 (as_show_where): Ditto, for line number too.
3922 (as_warn_internal, as_bad_internal): Ditto.
3924 * input-file.c (input_file_open): If the input file can't be
3925 opened, consider it an error.
3929 * config/tc-mips.c: Added mips_4100 control, and support for
3930 accepting the 4100 as a MIPS architecture variant (md_begin,
3931 macro_build, mips_ip, md_parse_option). Adding suitable
3932 command-line OPTIONs, and updating the help text (md_show_usage).
3936 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
3937 BFD_ASSEMBLER is defined.
3939 * Use one active frag and one obstack per frag chain:
3940 * frags.c (frags): Variable deleted.
3941 (frag_alloc): New function.
3942 (frag_grow, frag_more, frag_variant, frag_now_fix,
3943 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
3945 (frag_new): Ditto. Verify that frch_last and frag_now match on
3946 entry and exit, and that old frag_now has non-zero type. Replace
3947 "know" uses with "assert". Use frag_alloc instead of mucking with
3949 * frags.h (frags): Declaration deleted.
3950 * subsegs.h (struct frchain): Add new field frch_frag_now.
3951 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
3953 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
3954 give it any stricter alignment than frchainS structures need. Do
3955 not initialize frags obstack. Set frag_now to point to
3956 dummy_obstack. Initialize absolute_frchain.
3957 (subseg_set_rest): Save and restore frag_now in frch_frag_now
3958 field of frchainS. Don't create new frags on section switch, and
3959 use frag_alloc when creating a new frag chain. For absolute
3960 section, set frchain_now to absolute_frchain. Verify that
3961 frch_last and frag_now match on entry and exit. Initialize
3962 per-chain obstack, and under gcc, set required alignment to that
3963 needed by fragS structure.
3965 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
3967 * stabs.c (get_stab_string_offset): Only copy input string if a
3968 fresh copy is needed, not if the section already exists.
3969 (s_stab_generic): Cache stab section name to bypass lookups, since
3970 usually it will match. Could be made faster still by changing the
3971 memory allocation rules.
3972 (s_xstab): Cache section name to bypass repeated string
3975 * frags.c (frag_new): Deleted register declarations.
3977 * listing.c (frag_now): Don't declare.
3979 * as.c (chunksize): New variable.
3980 (debug_memory): New variable.
3981 (main): If debug_memory is set, reduce chunksize and
3983 * as.h (chunksize): Declare it.
3984 * read.c (read_begin): Use it.
3986 * config/tc-alpha.c (md_shortopts): Include 'g'.
3987 (md_parse_option): Ignore it.
3989 * Makefile.in (distclean): Remove Makefile and config.status from
3990 testsuite directory.
3991 (clean-here): Don't delete testsuite. Instead, delete only the
3992 files within it that would be generated by running tests.
3994 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
3995 frag_now_fix instead of accessing obstack info directly.
3996 * config/tc-arm.c (s_ltorg): Ditto.
3997 (md_assemble): Ditto.
3999 * config/tc-i386.c (md_assemble): Call frag_grow instead of
4004 * stabs.c (aout_process_stab): Insert debug symbol into symbol
4005 chain after parsing value expression, if any, to avoid separating
4006 continued .stabs lines.
4010 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
4011 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
4015 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
4016 -mcom support and PPC_OPCODE_ANY for -many.
4017 (md_parse_option): Ditto.
4019 (md_begin): For duplicate instructions, print all duplicates
4024 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
4025 common mode operation.
4026 (md_show_usage): Add -mcom to usage message.
4030 * expr.c (op_rank): Add O_symbol_rva.
4031 * expr.h (operatorT): Add O_symbol_rva.
4032 * read.c (cons_worker): Set O_symbol_rva when necessary.
4033 * write.c (fix_new_exp): Understand O_symbol_rva.
4037 * config/tc-mips.c: Correct s_cons arguments. From Michael
4042 * config/tc-ppc.c (ppc_macro): Make count unsigned.
4043 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
4044 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
4046 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
4047 make non-static. Change all users.
4048 * config/obj-coff.h (coff_n_line_nos): Declare.
4054 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
4056 * config/obj-elf.c (obj_elf_common): Convert specified byte
4057 alignment to power of two. Set size of local bss symbol.
4059 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
4063 * sb.c, macro.c: Decide whether to include <string.h> or
4064 <strings.h> just as as.h does.
4068 * Makefile.in (site.exp): Fix setting of $srcdir.
4070 * config/tc-arm.c (md_atof): Fix little-endian output.
4071 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
4076 * doc/as.texinfo: Split out the NS32k family documentation,
4077 despite its being commented out for now.
4078 * doc/c-ns32k.texi: New file.
4080 * sb.c, macro.c: Include string.h.
4082 * Makefile.in (comparison): Only check *.o; we don't care if
4083 timestamps inserted by the native linker differ.
4085 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
4086 if alignment stricter than 4 bytes is requested; in that case,
4087 align to a 4-byte boundary first.
4091 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
4092 uninitialized" warning for `Max_Source_Offset'.
4096 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
4097 qualifiers and detect errors if present.
4098 (md_assemble): Add call to parse_toc_entry. Also added some support
4099 for the [tocv] qualifier.
4100 (ppc_pe_tocd): New function to support data in the toc section.
4104 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
4109 * config/tc-m68k.c (isword): Accept all values from -65536 to
4110 +65535, so ~VAL will not be rejected.
4112 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
4113 skip characters after the pseudo-op.
4115 * read.c (get_line_sb): Don't look past buffer_limit.
4116 (s_include): In MRI mode, skip characters after the file name.
4120 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
4121 periods in register names.
4123 For older changes see ChangeLog.1.