3 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
4 pass on the new file hook.
6 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
7 !alpha_force_relocation, as local LITERALs can be adjusted to be
8 relative to the section.
12 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
14 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
15 aligned to odd boundary.
16 (tc_frob_coff_symbol): Define.
18 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
20 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
22 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
25 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
27 (md_assemble): If the size is 'B', set fx_signed.
28 (md_apply_fix_2): Use fx_signed when checking for overflow.
30 * write.h (struct fix): Add fx_signed field.
31 * write.c (fix_new_internal): Initialize fx_no_overflow and
33 (fixup_segment): Use fx_signed when checking for overflow.
34 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
35 fx_signed when checking for overflow.
39 * NOTES, NOTES.config: Removed. These are rarely, if ever,
40 updated, and all the useful information is in doc/internals.texi.
43 * read.c (read_a_source_file): Check for conditional operators
44 before doing an MRI pending alignment.
45 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
46 (tc_conditional_pseudop): Define.
47 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
48 * doc/internals.texi (CPU backend): Describe
49 tc_conditional_pseudoop.
52 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
54 (md_begin): Check m68k_rel32_from_cmdline before setting
56 (m68k_mri_mode_change): Likewise.
57 (md_longopts): Add --disp-size-default-16 and
58 --disp-size-default-32.
59 (md_parse_option): Handle new options.
60 (md_show_usage): Mention new options.
61 * doc/c-m68k.texi (M68K-Opts): Document new options.
64 * config/tc-m68k.c (m68k_index_width_default): New static
66 (m68k_ip): Use m68k_index_width_default to set the size of a base
67 register whose size was not given.
68 (md_longopts): Add --base-size-default-16 and
69 --base-size-default-32.
70 (md_parse_option): Handle new options.
71 (md_show_usage): Mention new options.
72 * doc/c-m68k.texi (M68K-Opts): Document new options.
74 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
76 * configure.in: Recognize mips-*-linux* target.
79 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
80 work if valueT is only 32 bits.
82 * config/tc-mips.c: Throughout, check target_big_endian rather
85 (mips_init_after_args): Remove.
86 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
87 (mips_init_after_args): Don't declare.
88 (tc_init_after_args): Don't define.
90 * config/tc-mips.h (tc_frob_after_relocs): Define if
92 (mips_elf_final_processing): Likewise.
93 (ELF_TC_SPECIAL_SECTIONS): Likewise.
98 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
99 sld.{h,w} and the operand is relocatable, adjust the adend by
100 shifting it right one bit.
105 * read.c (read_a_source_file): Check mri_pending_align after
106 checking for a macro. From Ronald F. Guilmette
109 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
111 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
112 add segment argument. If OBJ_ELF, treat a relocation against a
113 symbol in a linkonce section like a relocation against an external
115 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
119 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
123 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
124 includes config.h instead of host.h, tc.h instead of tp.h, and
125 targ-env.h instead of target-environment.h.
126 Also, obj-format.h includes targ-cpu.h instead of
129 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
130 * configure: Rebuild with autoconf.
131 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
132 (TARGET_FORMAT): Define to "coff-tic80".
133 * config/tc-tic80.c: New file for TIc80 support.
134 * config/tc-tic80.h: New file for TIc80 support.
139 * doc/as.texinfo (M): Mention explicitly that -M changes macro
144 * write.c (adjust_reloc_syms): If the fixup symbol has been
145 equated to an undefined symbol, convert the fixup to being against
146 the target symbol. Remove obsolete code handling a special case
151 * mpw-make.sed: Use NewFolderRecursive for installation.
156 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
160 * config/tc-d10v.c (md_assemble): Fix bug which caused
161 second instruction in a line to be case sensitize. PR11312
166 * config/tc-mn10200.c (mn10200_insert_operand): Don't
167 range check operands with MN10200_OPERAND_NOCHECK set.
168 (check_operand): Likewise.
172 * config/tc-mips.c: Undo part of last Friday's alignment changes.
173 (md_begin): Always align the text section to a four byte
175 (append_insn): Remove call to record_align.
177 * config/tc-mips.c (insn_label): Remove.
178 (struct insn_label_list): Define.
179 (insn_labels, free_insn_labels): New static variables.
180 (mips_clear_insn_labels): New static function.
181 (append_insn): Mark all mips16 text labels, and make them odd.
182 Handle all labels after emitting a nop, not just one. Call
183 mips_clear_insn_labels rather than just clearing insn_label.
184 (mips_emit_delays): Add insns parameter, and use it to decide
185 whether to mark mips16 labels. Handle all labels, not just one.
186 Force mips16 labels to be odd. Change all callers.
187 (mips16_immed): Don't check for an odd branch target.
188 (md_apply_fix): Don't check mips16 mode for a branch reloc.
189 (mips16_extended_frag): Ignore the low bit in a branch target.
190 (md_convert_frag): Likewise.
191 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
193 (mips_align, mips_flush_pending_output, s_cons): Likewise.
194 (s_float_cons, s_gpword): Likewise.
195 (s_align): Use insn_labels rather than insn_label.
196 (s_cons, s_float_cons, s_gpword): Likewise.
197 (mips_frob_file_after_relocs): New function.
198 (mips_define_label): Rewrite to add to insn_labels list.
199 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
200 * ecoff.c (ecoff_build_symbols): If the size of a function comes
201 out odd, increment it.
203 * config/tc-mips.c (append_insn): Only update prev_insn when not
204 reordering if place is NULL.
206 * config/tc-mips.c (mips16_ip): Check for a missing expression
207 when using the register indirect addressing mode.
211 * config/tc-mn10200.c (mn10200_insert_operand): Don't
212 check 24bit operands for overflow.
213 (check_operand): Likewise.
217 * doc/as.texinfo (Section): Document how to use the .section
218 pseudo-op for COFF and ELF.
222 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
226 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
227 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
228 store them. Adjust other RELAX_MIPS16 macros.
229 (RELAX_MIPS16_DSLOT): Define.
230 (RELAX_MIPS16_JAL_DSLOT): Define.
231 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
232 handling of whether previous instruction has a fixup. Set
233 prev_insn_reloc_type.
234 (mips_no_prev_insn): Clear prev_insn_reloc_type.
235 (mips16_extended_frag): Use the right base address for a PC
236 relative add or load.
237 (md_convert_frag): Likewise. If a PC relative add or load is
238 used, record the alignment for the section.
242 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
243 linkonce section into a reloc against the section symbol.
245 * config/tc-mips.c (mips16_macro): Remove nop instructions after
248 * config/tc-mips.c (md_begin): If configured for an embedded ELF
249 system, don't set the section alignment to 2**4.
250 (s_change_sec): Likewise.
251 (append_insn): Call record_alignment for the section.
252 (md_section_align): Don't align the section size for an embedded
257 * write.c (adjust_reloc_syms): Make sure that symbols are
258 resolved; expression symbols may have been skipped.
259 * config/obj-coff.c (fixup_segment): Likewise.
263 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
264 BFD_RELOC_24_PLT_PCREL relocation.
265 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
270 * config/tc-d10v.c (write_2_short): Remove code that called
271 parallel_ok() when the programmer specified parallel instructions.
276 * config/tc-mn10300.c (md_assemble): Update to handle endianness
279 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
280 * config/tc-mn10300.c (md_assemble): Likewise.
284 * config/tc-mips.c (append_insn): Make sure there is enough room
285 in a frag after a mips16 instruction to switch it with a jump
288 * config/tc-mips.c (mips16_extended_frag): Give an error for an
289 attempt to use a non absolute symbol in an extending frag.
293 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
294 * config/tc-mn10200.h: Likewise.
298 * app.c (do_scrub_chars): At the end of a C comment, pass space to
299 UNGET rather than PUT. Set old_state before setting state to -2.
301 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
302 when extending because the value is exactly maxtiny + 1.
304 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
305 arguments, and store them. Adjust other RELAX_MIPS16 macros.
306 (RELAX_MIPS16_USER_SMALL): Define.
307 (RELAX_MIPS16_USER_EXT): Define.
308 (mips16_small, mips16_ext): New static variables.
309 (append_insn): Pass mips16_small and mips16_ext to
311 (mips16_ip): Set mips16_small and mips16_ext.
312 (mips16_immed): Don't check mips16_autoextend.
313 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
315 * write.c (write_relocs): Print an error for an out of range
316 fixup, rather than calling abort.
318 * as.c (main): Unlink the output file if there are errors while
319 generating the fixups.
323 * config/tc-mips.c (mips16_extended_frag): Don't call
325 (md_convert_frag): Call resolve_symbol_value before calling
326 S_GET_VALUE, and don't add in the frag address.
328 * config/tc-mips.c (mips16_immed): Add file and line parameters,
329 and use them when reporting errors. Change all callers.
333 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
334 Remove '$' prefixing for registers.
339 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
340 is initialized before using it.
341 (d10v_cleanup): No longer uses its argument, so make it void.
343 * config/tc-d10v.h (d10v_cleanup): Change prototype.
348 * write.c (fixup_segment): Don't discard the symbol for a PC
349 relative fixup to an absolute symbol.
354 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
355 with multiple sections.
360 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
361 no-mips-16 to no-mips16.
362 (s_mipsset): Accept .set mips16 and .set nomips16.
366 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
367 argument, and check for +/- constant following the suffix, folding
368 it into the expression.
369 (ppc_elf_cons): Change ppc_elf_suffix calls.
370 (md_assemble): Ditto.
371 (shlib): Replace boolean mrelocatable with enumeration shlib.
372 (md_parse_option): Discriminate between PIC style shared libraries
374 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
379 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
380 change the comment characters.
382 * config/tc-ppc.c (comment_chars): Delete in favor of
384 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
386 (ppc_comment_chars): Select appropriate comment chars by default.
387 (msolaris): New flag for -m{,no-}solaris.
388 (md_parse_option): Recognize -K pic. Add support for
390 (md_show_usage): Update.
391 (md_begin): Do not set ELF flags if Solaris.
392 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
393 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
397 * as.c (main): Correct handling of flag_always_generate_output.
401 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
402 fx_offset, not fx_addnumber.
404 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
409 * config/tc-i386.c: Correct misspelling: balenced to balanced.
413 * config/tc-mips.c (md_section_align): Check for an alignment of
414 4, not an alignment of 16. Corrects August 7 patch.
418 * configure, conf.in: Rebuild with autoconf 2.12.
420 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
421 alignment is specified.
423 Add support for mips16 (16 bit MIPS implementation):
424 * config/tc-mips.c: Extensive additions for mips16 support, not
426 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
427 (insn_uses_reg): Change last parameter to an enum.
428 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
429 (md_relax_frag): Define as mips_relax_frag.
430 (mips_relax_frag): Declare.
431 (struct mips_cl_insn): Add use_extend and extend fields.
432 (tc_fix_adjustable): Define.
433 * config/obj-elf.h (S_GET_OTHER): Define.
434 (S_SET_OTHER): Define.
438 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
439 o and p operand specifiers.
443 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
444 (relax_segment, case rs_org): Move code inside braces. Move locals
445 target,after inside too.
446 (relax_segment, case rs_machine_dependent): Guts moved to ...
447 (relax_frag): New function.
448 Call md_prepare_relax_scan if defined.
449 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
454 * config/tc-mn10300.c (address_registers): Use '$' as register
455 prefix instead of '%'.
456 (data_registers, other_registers, md_assemble): Likewise.
458 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
459 (data_registers, other_registers, md_assemble): Likewise.
461 * config/tc-mn10300.c (md_assemble): Correctly determine the
462 correct location and type for each relocation.
463 (md_pcrel_from): Simplify.
467 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
468 converted into branch around branch.
472 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
473 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
475 * config/tc-mn10300.c (md_begin): Set linkrelax.
476 (md_assemble): Create fixups as needed.
477 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
482 * config/tc-d10v.c (parallel_ok): When automatically converting
483 serial ops to parallel, do not consider a branch as the first
489 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
493 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
495 (check_operand): Likewise.
500 * config/tc-d10v.c (parallel_ok): Branch and link instructions
502 (write_2_short): Call parallel_ok to check whether two short
503 instructions the user requested execute in parallel, can be
508 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
509 allow a branch and link in parallel with an exe instruction.
513 * doc/c-d10v.texi: Add info on @word modifier.
518 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
519 operands are assumed to be 32bits. Use "bits" field to hold the
520 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
521 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
524 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
525 of a MN10300_OPERAND_SPLIT operand by operand->shift.
527 * config/tc-mn10300.c (mn10300_insert_operand): Handle
528 MN10300_OPERAND_SPLIT.
532 * config/tc-mn10300.c (md_assemble): Insert operands into
533 the extension part of the instruction if necessary.
534 (mn10300_insert_operand): Accept pointer to extension word
535 argument. Make insn a pointer argument too. Return type
536 is now void. All callers changed.
540 * config/tc-mn10300.c (mn10300_insert_operand): Handle
541 repeated register operands.
545 * doc/as.texinfo: Added section on reporting bugs.
547 * config/tc-alpha.c: Change uses of void * to PTR. Change the
548 alpha_macro emit field to expect a const argument, and change the
549 arg field to be const. Fix some spacing to follow the GNU
554 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
555 (pca56) and 21264 (ev6) cpus.
556 (md_apply_fix): Private relocation types are now negative.
557 (alpha_force_relocation): Likewise.
558 (tc_gen_reloc): Likewise.
559 (emit_insn): Likewise.
560 (emit_ldXu): Do the right thing when the hardware can do byte insns.
561 (emit_stX): Likewise.
562 (emit_sextX): Likewise.
566 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
567 a symbol found in a reloc.
569 * symbols.c (resolve_symbol_value): Improve the error message if
570 an undefined symbol is used in an expression.
574 * doc/internals.texi: Rewrite, and add a lot of documentation.
575 * doc/Makefile.in (internals.info): New target.
580 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
581 against weak symbols.
585 * config/tc-v850.c (md_assemble): Don't lose for relaxable
588 * config/tc-v850.c (md_convert_frag): Make sure we insert the
589 fixup at the right address within the frag.
591 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
592 to an absolute value, instead increment it as needed.
594 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
595 * config/tc-v850.c: Fix some indention problems.
596 (md_relax_table): Define for D9->D99 branch displacement
598 (md_convert_frag): Do something useful instead of aborting.
599 (md_estimate_size_before_relax): Likewise.
600 (md_assemble): Note if the matching instruction has a relaxable
601 operand. If it does, allocate frag with frag_var and don't
608 * config/tc-d10v.h (md_cleanup): New function. This is needed to
609 write out any buffered instructions when a ".end" is found.
614 * read.c (read_a_source_file): New hook md_cleanup().
618 * write.c (fix_new_exp): Use make_expr_symbol to build an
619 expression symbol for a complex fixup.
624 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
625 hacks to improve parsing of complex hi, lo, zda, etc
627 (md_assemble): Don't demand and eat a trailing ')' after finding
628 a v850 relocation prefix. Sign extend the constant in a
629 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
631 (parse_cons_expression_v850): Don't eat a trailing ')' after
632 finding a v850 relocation prefix.
634 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
635 (TC_CONS_FIX_NEW): Likewise.
636 * config/tc-v850.c (parse_cons_expression_v850): New function.
637 (cons_fix_new_v850): Likewise.
639 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
644 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
645 for an unknown relocation type.
650 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
655 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
656 differently for movea & sst/sld insns.
662 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
668 * obj.h (struct format_ops): Add frob_file_after_relocs field.
669 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
670 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
671 frob_file_after_relocs field.
672 * config/obj-elf.c (elf_format_ops): Likewise.
673 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
678 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
679 specially if not ELF.
680 (md_apply_fix): Handle BFD_RELOC_64.
681 (tc_gen_reloc): Handle BFD_RELOC_64.
683 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
684 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
687 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
688 Don't check S_GET_OTHER.
692 * config/tc-mips.c (mips_ip): Accept an odd floating point
693 register with l.s or s.s.
695 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
697 (obj_aout_type): New static function.
701 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
706 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
707 and sdaoff expressions.
709 * write.c (fixup_segment): Don't add symbol value to addend if
711 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
714 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
715 (md_pcrel_from_section): Likewise.
716 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
720 * config/tc-v850.c (md_pcrel_from): Delete unused function.
721 (md_pcrel_from_section): New function.
722 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
727 * config/tc-mips.c (load_register): Add cast to offsetT when using
732 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
733 (elf_frob_file_after_relocs): ... here. New function.
734 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
735 * write.c (write_object_file): Call *frob_after_relocs after the
736 call to write_relocs.
738 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
740 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
741 symbol, as this messes up linking. Instead, expand the recursive call
742 inline and change up the appropriate bits to get the 0x8000 offset
747 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
748 two symbols in the same segment to be adjusted.
750 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
751 * configure: Rebuild.
755 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
757 (ppc_elf_validate_fix): Ditto.
758 (md_assemble): Ditto.
760 (ppc_pe_section): Ditto.
761 (ppc_frob_symbol): Ditto.
765 * config/mn10300.c (md_assemble): Pass an extra shift count
766 to mn10300_insert_operand based on the opcode format.
767 (mn10300_insert_operand): Accept and use extra shift count
770 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
771 formats rather than hard-coded constants.
773 * config/tc-mn10300.c (md_assemble): Format D5 instructions
774 are 7 bytes long. Write out instructions in big-endian format.
778 * config/tc-mn10300.c (md_assemble): Tweak further so
779 that all instructions are parsed correctly.
783 * as.h: Include libiberty.h.
784 (xmalloc, xrealloc): Don't declare.
785 * as.c: Don't include libiberty.h.
786 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
787 * config/tc-mips.c: Likewise.
788 * messages.c: Likewise.
789 (xstrerror): Don't declare.
794 * config/tc-mn10300.h (pre_defined_registers) Remove.
795 (system_registers, cc_names): Likewise.
796 (address_registers, data_registers, other_registers): New register
798 (register_name, system_register_name, cc_name): Remove.
799 (mn10300_reloc_prefix): Likewise.
800 (data_register_name): New function.
801 (address_register_name, other_register_name): Likewise.
802 (md_assemble): Rough cut at parsing operands. Remove lots of
804 (md_apply_fix3): Disable for now.
808 * config/tc-m68k.c (select_control_regs): New function, extracted
809 out of m68k_init_after_args.
810 (m68k_init_after_args): Use it.
811 (mri_chip): Use it here as well to update set of allowed control
816 * config/obj-elf.c (elf_begin): New function.
817 (obj_elf_section): Add the section symbol to the symbol table.
818 * config/obj-elf.h (obj_begin): Define.
819 (elf_begin): Declare.
820 * as.c (perform_an_assembly_pass): Call obj_begin if it is
825 * config/obj-coff.c (fixup_segment): Subtract the section address
826 from a PC relative reloc if TC_M68K.
830 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
831 .uaxword available even if not OBJ_ELF.
832 (md_atof): Remove unused local variable wordP.
836 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
837 for Matsushita MN10x00 support.
838 * configure.in: Recognize mn10x00-*-*
839 * configure: Rebuilt.
843 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
844 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
845 (load_expression): track clobbering of base reg before jmp/jsr.
846 (s_alpha_file): pass case_hack flags and source filename via
848 * tc-alpha.h (TC_CONS_FIX_NEW): define
852 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
853 * configure: Rebuild.
857 * read.c (s_macro): Warn if a macro has the same name as a
859 (s_space): In m68k MRI mode, align to a word boundary.
860 * macro.c (define_macro): Add namep parameter. Change all
862 * macro.h (define_macro): Update declaration.
864 * as.c (show_usage): Print bug report address.
865 (parse_args): Change version printing to match current GNU
867 * gasp.c (show_usage): Print bug report address.
868 (main): Change version printing to match current GNU standards.
870 * config/tc-m68k.c (init_table): Correct access control unit
873 * config/tc-alpha.c: Add some static function prototypes.
874 (alpha_macros): Move to top of file. Make static.
875 (alpha_num_macros): Move to top of file.
880 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
886 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
887 and just call sprintf_vma.
891 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
895 * read.c (float_cons): Call md_flush_pending_output if it is
901 * config/tc-d10v.c (md_operand): Created. Allows operands to
903 * config/tc-d10v.h (md_operand): Undefined.
908 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
909 (md_assemble): A fixup width of '3' means a 1 byte reloc.
913 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
914 reloc for the i960 for a reloc in the same section. This undoes
915 one of the two changes made Aug 19.
919 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
920 symbols to the position of the debugging information.
924 * expr.c (expr): Always use unsigned right shifts for >>.
928 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
929 slots when processing BL fixups.
931 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
932 on the first half of the instruction.
936 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
937 stabs expression, rather than giving an error.
939 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
944 * config/tc-mips.c (append_insn): Fill in the value for a constant
945 jump, rather than creating a reloc.
949 * config/tc-mips.c (append_insn): Don't swap an instruction which
950 sets a condition code with an instruction which uses a condition
952 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
955 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
956 (mips_ip): Give a better error message if the ISA level is wrong.
957 (md_parse_option): Recognize -mcpu=[v][r]5000.
961 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
963 (load_register): Ensure hi32 bits are not lost during lo32bit
964 processing. Fix shift offset that was overflowing into the next
965 instruction field. Add code to generate shorter sequences for
966 constants with a single contiguous seqeuence of ones.
971 * config/tc-d10v.c (d10v_dot_word): New function to support
972 "@word" with the word pseudo-op.
973 (md_apply_fix3): Cleanup and changes to support correct sizes
974 for 16 and 18-bit relocs.
979 * configure.in (sparc-*-aout): Set `em'.
980 * configure: Regenerated.
981 * config/te-sparcaout.h: New file.
982 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
983 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
984 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
985 (SPECIAL_CASE_{SETSW,SETX}): Define.
986 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
987 (md_begin): Delete setting of `target_big_endian'.
988 (output_insn): New function.
989 (md_assemble): Rewrite. Add `setx' support.
990 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
992 (md_atof): Add little endian support.
993 (md_number_to_chars): Likewise.
994 (md_apply_fix): Likewise.
995 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
996 (md_parse_option): Likewise.
997 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
1001 * ecoff.c (ecoff_new_file): New function.
1002 * ecoff.h (ecoff_new_file): Declare.
1003 * config/obj-ecoff.h (obj_app_file): Define.
1007 * config/tc-alpha.c (load_expression): Bias the .lit8 section
1008 symbol by 32k so that our 16-bit signed offset can address the
1013 * config/tc-mips.c (load_register): Remove unused variable tmp.
1017 * config/tc-mips.c (load_register): Remove unnecessary code that
1018 was causing the high 32bits of 64bit constants to be lost.
1023 * config/tc-d10v.c: Added changes to support function
1024 pointers and "@word" syntax.
1030 * config/tc-v850.c: Remove commented out and #if 0'd code.
1031 (v850_reloc_prefix): Provide prototype.
1032 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
1034 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
1035 (md_assemble): Add default to case statement.
1039 * config/tc-v850.c (md_assemble): Compute size of the instrction
1042 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
1047 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
1048 routines to fetch/store the updated instruction from/to memory.
1049 (v850_insert_operand): If the operand has a specialized insert
1054 * config/tc-v850.c (reg_name_search): Align calling convention to
1055 be like identical function found in tc-ppc.c.
1056 (get_reloc): Removed.
1057 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
1058 (md_assemble): emit fixups.
1059 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
1061 (md_apply_fix3): handle fixups/relocs.
1062 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
1068 * configure.in (sh-*-elf*): New target.
1069 * config/tc-sh.h (TARGET_ARCH): Define.
1070 (WORKING_DOT_WORD): Define.
1071 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
1072 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
1073 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
1074 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
1075 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
1076 (SUB_SEGMENT_ALIGN): Likewise.
1077 (RELOC_32): Don't define.
1078 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
1079 (target_big_endian): Declare if OBJ_ELF.
1080 (TARGET_FORMAT): Define if OBJ_ELF.
1081 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
1083 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
1084 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
1085 (struct sh_count_relocs): Define.
1086 (sh_count_relocs): New static function, broken out of
1087 sh_frob_file. Add BFD_ASSEMBLER code.
1088 (sh_frob_section): Likewise.
1089 (sh_frob_file): Call sh_frob_section.
1090 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
1091 call section_symbol rather than seg_info (seg)->dot.
1092 (md_section_align): Add OBJ_ELF version.
1093 (SWITCH_TABLE_CONS): Define.
1094 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
1095 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
1096 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
1098 (struct reloc_map): Define if not BFD_ASSEMBLER.
1099 (coff_reloc_map): Likewise.
1100 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
1101 (tc_gen_reloc): New function if BFD_ASSEMBLER.
1102 * write.c (write_relocs): Ifdef out fx_where test which triggers
1103 inappropriately for SH ELF.
1104 (write_object_file): Call tc_frob_file_before_adjust and
1105 obj_frob_file_before_adjust if they are defined.
1107 * write.c (write_object_file): Use BFD_RELOC_16, not
1108 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
1110 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
1116 * config/tc-d10v.c (find_opcode): Fix problem with calculating
1117 branch sizes in across sections.
1123 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
1128 * config/tc-v850.c (md_begin): Deal with end of opcode
1135 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
1136 the wrong opcode for cases like st2w where there are many forms
1137 of the same instruction.
1142 * expr.c (operand): If md_parse_name is defined, call it before
1143 calling symbol_find_or_make.
1144 * config/tc-ppc.h (md_parse_name): Define.
1145 (ppc_parse_name): Declare.
1146 * config/tc-ppc.c (reg_name_search): Add regs and regcount
1148 (register_name): Update call to reg_name_search.
1149 (cr_operand): New static variable.
1150 (cr_names): New static const array.
1151 (ppc_parse_name): New function.
1152 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
1153 cr_operand before calling expression.
1157 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
1158 hppa_gen_reloc_type call.
1163 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
1164 on addresses. Improved error messages.
1165 * doc/c-d10v.texi: Added docs for register pairs.
1169 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
1175 * ecoff.c (init_file): Initialize fMerge to 1.
1176 (add_file): Restore old file merging code, but only merge files if
1178 (ecoff_directive_loc): Clear fMerge field of current file.
1179 (ecoff_generate_asm_lineno): Likewise.
1184 * doc/c-d10v.texi: Fix typo.
1190 * config/tc-v850.c (md_assemble): Correct bit masking for
1191 hi and lo expressions.
1193 * config/tc-v850.c (md_assemble): Rough cut at demanding
1194 "ep" or "r30" in sst and sld instructions.
1195 (md_apply_fix3): Don't abort. Just warn that we don't
1198 * config/tc-v850.c (CC_NAME_CNT): Define.
1199 (cc_name): New function.
1200 (md_assemble): Handle V850_OPERAND_CC correctly.
1202 * config/tc-v850.c (md_assemble): Don't forget to initialize
1205 * config/tc-v850.c (reg_name_search): Generalize to search
1206 any given register table.
1207 (register_name): Pass appropriate table and size to reg_name_search.
1208 (system_register_name): New function.
1209 (SYSREG_NAME_CNT): Define.
1210 (md_assemble): Handle operands which are system registers.
1212 * config/tc-v850.c (md_assemble): If we find a register, but the
1213 opcode doesn't want a register, then we don't have a match.
1214 (md_assemble): Get size of the instruction from the opcode table.
1219 * configure.in: Set and substitute HLDENV.
1220 * configure: Rebuild.
1221 * Makefile.in (HLDENV): New variable.
1222 (as.new): Use $(HLDENV).
1224 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
1225 if an array dimension is not known.
1230 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
1231 * doc/c-d10v.texi: Cleanup.
1237 * config/tc-v850.c: New file.
1238 * config/tc-v850.h: New file.
1239 * configure (v850-*-elf): New target.
1240 * configure.in (v850-*-elf): New target.
1246 * doc/c-d10v.texi: New file.
1247 * doc/all.texi: Added D10V stuff.
1248 * doc/as.texinfo: Added D10V stuff.
1252 * config/tc-d10v.c: All references to defined symbols should
1253 now use the optimal instruction. .float and .double now work.
1258 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
1259 section address for the i960 as is done for the i386.
1263 * mpw-config.in: Add wildcards for config matching, add mips-*-*
1264 case, forward-include bfd/elf-bfd.h.
1269 * config/tc-d10v.c: Add additional information to the opcode
1270 table to help determinine which instructions can be done
1276 * config/tc-arm.c: Major changes to add Thumb support, with lots
1278 Reverted to INSN_SIZE macro, rather than insn_size variable.
1279 (insns): Added ARM "bx" instruction support.
1280 (tinsns): Added Thumb instruction definition structure.
1281 (arm_tops_hsh): Added hash structure for Thumb opcodes.
1282 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
1283 (opcode_select,s_arm,s_thumb,s_code): Added.
1284 (decode_shift): Allow upper-case RRX.
1285 (do_ldst): Simpler halfword support.
1286 (do_ldmstm): Improved.
1287 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
1288 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
1289 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
1290 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
1291 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
1293 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
1294 (md_parse_option): Add support for -mthumb.
1295 (md_show_usage): Updated to reflect new command line option.
1296 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
1297 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
1298 fragment information to be held.
1302 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
1303 non-null, but is a constant.
1304 (fix_new_arm): Call make_expr_symbol to make the expression symbol
1305 so that error reporting will work correctly.
1309 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
1310 against weak symbols.
1314 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
1315 (ppc_force_relocation): Declare if OBJ_XCOFF.
1316 * config/tc-ppc.c (ppc_force_relocation): New function if
1321 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
1326 * config/tc-d10v.c: Fix problem with relocs.
1331 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
1332 with nops if not in data_section or bss_section.
1336 Add support for openVMS/Alpha.
1337 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
1338 * config/obj-evax.c: New file.
1339 * config/obj-evax.h: New file.
1340 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
1342 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
1343 defined. Add case for bfd_target_evax_flavour.
1344 * config/vms-a-conf.h: New file.
1345 * conf-a-gas.com: New file.
1346 * configure.in: Add target alpha-*-*vms*.
1347 * configure: Rebuild.
1348 * makefile.vms: New file.
1349 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
1350 Don't call ffs on openVMS/Alpha.
1354 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
1355 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
1357 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
1358 (NEED_DECLARATION_SBRK): New macro.
1359 * configure, conf.in: Rebuild.
1360 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
1361 (strstr): Declare if NEED_DECLARATION_STRSTR.
1362 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
1366 * symbols.c (resolve_symbol_value): Handle addition or subtraction
1367 by a constant before entering the main switch. Reject attempts to
1368 apply an arithmetic function to non-absolute symbols, except for
1369 the special case of subtraction of two symbols in the same
1372 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
1373 not to more than a 16 byte boundary.
1375 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
1377 (tc_gen_reloc): If out of memory call as_fatal rather than
1378 assert. If no howto found, call as_bad_where rather than
1379 as_fatal. Change the error message slightly. Set howto to a
1380 non-NULL value in order to keep going.
1385 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
1390 * config/tc-h8300.c (get_specific): New operand "size" derived
1391 from ".b", ".w" and ".l" extensions. All callers changed. If
1392 the base instruction has no operands, then use the size to
1393 determine which specific instruction to use.
1397 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
1401 * config/tc-mips.c (md_section_align): Don't change addr if
1406 * config/tc-hppa.c: Revert yesterday's changes.
1411 * config/tc-d10v.c: Disable range checking on 16-bit values.
1416 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
1417 not just the ones that don't match.
1422 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
1427 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
1428 pre-cursor to adding Thumb support. Also added cpu_variant flag
1429 information to each of the asm_flg structures.
1430 (md_parse_option): Updated ARM7 parsing to allow 't' for
1431 thumb/halfword support, aswell as 'm' for long multiply.
1432 (md_show_usage): Updated help message.
1433 (md_assemble): Check that instruction flags are applicated to the
1434 current cpu variant.
1435 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
1436 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
1437 signextension instructions.
1438 (do_ldst): Generate halfword and signextension variants if
1439 mnemonic flags match.
1440 (ldst_extend): Do not allow shifts in the offset field of halfword
1441 or signextension instructions.
1442 (validate_offset_imm): Provide check on halfword and signextension
1444 (add_to_lit_pool): Merge identical literal pool values.
1448 * config/tc-hppa.c (selector_table): Add 'E' selector.
1449 (cons_fix_new_hppa): Don't coke on e_esel.
1450 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
1451 to help generate exception handling tables.
1452 (md_apply_fix): Don't try to apply fixups with an e_esel
1454 (hppa_fix_adjustable): Fixups with e_esel selectors
1459 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
1465 * config/tc-d10v.c: Added lots of error checking. Added hacks
1466 to support accumulator shifts.
1471 * symbols.c (S_SET_EXTERNAL): Let .weak override.
1472 (S_CLEAR_EXTERNAL): Likewise.
1473 (S_SET_WEAK): Remove error; just let .weak override.
1478 * config/tc-d10v.c (md_assemble): Now handles multiline
1483 * config/tc-d10v.c: Fix packaging bug. Added range checking.
1484 Added kludge for divs instruction. Fixed minor problem with
1485 multiple text sections.
1486 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1490 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
1491 addresses to be right-shifted by 2.
1495 * config/tc-d10v.c: Many changes to get relocs working.
1496 (register_name): No longer creates a symbol for register names.
1497 (pre_defined_registers): moved to opcodes/d10v-opc.c.
1498 (d10v_insert_operand): Now works correctly for either container.
1499 * config/tc-d10v.h (d10v_cleanup): Declare.
1504 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
1505 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
1509 * tc-alpha.c: Patches to track current minimum alignment to reduce
1510 the number of fragments created with frag_align.
1511 (alpha_current_align): New static variable.
1512 (s_alpha_text): Reset alignment to 0.
1513 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
1514 (s_alpha_stringer, s_alpha_space): New functions.
1515 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
1516 (alpha_cons_align): New function to replace both of them.
1517 (emit_insn): Only align if alpha_current_align is less than 2;
1518 reset alpha_current_align to 2.
1519 (s_alpha_gprel32): Likewise.
1520 (s_alpha_section): New function. Basically duplicate the other
1521 alpha section change hooks. Only define for ELF.
1522 (s_alpha_float_cons): Simplify alignment handling.
1523 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
1524 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
1525 Don't define the s_alpha_cons pseudo-ops. Do define
1526 s_alpha_stringer and s_alpha_space pseudo-ops.
1527 (alpha_align): Skip if less than current default alignment. Set
1529 * tc-alpha.h (md_flush_pending_output): Remove.
1530 (md_cons_align): Add.
1532 * tc-alpha.c: Add oodles of function description comments.
1533 (md_bignum_to_chars): Remove; there are no callers.
1534 (md_show_usage): Mention some more variants.
1539 * configure.in (sparc-*-sysv4*): New target.
1540 * configure: Rebuild.
1542 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
1543 uaxword to use s_uacons.
1544 (sparc_no_align_cons): New static variable.
1545 (s_uacons): New static function.
1546 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
1549 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
1551 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
1552 5 with regard to shared libraries.
1553 * configure: Rebuild.
1555 * config/tc-m68k.c (m68k_ip): Use the correct length when
1556 allocating space for the unsupported architecture error message.
1561 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
1565 * config/tc-d10v.c: New file.
1566 * config/tc-d10v.h: New file.
1567 * configure (d10v-*-elf): New target.
1568 * configure.in (d10v-*-elf): New target.
1573 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
1577 * config/tc-alpha.c (alpha_align): Change fill parameter
1578 to a pointer. Take NULL as 0 or nop depending on section. Change
1580 (s_alpha_align): Rename local variables.
1582 * doc/as.texinfo (.align): Document action of omitted
1587 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
1588 when an unsupported PC relative reloc is seen, rather than calling
1591 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
1592 Instead, check state when '#' comment is seen.
1596 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
1598 (tc_gen_reloc): If fixup was changed to be PC relative, change
1599 reloc type accordingly. Use name of reloc in error message.
1601 * as.h: Don't define const or volatile.
1602 * flonum.h: Don't define const.
1604 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
1605 if fx_pcrel is set. Correct setting the addend case in the
1606 OBJ_ELF case (from Andreas Schwab
1608 (md_show_usage): Correct -mfc5200 to -m5200.
1612 * doc/c-m68k.texi: Document -m5200 flag.
1613 * doc/as.texinfo: Likewise.
1615 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
1620 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
1621 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
1625 * Makefile.in (VERSION): Set to cygnus-2.7.1.
1627 * Released binutils 2.7.
1631 * config/tc-mips.c (mips_ip): Only perform range check when
1632 dealing with O_constant expressions.
1636 * m68k-parse.h (m68k_register): Add new coldfile control
1639 * config/tc-m68k.c (mcf5200_control_regs): New variable,
1640 array of control registers for the coldfire.
1641 (cpu_of_arch): Added mcf5200.
1642 (archs): Added mcf5200.
1643 (init_table): Add new control registers.
1644 (m68k_ip): Added support for new control registers.
1645 (m68k_init_after_args): Likewise.
1647 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
1651 * read.h (is_it_end_of_statement): Declare.
1652 * read.c (is_it_end_of_statement): Remove declaration.
1654 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
1656 (md_assemble): Fix handling of @l with an unsigned constant. Add
1657 default case to reloc switch.
1659 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
1662 * ecoff.c (last_lineno): New static variable.
1663 (add_procedure): Set last_lineno.
1664 (ecoff_directive_loc): Likewise.
1665 (ecoff_generate_asm_lineno): Likewise.
1666 (ecoff_fix_loc): New function.
1667 * ecoff.h (ecoff_fix_loc): Declare.
1668 * config/tc-mips.c (append_insn): When inserting nops, and using
1669 ECOFF debugging, call ecoff_fix_loc.
1673 * config/tc-h8300.c (build_bytes): If an operand type is
1674 marked as SRC_IN_DST retrieve it from the "destination" op.
1678 * configure.in (arm-*-riscix*): Set emulation to riscix.
1679 * configure: Rebuild.
1680 * config/te-riscix.h: New file to define TE_RISCIX.
1682 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
1686 * Makefile.in (config.status): Just run config.status as other
1691 * mpw-config.in (TARGET_OS): Add definition to conf.
1695 * config/tc-mips.c (append_insn): Parenthesize
1696 cop_interlocks expressions.
1700 * listing.c (listing_print): Close the listing file if it is not
1701 stdout. Close the other files opened for the listing.
1703 * config/tc-sparc.h (md_cons_align): Define.
1704 (sparc_cons_align): Declare.
1705 (HANDLE_ALIGN): Define.
1706 (sparc_handle_align): Declare.
1707 * config/tc-sparc.c (sparc_cons_align): New function.
1708 (sparc_handle_align): New function.
1709 * read.c (cons_worker): Call md_cons_align if it is defined.
1711 * as.h (struct frag): Add fr_file and fr_line fields.
1712 * frags.c (frag_new): Set fr_file and fr_line.
1713 (frag_var): Likewise.
1714 (frag_variant): Likewise.
1716 * as.h (struct frag): Remove unused align_mask and align_offset
1719 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
1724 * configure.in (mips-*-osf*): New target.
1725 * configure: Rebuild.
1727 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
1731 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
1732 between co-processor comparisons and branches for the VR4300.
1736 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
1737 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
1739 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
1740 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
1741 includedir): Use autoconf set values.
1746 * listing.c (listing_eject): Don't do anything if listing is 0.
1747 (listing_list): Likewise.
1748 (listing_source_line): Likewise.
1749 (listing_title): Don't save title if listing is 0.
1750 (listing_source_file): Check listing rather than listing_tail.
1752 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
1753 using shared libraries.
1754 * configure: Rebuild.
1758 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
1759 absolute expression if a relocation type was specified.
1763 * configure.in: Add support for *-*-rtems* configurations.
1764 * configure: Rebuild.
1768 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
1769 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
1770 opcodes library for alpha.
1771 * configure: Rebuild with autoconf 2.10.
1772 * config/tc-alpha.c: Substantial rewrite to add ELF support and
1773 use new opcode table.
1774 * config/tc-alpha.h (md_undefined_symbol): Don't define.
1775 (LOCAL_LABEL): Define differently if OBJ_ELF.
1776 (FAKE_LABEL_NAME): Define if OBJ_ELF.
1777 * config/alpha-opcode.h: Remove.
1778 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
1779 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
1780 include/opcode/alpha.h rather than config/alpha-opcode.h.
1784 * config/obj-aout.c (obj_emit_relocations): Give an error if the
1785 relocation symbol was not resolved.
1786 * config/obj-coff.c (do_relocs_for): Likewise.
1788 * write.c (adjust_reloc_syms): Refetch the symbol section after
1789 calling S_GET_VALUE, since it may have changed.
1791 * expr.c (struct expr_symbol_line): Define.
1792 (expr_symbol_lines): New static variable.
1793 (make_expr_symbol): Add entry to expr_symbol_lines.
1794 (expr_symbol_where): New function.
1795 * expr.h: Use extern on function declarations.
1796 (expr_symbol_where): Declare.
1797 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
1798 rather than printing the meaningless name of an expression
1803 * config/tc-i386.c (md_number_to_chars): Deleted.
1804 * config/tc-i386.h (md_number_to_chars): New macro.
1806 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
1808 (load_symbol_address, load_expression): Use build_mem.
1809 (build_operate): New function.
1810 (emit_addq_r): Use it.
1814 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
1818 * config/tc-vax.h (NOP_OPCODE): Define.
1822 * config/obj-vms.h (S_IS_COMMON): Define.
1823 (S_IS_LOCAL): Check for \002 as well as \001.
1824 (LONGWORD_ALIGNMENT): New macro.
1825 (SUB_SEGMENT_ALIGN): Use it.
1829 * config/vms-conf.h: Reconcile with conf.in.
1833 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
1836 * doc/as.texinfo: Document that any number of hex digits can
1839 * as.c (struct defsym_list): Define.
1840 (defsyms): New static variable.
1841 (parse_args): Just put --defsym arguments on defsyms list, rather
1843 (main): Define defsyms after output file is created.
1845 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
1848 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
1849 insn_error rather than calling check_absolute_expr.
1851 * as.c (emulation_name): Remove unused static variable.
1852 (default_emul_bfd_name): Add return NULL to avoid warning.
1853 * ecoff.c (ecoff_stab): Remove unused variables name and
1855 * frags.c (frag_new): Remove unused variable tmp.
1856 * hash.c (hash_grow): Parenthesize + within <<.
1857 (hash_print_statistics): Use %lu, not %d, to print unsigned
1859 * messages.c: Include "libiberty.h".
1860 (fprint_value): Add cast to avoid printf warning.
1861 (sprint_value): Likewise.
1862 * read.c: Include "ecoff.h".
1863 (emit_expr): Add casts to avoid printf warnings.
1864 * read.h: Use extern for function declarations.
1865 (pop_insert): Declare.
1866 * stabs.c: Include "ecoff.h".
1867 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
1868 former_last_fragP, and new_fragP.
1869 * subsegs.h (subsegs_print_statistics): Declare.
1870 * symbols.c (debug_verify_symchain): Change macro to discard
1872 * write.c (dump_section_relocs): Likewise.
1873 * write.h: Use extern for function declarations.
1874 (write_print_statistics): Declare.
1875 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
1877 * config/e-mipself.c (mipself_bfd_name): Likewise.
1878 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
1880 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
1881 emit relocations against external symbols.
1883 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
1884 message if bfd_reloc_type_lookup fails, rather than calling
1887 * config/tc-alpha.c (alpha_force_relocation): Add
1888 BFD_RELOC_12_PCREL to switch.
1892 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
1893 TE_PE (Lfoo, not .Lfoo).
1897 * read.c (s_fill): Don't warn about a zero repeat count.
1899 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
1900 coprocessor register.
1902 * config/tc-i386.c (md_assemble): When checking the size of a
1903 register to set the size of an instruction, do a bitwise and with
1904 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
1909 * config/tc-h8300.c (parse_reg): Tweak error messages.
1910 (build_bytes): Likewise.
1911 (skip_colonthing): Handle :32 suffix.
1912 (get_specific): Promote L_24 to L_32 if it makes a match.
1913 Don't always promote L_8 to L_16.
1914 (do_a_fix_imm): Clean up L_32 and L_24 handling.
1916 * config/tc-h8300.c (Smode): New variable.
1917 (h8300hmode): Turn off Hmode.
1918 (h8300smode): New function. Turn on Smode and Hmode.
1919 (md_pseudo_table): New ".h8300s" pseudo-op.
1920 (parse_reg): Handle "exr" register.
1921 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
1922 Handle "mach" and "machl" operands for ldmac.
1923 (get_specific): Handle "stm.l" and "ldm.l".
1924 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
1925 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
1930 * doc/as.texinfo: Reorder chapter of machine dependent options so
1931 that it is sorted by chip name.
1933 * doc/as.texinfo: Use consistant spelling of Vax.
1934 * doc/c-vax.texi: Likewise.
1938 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
1940 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
1941 (md_apply_fix): Likewise.
1942 (pa_try): New function.
1943 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
1947 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
1948 (pa_level): New function.
1952 * listing.c (listing_newline): Don't do anything if listing is 0.
1956 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
1957 symbol_create, not symbol_new.
1961 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
1962 absolute symbol unless TC_FORCE_RELOCATION returns true.
1964 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
1966 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
1967 the symbol to a pointer. Don't set previous_file_symbol.
1968 Simplify symbol list rearrangement.
1969 (coff_frob_symbol): Don't do anything with C_FILE symbols.
1970 (coff_adjust_symtab): Don't check previous_file_symbol.
1974 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
1976 (md_pseudo_table): Add ppc_elf_lcomm.
1980 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
1981 d0{24:1},d0' works without an immediate prefix on the bit numbers.
1982 (md_begin): Add digits to alt_notend_table.
1983 (md_parse_option): Make s a const pointer.
1985 * config/tc-sparc.c (md_pseudo_table): Add "empty".
1986 (s_empty): New static function.
1988 * config/obj-coff.c (struct filename_list): Only define if not
1990 (filename_list_head, filename_list_tail): Likewise.
1991 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
1992 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
1993 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
1994 have a section of N_ABS rather than N_DEBUG. If we do a merge,
1995 remove the new symbol from the list.
1996 (obj_coff_endef, both versions): Call tag_insert even if there is
1997 an old symbol with the same name, if the old symbol does not
1999 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
2000 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
2001 symbol that will be moved to the end of the symbol list.
2002 (coff_adjust_section_syms): Always call section_symbol for .text,
2004 (coff_frob_section): Likewise. Also, remove unused variable
2007 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
2008 manipulating frags directly.
2009 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
2012 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
2013 whether the next character is '*' before checking whether we are
2014 at the start of a line. Permit LINE_COMMENT_START to start a
2015 comment in state 1 (seen some whitespace) as well, to match the
2018 * gasp.c (do_align): Permit a fill value for .align.
2022 * read.c (next_char_of_string): Warn if a newline is seen in the
2023 middle of a string. Call bump_line_counters when appropriate.
2027 * symbols.c (colon): Use LOCAL_LABEL.
2031 * Makefile.in (install): Don't check to see if tooldir exists.
2032 Make $(tooldir) and $(tooldir)/bin.
2036 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
2041 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
2042 insn_error. Don't put the string "ERROR" in insn_error. Set
2043 insn_error rather than calling as_warn for an unsupported opcode.
2047 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
2048 before permitting -64.
2049 * output-file.c (output_file_create): Remove duplicate
2054 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
2055 section rather than a .reginfo section.
2056 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
2059 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
2060 bit value with the high bit set.
2064 * read.c (s_lcomm): Set section flags for .sbss section.
2066 * config/tc-mips.c (mips_64): New static variable.
2067 (mips_target_format): If mips_64, return elf64 targets rather than
2069 (md_longopts): Add "32" and "64".
2070 (md_parse_option): Handle -32 and -64.
2071 (md_show_usage): Mention -32 and -64.
2072 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
2077 * config/tc-ppc.c (comment_chars): Make '!' a comment character
2078 for Solaris compatibility.
2080 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
2081 .stabd with 4 arguments into a .stabn.
2085 * config/tc-mips.c (macro): When passing X_add_number to
2086 macro_build, cast it to int first.
2090 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
2091 created by emit_expr.
2093 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
2094 as_perror rather than assert.
2098 * config/tc-mips.c (mips_ip): Mark sections created to hold
2099 floating point information as read only.
2103 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
2107 * read.c (potable): Add .skip as a synonym for .space.
2109 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
2110 4 arguments, providing the 4th argument is 0, to allow
2111 compatibility with the Solaris assembler.
2115 * config/tc-sh.h (struct sh_segment_info_type): Define.
2116 (TC_SEGMENT_INFO_TYPE): Define.
2117 (sh_frob_label): Declare.
2118 (tc_frob_label): Define.
2119 (sh_flush_pending_output): Declare.
2120 (md_flush_pending_output): Define.
2121 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
2122 reloc before the instruction if necessary.
2123 (sh_frob_label): New function.
2124 (sh_flush_pending_output): New function.
2125 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
2126 when looking for the reloc for the target of .uses.
2127 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
2128 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
2130 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
2131 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
2132 the absolute symbol.
2134 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
2135 TC_SEGMENT_INFO_TYPE is defined.
2139 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
2140 matches the register size.
2144 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
2146 (do_relocs_for): Likewise.
2147 (fixup_segment): Don't just quit if linkrelax is set. Try to
2148 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
2153 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
2154 relative relocations with -mrelocatable. Also allow unfixed
2155 relocs in .ex_shared.
2159 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
2160 non-zero before assuming this is a long file name.
2161 (w_strings): Likewise.
2162 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
2164 * config/obj-coff.c (w_strings): Move declaration of i inside
2165 #ifdef block which uses it.
2169 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
2170 "relaxmode". Output relocs which identify various relaxing
2171 possibilities for mov.[bwl] instructions.
2172 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
2176 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
2177 (MD_APPLY_FIX3): Define.
2178 * config/tc-arm.c (my_get_expression): Only watch for bad segments
2180 (md_apply_fix3): Renamed from md_apply_fix.
2181 If pcrel reloc and symbol is in different section, undo effects
2186 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
2187 any reloc with an LR% or RR% field selector for SOM.
2191 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
2196 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
2198 (yank_symbols): Likewise.
2202 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
2203 BFD_ASSEMBLER, add name field.
2204 * config/obj-coff.c: Include "libiberty.h".
2205 (coff_header_append): Handle long section names.
2206 (crawl_symbols): Just use the name field for the symbol name,
2207 without worrying about null byte termination.
2208 (w_strings): Handle long section names.
2209 (write_object_file): Likewise. Also, use the name field, rather
2211 (obj_coff_add_segment): Permit long section names.
2212 (obj_coff_init_stab_section): Use the name field, rather than
2214 (adjust_stab_section): Likewise.
2215 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
2217 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
2218 (mem_fmt): Likewise.
2219 (md_apply_fix): Don't check fx_im_disp field.
2223 * configure.in: Add * after sparc*-*-vxworks.
2224 * configure: Rebuild.
2226 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
2227 define it to comment_chars. Use tc_comment_chars rather than
2229 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
2230 * config/tc-m68k.h (tc_comment_chars): Define.
2231 (m68k_comment_chars): Declare.
2232 * config/tc-m68k.c (m68k_comment_chars): Rename from
2233 comment_chars. Change into a pointer rather than an array.
2234 (md_longopts): Add "bitwise-or".
2235 (md_parse_option): Handle OPTION_BITWISE_OR.
2236 (md_show_usage): Mention --bitwise-or.
2237 * doc/c-m68k.texi: Document --bitwise-or.
2241 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
2244 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
2245 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
2246 correct handling of input line pointer, and call
2247 demand_empty_rest_of_line.
2251 * config/tc-sparc.c (in_bitfield_range): New static function.
2252 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
2253 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
2254 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
2255 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
2256 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
2260 * config/obj-coff.c: BFD_ASSEMBLER:
2261 (coff_last_bf): New static variable.
2262 (coff_frob_symbol): Set endndx of a .bf symbol.
2264 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
2265 (last_bfP): New static variable.
2266 (yank_symbols): Set endndx of a .bf symbol.
2270 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
2271 and -s options. Add -be for good measure.
2275 * read.c (s_space): Support non-constant fill value. Handle fill
2276 value correctly for a size other than 1.
2280 * config/tc-arm.c (my_get_float_expression): Update call to
2281 gen_to_words, X_PRECISION changed from 6 to 5.
2285 * config/tc-ppc.c (register_name,reg_name_search): Move register
2286 name lookup from PE specific code to all targets. Add support for
2287 -mregnames/-mno-regnames to control whether register names are
2289 (md_assemble): Call register_name for all platforms.
2290 (md_parse_option): Add support for -mregnames/-mno-regnames.
2292 * configure.in (powerpcle*-*-solaris): Add support.
2293 (powerpc*-*-linux): Ditto.
2294 * configure: Regenerate.
2296 * config/ppc-sol.mt: New config file for PowerPC Solaris.
2300 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
2301 be associated with a single %lo reloc.
2303 * config/tc-mips.c (load_address): Cast X_add_number to valueT
2304 before comparing against MAX_GPREL_OFFSET, so that negative
2305 numbers are handled correctly.
2310 * config/tc-sparc.c (last_insn): New static variable.
2311 (md_assemble): Warn about putting floating point branches in a
2312 delay slot. If architecture is less than v9, insert NOP
2313 instructions between floating point instructions and floating
2314 point branches. (The SunOS assembler does both these operations.)
2315 Save the last instruction opcode.
2316 (sparc_ip): Add pinsn parameter. Change caller.
2318 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
2319 for byte jump to next instruction to skip empty frags.
2323 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
2324 for addq with a 16 bit signed value, just emit a lda instruction
2329 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
2334 * configure.in: Permit --enable-shared to specify a list of
2336 * configure: Rebuild.
2340 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
2341 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
2345 * config/atof-ieee.c: Fix handling of denormalized extended
2346 precision numbers and overflow/underflow detection.
2347 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
2348 not include the 16 bit gap in the m68k extended precision format.
2352 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
2354 * configure: Rebuild.
2358 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
2359 be accessed using the $gp register.
2360 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
2364 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
2365 bfd_mach_sparc_sparclet if current_architecture is sparclet.
2369 * read.c (get_line_sb): Bump line counters based on
2370 input_line_pointer[-1], not *input_line_pointer. Don't bother to
2371 call LISTING_NEWLINE.
2372 (s_macro): Don't call demand_empty_rest_of_line.
2373 * app.c (do_scrub_chars): When handling C style comments, unget
2378 * read.h (enum linkonce_type): Define.
2379 (s_linkonce): Declare.
2380 * read.c (potable): Add "linkonce".
2381 (s_linkonce): New function.
2382 * subsegs.h (segment_info_type): Add linkonce field to
2383 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
2384 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
2385 (obj_coff_pe_handle_link_once): Declare if TE_PE.
2386 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
2388 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
2389 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
2390 entry based on the linkonce field in segment_info.
2391 * doc/as.texinfo: Document .linkonce.
2395 * doc/as.1: Changed to be recognized by catman -w on Solaris.
2399 * stabs.c (s_stab_generic): Call the listing functions before
2400 doing the rest of the processing, which may involve freeing the
2401 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
2402 SEPARATE_STAB_SECTIONS case.
2404 * config/tc-hppa.c: Remove nested comment.
2405 (tc_gen_reloc): Move label done inside the ifdef in which it is
2407 (md_apply_fix): Pass pointers to correct types to libhppa.h
2408 functions. Always return a value.
2410 * config/tc-mips.h (tc_frob_file): Define.
2411 (mips_frob_file): Declare.
2412 * config/tc-mips.c (struct mips_hi_fixup): Define.
2413 (mips_hi_fixup_list): New static variable.
2414 (imm_unmatched_hi): New static variable.
2415 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
2416 offset_reloc. Pass imm_unmatched_hi to append_insn.
2417 (append_insn): Add unmatched_hi parameter. If it is set, add the
2418 new fixup to mips_hi_fixup_list. Change all callers.
2419 (mips_ip): Set imm_unmatched_hi when appropriate.
2420 (mips_frob_file): New function.
2424 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
2425 * configure: Regenerated.
2429 * as.c (main): Call bfd_set_error_program_name.
2433 * as.h (strdup): Don't declare.
2434 * stabs.c: Include libiberty.h
2435 (get_stab_string_offset): Use xstrdup rather than strdup.
2436 (s_stab_generic): Likewise.
2437 * as.c (parse_args): Likewise.
2438 * read.c (s_mri_sect): Likewise.
2440 * gasp.c (change_base): Recognize \(...) construct documented to
2441 pass through enclosed characters literally through to the output.
2442 (process_assigns): Likewise. Also, be more careful to avoid
2443 looking past the end of the buffer.
2447 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
2452 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
2456 * config/alpha-opcode.h: Added cvtst instruction.
2460 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
2461 string. Extract the addend from the result of expression.
2465 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
2466 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
2467 mode, change the state to 3 rather than 1.
2471 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
2472 correspond to 11 March change.
2476 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
2477 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
2481 * configure: Rebuild with autoconf 2.8.
2485 * config/atof-ieee.c (gen_to_words): Improve handling of
2486 X_PRECISION numbers. Based on patches from Andreas Schwab
2491 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
2493 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
2494 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
2495 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
2496 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
2501 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
2505 * config/tc-hppa.c (SUBSEG_MILLI): Define.
2506 (pa_def_subspaces): Add $MILLICODE$.
2507 (pa_spaces_begin): Set section flags for $MILLICODE$.
2511 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
2512 not SHT_NOBITS. Don't tamper with flags based on type if a
2513 special section was found (revert Feb 29 change).
2515 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
2516 if aligning to a longword boundary or greater.
2520 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
2524 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
2525 * configure: Regenerated.
2529 * config/tc-hppa.c: Add default definitions for R_N0SEL and
2530 R_N1SEL since they're not defined for old versions of hpux.
2532 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
2533 Set "sym_ptr_ptr" and "addend" fields to dummy values for
2534 R_N0SEL and R_N1SEL.
2538 * flonum-konst.c: Add two more constants for 1e+-2048 and
2539 1e+-4096, and correct the other constants.
2541 * symbols.c (resolve_symbol_value): Handle O_logical_not.
2545 * config/obj-elf.c (obj_elf_section): Allow predefined section
2546 types to set the nobits type. Avoid a shadowed declaration.
2550 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
2551 reduce relocs using e_nlrsel field selectors.
2553 * write.c (fix_new_exp): Don't use #elif. Some compilers
2556 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
2558 (pa_chk_field_selector): Handle new field selectors for SOM.
2562 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
2563 * configure: Rebuild.
2564 * config/te-linux.h (LOCAL_LABELS_FB): Define.
2565 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
2567 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
2572 * mpw-make.sed: Update to handle shared library support.
2576 * config/tc-sparc.c (sparc_ip): Print all architectures that support
2577 the insn on mismatch.
2581 * configure.in: Add support for a29-coff.
2582 * configure: Rebuild.
2586 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
2587 of the section when looking for the R_SH_USES fixup, because the
2588 frag addresses have not yet been adjusted.
2590 * gdbinit.in: Set a breakpoint on as_warn_where.
2592 * config/tc-mips.c (macro): Add missing arguments to macro_build
2597 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
2598 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
2602 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
2606 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
2607 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
2608 Recognize [uU] format args as sparclet cpregs.
2612 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
2617 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
2618 well as S_IS_EXTERNAL.
2619 (tc_fix_adjustable): Likewise.
2620 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
2621 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
2623 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
2624 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
2628 * config/tc-sparc.c (max_architecture): Change to sparclite for
2630 (default_compatible): Delete.
2631 (sparc_ffs): New function.
2632 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
2633 (sparc_ip): Rewrite architecture match and bump logic.
2637 * configure.in: Check for 'do not mix' from native linker before
2638 trying to use -rpath.
2639 * configure: Rebuild.
2643 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2644 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2645 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2646 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
2648 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
2649 to the symbol lnnoptr field, to get the correct file offset.
2653 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
2654 symbols that are not function, file, or section symbols to be
2659 * configure: Set and substitute RPATH_ENVVAR.
2660 * configure: Rebuild.
2661 * Makefile.in (RPATH_ENVVAR): New variable.
2662 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
2665 doesn't need opcodes. If configuring shared, opcodes needs bfd.
2666 * configure: Rebuild.
2670 * read.c (s_mri_sect): Don't return '\0' in type. Set all
2671 appropriate flags in BFD section.
2673 * configure.in (m68k-*-psos*): New target.
2674 * configure: Rebuild.
2675 * config/te-psos.h: New file.
2676 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
2682 * configure.in: Remove duplicate setting of cpu_type. Check
2683 whether opcodes library is required for on all targets, not just
2685 * configure: Rebuild.
2687 * config/tc-mips.c (mips_big_got): New static variable.
2688 (s_extern): Don't declare.
2689 (reg_needs_delay): New static function.
2690 (macro_build): Permit GOT/CALL_HI/LO relocs.
2691 (macro_build_lui): If place is not NULL, use the number in the
2693 (load_address): Handle mips_big_got case.
2694 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
2696 (OPTION_XGOT): Define.
2697 (md_longopts): Add "xgot" if OBJ_ELF.
2698 (md_parse_option): Handle -xgot.
2699 (md_show_usage): Mention -xgot.
2700 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
2701 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
2705 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
2706 the byte relocation, point it to the low byte of the word.
2710 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
2711 * configure: Rebuild.
2715 * configure.in: Recognize any sparc* cpu.
2716 * configure: Regenerated.
2720 * read.c (potable): Add "mri" and ".mri".
2721 (s_mri): New function.
2722 * read.h (s_mri): Declare.
2723 * app.c (scrub_m68k_mri): New static variable.
2724 (mri_pseudo): New static variable.
2725 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
2726 flag_m68k_mri. Initialize scrub_m68k_mri.
2727 (mri_state, mri_last_ch): New static variables.
2728 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
2730 (app_push): Save new fields.
2731 (app_pop): Restore new fields.
2732 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
2733 If TC_M68K, use a trivial state machine to look for occurrences of
2734 the .mri pseudo-op, and change the mode appropriately.
2735 * as.h (do_scrub_begin): Update prototype.
2736 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
2738 * config/tc-m68k.c (reg_prefix_optional_seen): New static
2740 (m68k_mri_mode_change): New function.
2741 (md_parse_option): Set reg_prefix_optional_seen.
2742 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
2743 (MRI_MODE_CHANGE): Define.
2744 * doc/as.texinfo: Document .mri pseudo-op.
2746 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
2751 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
2752 * config/m68k-parse.h (enum pic_relocation): Define.
2753 (struct m68k_exp): Add pic_reloc field.
2754 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
2755 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
2756 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
2758 * config/tc-m68k.c: Delete definition of NO_RELOC.
2759 (struct m68k_it): Add pic_reloc field.
2760 (add_fix): Copy over pic_reloc field.
2761 (md_pseudo_table): Interpret .align parameter as byte count.
2762 (mote_pseudo_table): Likewise.
2763 (tc_m68k_fix_adjustable): New function.
2764 (get_reloc_code): New function.
2765 (md_assemble): Use it as last argument to fix_new_exp.
2766 (md_apply_fix_2): For a relocation against a symbol don't put the
2767 addend into the data.
2768 (tc_gen_reloc): Different addend computation for OBJ_ELF.
2769 (m68k_ip): Don't relax an operand that requires pic relocation.
2770 (md_begin): Align .text, .data and .bss on 4 byte boundary by
2772 * write.c (fixup_segment): Don't add symbol value to addend if
2773 TC_M68K and OBJ_ELF.
2774 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
2775 (motorola_operand): Add rule for `(zapc, EXPR)'.
2779 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
2780 *really* is permissible before the comma.
2784 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
2785 inserting nop instructions.
2789 * config/te-aux.h: Change include of aux.h to aux-coff.h.
2793 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
2794 class setting for a CALLNAME symbol in COFF.
2796 * read.c (potable): Pass negative numbers for new .balign[wl] and
2797 .p2align[wl] pseudo-ops.
2798 (s_align_bytes): Treat a negative argument as specifying the fill
2800 (s_align_ptwo): Likewise.
2804 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
2805 (do_align): Take new len parameter. Change all callers. Pass it
2807 (s_align_bytes): Arg now indicates the length of the fill pattern.
2808 (s_align_ptwo): Likewise.
2809 * config/obj-coff.c (write_object_file): Pass length to
2811 * config/tc-i386.h (md_do_align): Take new len parameter.
2812 * config/tc-m88k.h (md_do_align): Likewise.
2813 * config/tc-m88k.c (m88k_do_align): Likewise.
2814 * config/tc-sh.h (md_do_align): Likewise.
2815 * config/tc-sh.c (sh_do_align): Likewise.
2816 * doc/as.texinfo: Document new pseudo-ops.
2818 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
2819 done in cvt_frag_to_fill.
2821 * config/tc-sh.h (sh_do_align): Declare.
2822 (md_do_align): Define.
2823 * config/tc-sh.c (sh_do_align): New function.
2825 * ecoff.c (ecoff_build_lineno): Don't try to store the address
2826 difference if the next address is before the current one.
2828 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
2829 (archs): Initialize new field.
2830 (m68k_ip): Don't list alias names when listing CPUs which support
2833 * as.c (main): Call parse_args before read_begin.
2834 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
2835 front of generated pseudo-ops.
2836 * read.c (potable): Ignore "name".
2837 (s_app_file): Permit a single quote after the string, since one
2838 may appear in m68k MRI mode.
2840 * configure.in: Check for --enable-shared. If linking against
2841 shared BFD and opcodes, fix library name on SunOS, and try to set
2843 * configure: Rebuild.
2847 * as.h (flag_m68k_mri): Declare.
2848 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
2849 * Many files: For MRI syntax that is specific to the m68k MRI
2850 assembler, check flag_m68k_mri rather than flag_mri or
2851 MRI_MODE_NEEDS_PSEUDO_DOT.
2855 * config/tc-i960.c (ARCH_HX): Define.
2857 (targ_has_sfr): Handle ARCH_HX.
2858 (targ_has_iclass): Handle ARCH_HX.
2859 (tc_coff_fix2rtype): Add return 0 to avoid warning.
2860 (tc_headers_hook): If the architecture was specified explicitly,
2861 use it when setting the flags. Set the extern variable coff_flags
2862 rather than headers->filehdr.f_flags, since the latter is set
2863 unconditionally in obj-coff.c.
2864 (i960_handle_align): Remove unused variable fixp.
2866 Support for building bfd and opcodes as shared libraries, based on
2868 * configure.in: Set OPCODES and BFD to search directories.
2869 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
2870 * configure: Rebuild.
2871 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
2872 (LIBDEPS): New variable.
2873 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
2874 $(HLDFLAGS) in link.
2875 (check): Set LD_LIBRARY_PATH in the environment.
2879 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
2880 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
2881 read-only, not read/write.
2885 * Makefile.in (INSTALL_XFORM): Remove -e.
2889 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
2890 than comparing S_GET_SEGMENT to undefined_section.
2891 (write_object_file): Skip symbols which were equated to an
2892 undefined or common symbol.
2893 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
2894 S_IS_COMMON rather than comparing S_GET_SEGMENT to
2896 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
2897 or S_IS_COMMON call S_GET_VALUE.
2898 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
2899 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
2900 to an undefined or common symbol, convert the reloc to be against
2902 (obj_crawl_symbol_chain): Skip symbols which were equated to an
2903 undefined or common symbol.
2904 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
2905 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
2906 to an undefined or common symbol, convert the reloc to be against
2908 (obj_crawl_symbol_chain): Skip symbols which were equated to an
2909 undefined or common symbol.
2910 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
2911 S_IS_COMMON rather than comparing S_GET_SEGMENT to
2913 (yank_symbols): Skip symbols which were equated to an undefined or
2918 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
2919 * config/obj-bout.h (S_IS_LOCAL): Likewise.
2921 * configure.in: Make sure we only add m68k-parse.o to
2922 ${extra_objects} once, no matter how many m68k targets have been
2924 * configure: Rebuild.
2928 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
2929 * configure: Rebuild.
2933 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
2934 * doc/c-m68k.texi: Document .extend and .ldouble.
2936 * configure.in (m68*-apple-aux*): New target.
2937 * config/te-aux.h: New file.
2938 * config/obj-coff.c (compare_external_relocs): New static function
2940 (do_relocs_for): Sort relocs if TE_AUX.
2941 (fixup_segment): If TE_AUX, store common symbol value in segment.
2942 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
2946 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
2948 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
2950 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
2951 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
2953 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
2954 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
2955 an undefined symbol and the size of a function symbol.
2956 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
2957 for all common symbols.
2961 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
2964 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
2969 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
2972 * ecoff.c (localsym_t): Add addend field.
2973 (add_ecoff_symbol): Add addend argument. Change all callers.
2974 (coff_sym_value): Make static.
2975 (coff_sym_addend): New static variable.
2976 (ecoff_directive_def): Initialize coff_sym_addend.
2977 (ecoff_directive_val): Accept symbol + constant.
2978 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
2979 (ecoff_build_symbols): Include the addend in the symbol value.
2983 * config/tc-ppc.c (md_assemble): Ignore overflow on
2984 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
2988 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
2989 bytes in size, so offset appropriately in big endian mode when
2990 writing the bottom 2 bytes.
2994 * config/tc-sparc.c (default_compatible): New static local.
2995 (md_begin): Initialize it. Rewrite warn_on_bump handling.
2996 (sparc_ip): If no architecture or -bump specified, don't mark as
2997 mismatched those in default_compatible.
3002 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
3003 * configure: Rebuild.
3004 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
3005 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
3006 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
3007 (sco_id): Declare if SCO_ELF.
3011 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
3012 ({max,warn_after}_architecture): New static locals.
3013 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
3014 If both architecture and -bump requested, set max_architecture to max.
3015 (sparc_md_end): Simplify.
3016 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
3017 Rewrite code to bump architecture and check for conflicts.
3018 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
3019 with Solaris assembler.
3020 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
3021 (md_show_usage): Update.
3025 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
3029 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
3030 #ifdef sparcv9 when choosing value.
3033 (sparc_md_end): Declare.
3034 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
3035 (initial_architecture): New static local.
3036 (can_bump_v9_p): Likewise.
3037 (NO_V9): Delete all occurrences.
3038 (sparc_md_end): New function.
3039 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
3040 Don't bump architecture to v9 unless can_bump_v9_p set.
3041 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
3042 * configure.in (sparc64 target cpu): Don't set obj_format here.
3043 (SPARC_V9): Renamed from sparcv9.
3044 (sparc64-*-elf*): Define SPARC_ARCH64.
3045 * configure: Regenerated.
3046 * acconfig.h (SPARC_V9): Renamed from sparcv9.
3047 (SPARC_ARCH64): Add.
3048 * conf.in: Regenerated.
3049 * config/vmsconf.h: Update.
3053 * config/tc-mips.c (load_register): Optimise "dli" loads.
3054 (md_show_usage): add "-mcpu=vr4100" to help text.
3058 * symbols.c (resolve_symbol_value): If a symbol is equated to an
3059 undefined symbol, preserve the X_op of O_symbol.
3060 (S_GET_VALUE): Fix check to permit this case.
3061 * write.c (write_relocs): If a reloc is against an undefined
3062 symbol equated to another symbol, change the reloc to be against
3064 * config/obj-coff.c (do_relocs_for): Likewise.
3066 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
3071 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
3072 computer is PowerPC.
3074 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
3076 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
3077 mentioned in the eabi.
3081 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
3082 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
3086 * config/tc-ppc.c (mapping): Add more relocation suffixes.
3090 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
3091 as a section it is ok to have unadorned -mrelocatable pointers in.
3095 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
3100 * config/obj-elf.c (obj_elf_section): Add hooks so machine
3101 dependent section attributes can be handled.
3103 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
3104 macros to add support for exclude section flag and ordered section
3107 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
3108 functions to add support for exclude section flag and ordered
3113 * subsegs.c (section_symbol): Don't try to look up the section
3114 symbol in the hash table. It should be possible to have a symbol
3115 with the same name as a section, but no connection to it.
3117 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
3120 * expr.c (operand): Skip whitespace after a close parenthesis.
3125 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
3126 as_warn_where for relocation overflow.
3127 (parse_reg): Accept register name only if next character is
3132 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
3133 addresses for static variables.
3137 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
3138 and xgot give the old toc16 relocation.
3139 (md_apply_fix3): Support all GOT relocations.
3143 * config/tc-mips.c (load_address): Correctly handle a constant in
3148 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
3149 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
3150 (build_Mbytes): Handle REG_B.
3154 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
3158 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
3162 * config/obj-coff.c (write_object_file): Set the s_align field to
3163 the number of bytes, rather than to the power of 2.
3167 * Makefile.in (DISTCLEAN_HERE): New variable.
3168 (distclean): Use it.
3169 (maintainer-clean): Depend upon clean-here rather than clean,
3170 distclean, and clean-info. Run make maintainer-clean in doc.
3171 Remove files listed in DISTCLEAN_HERE.
3172 * doc/Makefile.in (maintainer-clean realclean): Split out from
3173 distclean. Depend upon clean-info and distclean.
3177 * mac-as.r: Fix copyright and version strings.
3178 (cfrg): Use PROG_NAME instead of literal name.
3182 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
3184 * config/tc-a29k.h (tc_unrecognized_line): Define.
3185 * config/tc-a29k.c (a29k_unrecognized_line): New function.
3186 (md_operand): Handle a29k style local dollar labels.
3190 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
3194 * read.c (s_fill): If md_flush_pending_output is defined, call
3199 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
3200 Treat rs_align_code like rs_align.
3204 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
3205 converting an absolute address into a PC-relative one.
3209 * mpw-config.in: Don't always use te-generic.h for emulation.
3210 (powerpc-apple-macos): Use emulation te-macos.h.
3212 (mips-gm-magic): New configuration.
3214 * mpw-make.sed (install, install-only): Edit in Mac-specific
3219 * configure.in: Improve message about unsupported ELF targets.
3220 * configure: Rebuild.
3222 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
3223 pattern is zero. From Manfred Hollstein.
3227 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
3228 the various idata sections, we check the name on the .section pseudo.
3232 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
3233 add section's paddr to add_number; compatibility to native as and
3238 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
3240 * hash.c (struct hash_entry): Moved here...
3241 * hash.h (struct hash_entry): ...from here.
3243 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
3244 if it's already known to be null.
3248 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
3249 bit relocs converted to PC relative relocs. This turned out to
3250 add the offset from the beginning of .text twice.
3254 * stabs.c (s_stab_generic): In 's' case, free string from
3257 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
3259 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
3262 * subsegs.h (segment_info_type): Make bitfields unsigned.
3264 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
3265 and move them together. On most systems this will reduce the
3266 structure size by one word.
3267 (operatorT): Define O_max.
3268 * expr.c (expr_begin): Verify that X_op is wide enough to hold
3271 * read.c (pop_insert): Print error returned by hash table
3274 * as.c (dump_statistics): Split out from main; dump some hash
3275 table stats and target-specific stats.
3276 (start_time): No longer automatic to main.
3277 (main): Set file-level start_time and call dump_statistics at
3278 exit. Exit by calling xexit.
3279 (show_usage): Make --statistics description less specific.
3280 * subsegs.c (subsegs_print_statistics): New function.
3281 * write.c (write_print_statistics): New function.
3282 (n_fixups): New static variable.
3283 (fix_new_internal): Increment it.
3284 * read.c (read_print_statistics): New function.
3285 * read.h (read_print_statistics): Declare.
3286 * symbol.c (symbol_print_statistics): New function.
3287 * symbol.h (symbol_print_statistics): Declare.
3288 * hash.c (hash_print_statistics): New function.
3289 * hash.h (hash_print_statistics): Declare.
3290 * config/tc-i386.c (i386_print_statistics): New function.
3291 * config/tc-i386.h (i386_print_statistics): Declare.
3292 (tc_print_statistics): New macro.
3293 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
3296 * hash.c (DELETED): Rewrite to use a valid but unique address.
3297 (START_POWER): Reduce to 10.
3298 (enum stat_enum): New enumerator, replacing STAT_* index macros.
3299 Add new values for counting strcmp calls.
3300 (GROW_FACTOR): New macro.
3301 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
3303 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
3304 (hash_new): Use FULL_VALUE.
3305 (struct hash_control): Definition moved here.
3306 (hash_code): Don't mask to low bits.
3307 (hash_ask): Mask returned hash code here. Check hash value before
3308 calling strcmp; count strcmp calls.
3309 * hash.h (struct hash_control): Declare, don't define, here.
3310 (HASH_STATLENGTH): Deleted.
3311 (struct hash_entry): Add field for hash code.
3312 (hash_say, hash_apply): Don't declare.
3314 * hash.c (destroy): Return void.
3315 (applicatee): Ditto.
3316 (main): Fix declarations.
3317 (hash_apply): Return void. Argument `function' returns void. Put
3318 inside "#ifdef TEST".
3319 (hash_say): Define only if TEST is defined.
3320 * hash.h (hash_apply, hash_say): Declarations deleted.
3324 * configure: Rebuild with autoconf 2.7.
3328 * aclocal.m4 (AC_PROG_CC): Remove local definition.
3329 * configure: Rebuild with autoconf 2.6.
3333 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
3334 (ppc_stabx): Don't increment ppc_debug_name_section_size.
3336 (ppc_frob_file): Remove.
3337 * config/tc-ppc.h (tc_frob_file): Don't define.
3338 (ppc_frob_file): Don't declare.
3342 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
3343 * config/alpha-opcode.h: Include one-operand variants of jmp and
3346 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
3347 with te-delta.h in 8.3 file systems.
3348 * configure.in: Adjusted.
3352 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
3354 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
3356 * expr.c (operand): Handle MRI suffixes after unadorned 0.
3360 Version 2.6 released.
3361 * Makefile.in (VERSION): Updated to 2.6.
3363 * config/obj-coff.c (write_object_file): Change use of md_do_align
3364 to pass a pointer rather than a fill value, to match other uses.
3368 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
3370 * as.c (main): Move md_end call to just after call to
3371 perform_an_assembly_pass. Delete cpu-specific code here.
3372 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
3373 * config/tc-arm.c (md_end): Unused function deleted.
3374 * config/tc-ns32k.c (md_end): Ditto.
3376 * config/tc-i386.c (i386_align_code): New function, moved here
3377 from HANDLE_ALIGN macro.
3378 * config/tc-i386.h (HANDLE_ALIGN): Call it.
3382 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
3387 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
3388 multi-byte nops to avoid changing gcc. The necessary gcc change
3389 might break old assemblers.
3393 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
3394 generate optimal multi-byte nop instructions for ".align n"
3395 ".align n,0x90", and aligns requiring more than 15 bytes of
3396 padding still generate multiple 0x90's as before.
3400 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
3404 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
3405 * configure: Update.
3406 * mpw-make.sed: Reorder commands to make sed happier.
3407 * config/te-macos.h: New file.
3408 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
3412 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
3413 conditional branches.
3414 (md_apply_fix): Likewise.
3418 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
3419 generate a reloc which the linker will not handle correctly. Fix
3420 overflow checking--R_IREL is 18 bits, not 17.
3424 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
3425 from a PC relative reloc if TC_A29K.
3427 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
3428 code depends upon it.
3430 * config/tc-a29k.c (md_operand): Handle $float, $double, and
3431 $extend. Based on code from Eric Freudenthal
3433 * config/tc-a29k.h (LEX_DOLLAR): Define.
3434 * read.c (LEX_DOLLAR): Define if not defined.
3435 (lex_type): Use LEX_DOLLAR.
3439 * configure.in (a29k-nyu-sym1): New target, just like other a29k
3444 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
3448 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
3449 used only by "dlltool.c".
3453 * config/tc-alpha.c: Undefine inline if not __GNUC__.
3454 (md_pseudo_table): Don't define "extern".
3458 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
3460 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
3464 * macro.c (macro_expand_body): Don't warn about == with a
3465 nonexistent parameter, in case it is in a comment field.
3467 * as.c (main): On TC_A29K, call macro_init with macro_alternate
3469 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
3470 set, even if macro_alternate is set.
3471 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
3472 (sub_actual): If macro_strip_at, and kind is '@', don't look up
3473 the token unless it ended in '@'.
3474 * config/tc-a29k.c (line_separator_chars): Remove '@'.
3475 * doc/c-a29k.texi: Document macro usage on A29K.
3479 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
3481 (install_operand): Likewise.
3483 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
3487 * configure.in (m88k-motorola-sysv*): New target.
3488 * configure: Rebuild.
3489 * config/te-delta88.h: New file.
3490 * config/obj-coff.c (write_object_file): Use md_do_align if it is
3492 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
3493 (md_do_align): Define.
3494 * config/tc-m88k.c: Include "subsegs.h".
3495 (m88k_do_align): New function.
3497 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
3498 (COFF_NOLOAD_PROBLEM): Define.
3499 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
3503 * config/tc-i386.c (md_assemble): For a jump instruction with
3504 non-constant target, require 7 available bytes in the current
3509 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
3514 * configure.in (alpha-*-linux*): Use ecoff.
3515 * configure: Rebuild.
3516 * ecoff.c (ecoff_directive_extern): New function.
3517 (ecoff_directive_weakext): New function.
3518 (ecoff_build_symbols): Handle weak symbols.
3519 (ecoff_setup_ext): Likewise.
3520 (ecoff_frob_symbol): Warn about weak common symbols.
3521 * ecoff.h (ecoff_directive_extern): Declare.
3522 (ecoff_directive_weakext): Declare.
3523 * symbols.c (S_IS_WEAK): New function.
3524 * symbols.h (S_IS_WEAK): Declare.
3525 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
3527 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
3532 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
3534 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
3535 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
3536 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
3537 * config/obj-coff.c (coff_frob_symbol): Don't call
3538 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
3539 one. If OBJ_XCOFF, try to figure out whether the symbol is going
3542 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
3543 (ppc_stab_symbol): New static variable.
3544 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
3545 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
3546 sy_tc.real_name to the stab string.
3547 (ppc_bc, ppc_ec): New static functions.
3548 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
3550 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
3552 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
3555 * input-scrub.c (input_scrub_push): Reset sb_index.
3559 * listing.c (listing_newline): Don't create a frag in the absolute
3564 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
3565 (ppc_csect): Move most of the code to ppc_change_csect, and call
3567 (ppc_change_csect): New static function, taken from ppc_csect.
3568 (ppc_section): New static function.
3569 (ppc_saw_abs): New static varable.
3570 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
3571 a symbol has no csect.
3572 (ppc_adjust_symtab): New function.
3573 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
3574 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
3576 * write.c (write_object_file): If tc_adjust_symtab is defined,
3577 call it just before the call to obj_adjust_symtab.
3579 * symbols.c (symbol_find_or_make): Change name to be const.
3580 * symbols.h (symbol_find_or_make): Update declaration.
3584 * doc/as.texinfo (Align): Mention SH.
3585 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
3587 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
3588 (Z8000 Directives, even): Don't give numeric align value, instead
3593 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
3598 * Makefile.in (diststuff): Also make info.
3599 (maintainer-clean realclean): Also make clean-info.
3603 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
3604 (pa_subspace): For ".nsubspa", always create a new subspace
3605 with the given attributes, even if one already exists with the
3610 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
3611 TC_FORCE_RELOCATION, taking an additional section argument. If
3612 the section of the target symbol is not the same as the current
3613 section, always force the relocation to be used.
3614 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
3616 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
3617 md_pcrel_from function, taking an additional section argument.
3618 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
3620 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
3621 older TC_FORCE_RELOCATION if not defined.
3622 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
3623 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
3624 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
3625 TC_FORCE_RELOCATION.
3629 * input-scrub.c (as_where): Set name to null pointer if we don't
3631 * messages.c (identify): Only print filename if non-null.
3632 (as_show_where): Ditto, for line number too.
3633 (as_warn_internal, as_bad_internal): Ditto.
3635 * input-file.c (input_file_open): If the input file can't be
3636 opened, consider it an error.
3640 * config/tc-mips.c: Added mips_4100 control, and support for
3641 accepting the 4100 as a MIPS architecture variant (md_begin,
3642 macro_build, mips_ip, md_parse_option). Adding suitable
3643 command-line OPTIONs, and updating the help text (md_show_usage).
3647 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
3648 BFD_ASSEMBLER is defined.
3650 * Use one active frag and one obstack per frag chain:
3651 * frags.c (frags): Variable deleted.
3652 (frag_alloc): New function.
3653 (frag_grow, frag_more, frag_variant, frag_now_fix,
3654 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
3656 (frag_new): Ditto. Verify that frch_last and frag_now match on
3657 entry and exit, and that old frag_now has non-zero type. Replace
3658 "know" uses with "assert". Use frag_alloc instead of mucking with
3660 * frags.h (frags): Declaration deleted.
3661 * subsegs.h (struct frchain): Add new field frch_frag_now.
3662 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
3664 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
3665 give it any stricter alignment than frchainS structures need. Do
3666 not initialize frags obstack. Set frag_now to point to
3667 dummy_obstack. Initialize absolute_frchain.
3668 (subseg_set_rest): Save and restore frag_now in frch_frag_now
3669 field of frchainS. Don't create new frags on section switch, and
3670 use frag_alloc when creating a new frag chain. For absolute
3671 section, set frchain_now to absolute_frchain. Verify that
3672 frch_last and frag_now match on entry and exit. Initialize
3673 per-chain obstack, and under gcc, set required alignment to that
3674 needed by fragS structure.
3676 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
3678 * stabs.c (get_stab_string_offset): Only copy input string if a
3679 fresh copy is needed, not if the section already exists.
3680 (s_stab_generic): Cache stab section name to bypass lookups, since
3681 usually it will match. Could be made faster still by changing the
3682 memory allocation rules.
3683 (s_xstab): Cache section name to bypass repeated string
3686 * frags.c (frag_new): Deleted register declarations.
3688 * listing.c (frag_now): Don't declare.
3690 * as.c (chunksize): New variable.
3691 (debug_memory): New variable.
3692 (main): If debug_memory is set, reduce chunksize and
3694 * as.h (chunksize): Declare it.
3695 * read.c (read_begin): Use it.
3697 * config/tc-alpha.c (md_shortopts): Include 'g'.
3698 (md_parse_option): Ignore it.
3700 * Makefile.in (distclean): Remove Makefile and config.status from
3701 testsuite directory.
3702 (clean-here): Don't delete testsuite. Instead, delete only the
3703 files within it that would be generated by running tests.
3705 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
3706 frag_now_fix instead of accessing obstack info directly.
3707 * config/tc-arm.c (s_ltorg): Ditto.
3708 (md_assemble): Ditto.
3710 * config/tc-i386.c (md_assemble): Call frag_grow instead of
3715 * stabs.c (aout_process_stab): Insert debug symbol into symbol
3716 chain after parsing value expression, if any, to avoid separating
3717 continued .stabs lines.
3721 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
3722 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
3726 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
3727 -mcom support and PPC_OPCODE_ANY for -many.
3728 (md_parse_option): Ditto.
3730 (md_begin): For duplicate instructions, print all duplicates
3735 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
3736 common mode operation.
3737 (md_show_usage): Add -mcom to usage message.
3741 * expr.c (op_rank): Add O_symbol_rva.
3742 * expr.h (operatorT): Add O_symbol_rva.
3743 * read.c (cons_worker): Set O_symbol_rva when necessary.
3744 * write.c (fix_new_exp): Understand O_symbol_rva.
3748 * config/tc-mips.c: Correct s_cons arguments. From Michael
3753 * config/tc-ppc.c (ppc_macro): Make count unsigned.
3754 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
3755 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
3757 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
3758 make non-static. Change all users.
3759 * config/obj-coff.h (coff_n_line_nos): Declare.
3765 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
3767 * config/obj-elf.c (obj_elf_common): Convert specified byte
3768 alignment to power of two. Set size of local bss symbol.
3770 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
3774 * sb.c, macro.c: Decide whether to include <string.h> or
3775 <strings.h> just as as.h does.
3779 * Makefile.in (site.exp): Fix setting of $srcdir.
3781 * config/tc-arm.c (md_atof): Fix little-endian output.
3782 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
3787 * doc/as.texinfo: Split out the NS32k family documentation,
3788 despite its being commented out for now.
3789 * doc/c-ns32k.texi: New file.
3791 * sb.c, macro.c: Include string.h.
3793 * Makefile.in (comparison): Only check *.o; we don't care if
3794 timestamps inserted by the native linker differ.
3796 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
3797 if alignment stricter than 4 bytes is requested; in that case,
3798 align to a 4-byte boundary first.
3802 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
3803 uninitialized" warning for `Max_Source_Offset'.
3807 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
3808 qualifiers and detect errors if present.
3809 (md_assemble): Add call to parse_toc_entry. Also added some support
3810 for the [tocv] qualifier.
3811 (ppc_pe_tocd): New function to support data in the toc section.
3815 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
3820 * config/tc-m68k.c (isword): Accept all values from -65536 to
3821 +65535, so ~VAL will not be rejected.
3823 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
3824 skip characters after the pseudo-op.
3826 * read.c (get_line_sb): Don't look past buffer_limit.
3827 (s_include): In MRI mode, skip characters after the file name.
3831 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
3832 periods in register names.
3834 For older changes see ChangeLog.1.