4 * read.c (read_a_source_file): Check for conditional operators
5 before doing an MRI pending alignment.
6 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
7 (tc_conditional_pseudop): Define.
8 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
9 * doc/internals.texi (CPU backend): Describe
10 tc_conditional_pseudoop.
13 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
15 (md_begin): Check m68k_rel32_from_cmdline before setting
17 (m68k_mri_mode_change): Likewise.
18 (md_longopts): Add --disp-size-default-16 and
19 --disp-size-default-32.
20 (md_parse_option): Handle new options.
21 (md_show_usage): Mention new options.
22 * doc/c-m68k.texi (M68K-Opts): Document new options.
25 * config/tc-m68k.c (m68k_index_width_default): New static
27 (m68k_ip): Use m68k_index_width_default to set the size of a base
28 register whose size was not given.
29 (md_longopts): Add --base-size-default-16 and
30 --base-size-default-32.
31 (md_parse_option): Handle new options.
32 (md_show_usage): Mention new options.
33 * doc/c-m68k.texi (M68K-Opts): Document new options.
35 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
37 * configure.in: Recognize mips-*-linux* target.
40 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
41 work if valueT is only 32 bits.
43 * config/tc-mips.c: Throughout, check target_big_endian rather
46 (mips_init_after_args): Remove.
47 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
48 (mips_init_after_args): Don't declare.
49 (tc_init_after_args): Don't define.
51 * config/tc-mips.h (tc_frob_after_relocs): Define if
53 (mips_elf_final_processing): Likewise.
54 (ELF_TC_SPECIAL_SECTIONS): Likewise.
59 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
60 sld.{h,w} and the operand is relocatable, adjust the adend by
61 shifting it right one bit.
66 * read.c (read_a_source_file): Check mri_pending_align after
67 checking for a macro. From Ronald F. Guilmette
70 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
72 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
73 add segment argument. If OBJ_ELF, treat a relocation against a
74 symbol in a linkonce section like a relocation against an external
76 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
80 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
84 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
85 includes config.h instead of host.h, tc.h instead of tp.h, and
86 targ-env.h instead of target-environment.h.
87 Also, obj-format.h includes targ-cpu.h instead of
90 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
91 * configure: Rebuild with autoconf.
92 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
93 (TARGET_FORMAT): Define to "coff-tic80".
94 * config/tc-tic80.c: New file for TIc80 support.
95 * config/tc-tic80.h: New file for TIc80 support.
100 * doc/as.texinfo (M): Mention explicitly that -M changes macro
105 * write.c (adjust_reloc_syms): If the fixup symbol has been
106 equated to an undefined symbol, convert the fixup to being against
107 the target symbol. Remove obsolete code handling a special case
112 * mpw-make.sed: Use NewFolderRecursive for installation.
117 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
121 * config/tc-d10v.c (md_assemble): Fix bug which caused
122 second instruction in a line to be case sensitize. PR11312
127 * config/tc-mn10200.c (mn10200_insert_operand): Don't
128 range check operands with MN10200_OPERAND_NOCHECK set.
129 (check_operand): Likewise.
133 * config/tc-mips.c: Undo part of last Friday's alignment changes.
134 (md_begin): Always align the text section to a four byte
136 (append_insn): Remove call to record_align.
138 * config/tc-mips.c (insn_label): Remove.
139 (struct insn_label_list): Define.
140 (insn_labels, free_insn_labels): New static variables.
141 (mips_clear_insn_labels): New static function.
142 (append_insn): Mark all mips16 text labels, and make them odd.
143 Handle all labels after emitting a nop, not just one. Call
144 mips_clear_insn_labels rather than just clearing insn_label.
145 (mips_emit_delays): Add insns parameter, and use it to decide
146 whether to mark mips16 labels. Handle all labels, not just one.
147 Force mips16 labels to be odd. Change all callers.
148 (mips16_immed): Don't check for an odd branch target.
149 (md_apply_fix): Don't check mips16 mode for a branch reloc.
150 (mips16_extended_frag): Ignore the low bit in a branch target.
151 (md_convert_frag): Likewise.
152 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
154 (mips_align, mips_flush_pending_output, s_cons): Likewise.
155 (s_float_cons, s_gpword): Likewise.
156 (s_align): Use insn_labels rather than insn_label.
157 (s_cons, s_float_cons, s_gpword): Likewise.
158 (mips_frob_file_after_relocs): New function.
159 (mips_define_label): Rewrite to add to insn_labels list.
160 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
161 * ecoff.c (ecoff_build_symbols): If the size of a function comes
162 out odd, increment it.
164 * config/tc-mips.c (append_insn): Only update prev_insn when not
165 reordering if place is NULL.
167 * config/tc-mips.c (mips16_ip): Check for a missing expression
168 when using the register indirect addressing mode.
172 * config/tc-mn10200.c (mn10200_insert_operand): Don't
173 check 24bit operands for overflow.
174 (check_operand): Likewise.
178 * doc/as.texinfo (Section): Document how to use the .section
179 pseudo-op for COFF and ELF.
183 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
187 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
188 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
189 store them. Adjust other RELAX_MIPS16 macros.
190 (RELAX_MIPS16_DSLOT): Define.
191 (RELAX_MIPS16_JAL_DSLOT): Define.
192 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
193 handling of whether previous instruction has a fixup. Set
194 prev_insn_reloc_type.
195 (mips_no_prev_insn): Clear prev_insn_reloc_type.
196 (mips16_extended_frag): Use the right base address for a PC
197 relative add or load.
198 (md_convert_frag): Likewise. If a PC relative add or load is
199 used, record the alignment for the section.
203 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
204 linkonce section into a reloc against the section symbol.
206 * config/tc-mips.c (mips16_macro): Remove nop instructions after
209 * config/tc-mips.c (md_begin): If configured for an embedded ELF
210 system, don't set the section alignment to 2**4.
211 (s_change_sec): Likewise.
212 (append_insn): Call record_alignment for the section.
213 (md_section_align): Don't align the section size for an embedded
218 * write.c (adjust_reloc_syms): Make sure that symbols are
219 resolved; expression symbols may have been skipped.
220 * config/obj-coff.c (fixup_segment): Likewise.
224 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
225 BFD_RELOC_24_PLT_PCREL relocation.
226 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
231 * config/tc-d10v.c (write_2_short): Remove code that called
232 parallel_ok() when the programmer specified parallel instructions.
237 * config/tc-mn10300.c (md_assemble): Update to handle endianness
240 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
241 * config/tc-mn10300.c (md_assemble): Likewise.
245 * config/tc-mips.c (append_insn): Make sure there is enough room
246 in a frag after a mips16 instruction to switch it with a jump
249 * config/tc-mips.c (mips16_extended_frag): Give an error for an
250 attempt to use a non absolute symbol in an extending frag.
254 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
255 * config/tc-mn10200.h: Likewise.
259 * app.c (do_scrub_chars): At the end of a C comment, pass space to
260 UNGET rather than PUT. Set old_state before setting state to -2.
262 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
263 when extending because the value is exactly maxtiny + 1.
265 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
266 arguments, and store them. Adjust other RELAX_MIPS16 macros.
267 (RELAX_MIPS16_USER_SMALL): Define.
268 (RELAX_MIPS16_USER_EXT): Define.
269 (mips16_small, mips16_ext): New static variables.
270 (append_insn): Pass mips16_small and mips16_ext to
272 (mips16_ip): Set mips16_small and mips16_ext.
273 (mips16_immed): Don't check mips16_autoextend.
274 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
276 * write.c (write_relocs): Print an error for an out of range
277 fixup, rather than calling abort.
279 * as.c (main): Unlink the output file if there are errors while
280 generating the fixups.
284 * config/tc-mips.c (mips16_extended_frag): Don't call
286 (md_convert_frag): Call resolve_symbol_value before calling
287 S_GET_VALUE, and don't add in the frag address.
289 * config/tc-mips.c (mips16_immed): Add file and line parameters,
290 and use them when reporting errors. Change all callers.
294 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
295 Remove '$' prefixing for registers.
300 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
301 is initialized before using it.
302 (d10v_cleanup): No longer uses its argument, so make it void.
304 * config/tc-d10v.h (d10v_cleanup): Change prototype.
309 * write.c (fixup_segment): Don't discard the symbol for a PC
310 relative fixup to an absolute symbol.
315 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
316 with multiple sections.
321 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
322 no-mips-16 to no-mips16.
323 (s_mipsset): Accept .set mips16 and .set nomips16.
327 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
328 argument, and check for +/- constant following the suffix, folding
329 it into the expression.
330 (ppc_elf_cons): Change ppc_elf_suffix calls.
331 (md_assemble): Ditto.
332 (shlib): Replace boolean mrelocatable with enumeration shlib.
333 (md_parse_option): Discriminate between PIC style shared libraries
335 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
340 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
341 change the comment characters.
343 * config/tc-ppc.c (comment_chars): Delete in favor of
345 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
347 (ppc_comment_chars): Select appropriate comment chars by default.
348 (msolaris): New flag for -m{,no-}solaris.
349 (md_parse_option): Recognize -K pic. Add support for
351 (md_show_usage): Update.
352 (md_begin): Do not set ELF flags if Solaris.
353 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
354 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
358 * as.c (main): Correct handling of flag_always_generate_output.
362 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
363 fx_offset, not fx_addnumber.
365 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
370 * config/tc-i386.c: Correct misspelling: balenced to balanced.
374 * config/tc-mips.c (md_section_align): Check for an alignment of
375 4, not an alignment of 16. Corrects August 7 patch.
379 * configure, conf.in: Rebuild with autoconf 2.12.
381 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
382 alignment is specified.
384 Add support for mips16 (16 bit MIPS implementation):
385 * config/tc-mips.c: Extensive additions for mips16 support, not
387 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
388 (insn_uses_reg): Change last parameter to an enum.
389 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
390 (md_relax_frag): Define as mips_relax_frag.
391 (mips_relax_frag): Declare.
392 (struct mips_cl_insn): Add use_extend and extend fields.
393 (tc_fix_adjustable): Define.
394 * config/obj-elf.h (S_GET_OTHER): Define.
395 (S_SET_OTHER): Define.
399 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
400 o and p operand specifiers.
404 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
405 (relax_segment, case rs_org): Move code inside braces. Move locals
406 target,after inside too.
407 (relax_segment, case rs_machine_dependent): Guts moved to ...
408 (relax_frag): New function.
409 Call md_prepare_relax_scan if defined.
410 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
415 * config/tc-mn10300.c (address_registers): Use '$' as register
416 prefix instead of '%'.
417 (data_registers, other_registers, md_assemble): Likewise.
419 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
420 (data_registers, other_registers, md_assemble): Likewise.
422 * config/tc-mn10300.c (md_assemble): Correctly determine the
423 correct location and type for each relocation.
424 (md_pcrel_from): Simplify.
428 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
429 converted into branch around branch.
433 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
434 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
436 * config/tc-mn10300.c (md_begin): Set linkrelax.
437 (md_assemble): Create fixups as needed.
438 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
443 * config/tc-d10v.c (parallel_ok): When automatically converting
444 serial ops to parallel, do not consider a branch as the first
450 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
454 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
456 (check_operand): Likewise.
461 * config/tc-d10v.c (parallel_ok): Branch and link instructions
463 (write_2_short): Call parallel_ok to check whether two short
464 instructions the user requested execute in parallel, can be
469 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
470 allow a branch and link in parallel with an exe instruction.
474 * doc/c-d10v.texi: Add info on @word modifier.
479 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
480 operands are assumed to be 32bits. Use "bits" field to hold the
481 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
482 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
485 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
486 of a MN10300_OPERAND_SPLIT operand by operand->shift.
488 * config/tc-mn10300.c (mn10300_insert_operand): Handle
489 MN10300_OPERAND_SPLIT.
493 * config/tc-mn10300.c (md_assemble): Insert operands into
494 the extension part of the instruction if necessary.
495 (mn10300_insert_operand): Accept pointer to extension word
496 argument. Make insn a pointer argument too. Return type
497 is now void. All callers changed.
501 * config/tc-mn10300.c (mn10300_insert_operand): Handle
502 repeated register operands.
506 * doc/as.texinfo: Added section on reporting bugs.
508 * config/tc-alpha.c: Change uses of void * to PTR. Change the
509 alpha_macro emit field to expect a const argument, and change the
510 arg field to be const. Fix some spacing to follow the GNU
515 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
516 (pca56) and 21264 (ev6) cpus.
517 (md_apply_fix): Private relocation types are now negative.
518 (alpha_force_relocation): Likewise.
519 (tc_gen_reloc): Likewise.
520 (emit_insn): Likewise.
521 (emit_ldXu): Do the right thing when the hardware can do byte insns.
522 (emit_stX): Likewise.
523 (emit_sextX): Likewise.
527 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
528 a symbol found in a reloc.
530 * symbols.c (resolve_symbol_value): Improve the error message if
531 an undefined symbol is used in an expression.
535 * doc/internals.texi: Rewrite, and add a lot of documentation.
536 * doc/Makefile.in (internals.info): New target.
541 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
542 against weak symbols.
546 * config/tc-v850.c (md_assemble): Don't lose for relaxable
549 * config/tc-v850.c (md_convert_frag): Make sure we insert the
550 fixup at the right address within the frag.
552 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
553 to an absolute value, instead increment it as needed.
555 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
556 * config/tc-v850.c: Fix some indention problems.
557 (md_relax_table): Define for D9->D99 branch displacement
559 (md_convert_frag): Do something useful instead of aborting.
560 (md_estimate_size_before_relax): Likewise.
561 (md_assemble): Note if the matching instruction has a relaxable
562 operand. If it does, allocate frag with frag_var and don't
569 * config/tc-d10v.h (md_cleanup): New function. This is needed to
570 write out any buffered instructions when a ".end" is found.
575 * read.c (read_a_source_file): New hook md_cleanup().
579 * write.c (fix_new_exp): Use make_expr_symbol to build an
580 expression symbol for a complex fixup.
585 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
586 hacks to improve parsing of complex hi, lo, zda, etc
588 (md_assemble): Don't demand and eat a trailing ')' after finding
589 a v850 relocation prefix. Sign extend the constant in a
590 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
592 (parse_cons_expression_v850): Don't eat a trailing ')' after
593 finding a v850 relocation prefix.
595 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
596 (TC_CONS_FIX_NEW): Likewise.
597 * config/tc-v850.c (parse_cons_expression_v850): New function.
598 (cons_fix_new_v850): Likewise.
600 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
605 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
606 for an unknown relocation type.
611 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
616 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
617 differently for movea & sst/sld insns.
623 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
629 * obj.h (struct format_ops): Add frob_file_after_relocs field.
630 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
631 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
632 frob_file_after_relocs field.
633 * config/obj-elf.c (elf_format_ops): Likewise.
634 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
639 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
640 specially if not ELF.
641 (md_apply_fix): Handle BFD_RELOC_64.
642 (tc_gen_reloc): Handle BFD_RELOC_64.
644 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
645 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
648 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
649 Don't check S_GET_OTHER.
653 * config/tc-mips.c (mips_ip): Accept an odd floating point
654 register with l.s or s.s.
656 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
658 (obj_aout_type): New static function.
662 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
667 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
668 and sdaoff expressions.
670 * write.c (fixup_segment): Don't add symbol value to addend if
672 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
675 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
676 (md_pcrel_from_section): Likewise.
677 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
681 * config/tc-v850.c (md_pcrel_from): Delete unused function.
682 (md_pcrel_from_section): New function.
683 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
688 * config/tc-mips.c (load_register): Add cast to offsetT when using
693 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
694 (elf_frob_file_after_relocs): ... here. New function.
695 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
696 * write.c (write_object_file): Call *frob_after_relocs after the
697 call to write_relocs.
699 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
701 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
702 symbol, as this messes up linking. Instead, expand the recursive call
703 inline and change up the appropriate bits to get the 0x8000 offset
708 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
709 two symbols in the same segment to be adjusted.
711 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
712 * configure: Rebuild.
716 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
718 (ppc_elf_validate_fix): Ditto.
719 (md_assemble): Ditto.
721 (ppc_pe_section): Ditto.
722 (ppc_frob_symbol): Ditto.
726 * config/mn10300.c (md_assemble): Pass an extra shift count
727 to mn10300_insert_operand based on the opcode format.
728 (mn10300_insert_operand): Accept and use extra shift count
731 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
732 formats rather than hard-coded constants.
734 * config/tc-mn10300.c (md_assemble): Format D5 instructions
735 are 7 bytes long. Write out instructions in big-endian format.
739 * config/tc-mn10300.c (md_assemble): Tweak further so
740 that all instructions are parsed correctly.
744 * as.h: Include libiberty.h.
745 (xmalloc, xrealloc): Don't declare.
746 * as.c: Don't include libiberty.h.
747 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
748 * config/tc-mips.c: Likewise.
749 * messages.c: Likewise.
750 (xstrerror): Don't declare.
755 * config/tc-mn10300.h (pre_defined_registers) Remove.
756 (system_registers, cc_names): Likewise.
757 (address_registers, data_registers, other_registers): New register
759 (register_name, system_register_name, cc_name): Remove.
760 (mn10300_reloc_prefix): Likewise.
761 (data_register_name): New function.
762 (address_register_name, other_register_name): Likewise.
763 (md_assemble): Rough cut at parsing operands. Remove lots of
765 (md_apply_fix3): Disable for now.
769 * config/tc-m68k.c (select_control_regs): New function, extracted
770 out of m68k_init_after_args.
771 (m68k_init_after_args): Use it.
772 (mri_chip): Use it here as well to update set of allowed control
777 * config/obj-elf.c (elf_begin): New function.
778 (obj_elf_section): Add the section symbol to the symbol table.
779 * config/obj-elf.h (obj_begin): Define.
780 (elf_begin): Declare.
781 * as.c (perform_an_assembly_pass): Call obj_begin if it is
786 * config/obj-coff.c (fixup_segment): Subtract the section address
787 from a PC relative reloc if TC_M68K.
791 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
792 .uaxword available even if not OBJ_ELF.
793 (md_atof): Remove unused local variable wordP.
797 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
798 for Matsushita MN10x00 support.
799 * configure.in: Recognize mn10x00-*-*
800 * configure: Rebuilt.
804 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
805 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
806 (load_expression): track clobbering of base reg before jmp/jsr.
807 (s_alpha_file): pass case_hack flags and source filename via
809 * tc-alpha.h (TC_CONS_FIX_NEW): define
813 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
814 * configure: Rebuild.
818 * read.c (s_macro): Warn if a macro has the same name as a
820 (s_space): In m68k MRI mode, align to a word boundary.
821 * macro.c (define_macro): Add namep parameter. Change all
823 * macro.h (define_macro): Update declaration.
825 * as.c (show_usage): Print bug report address.
826 (parse_args): Change version printing to match current GNU
828 * gasp.c (show_usage): Print bug report address.
829 (main): Change version printing to match current GNU standards.
831 * config/tc-m68k.c (init_table): Correct access control unit
834 * config/tc-alpha.c: Add some static function prototypes.
835 (alpha_macros): Move to top of file. Make static.
836 (alpha_num_macros): Move to top of file.
841 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
847 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
848 and just call sprintf_vma.
852 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
856 * read.c (float_cons): Call md_flush_pending_output if it is
862 * config/tc-d10v.c (md_operand): Created. Allows operands to
864 * config/tc-d10v.h (md_operand): Undefined.
869 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
870 (md_assemble): A fixup width of '3' means a 1 byte reloc.
874 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
875 reloc for the i960 for a reloc in the same section. This undoes
876 one of the two changes made Aug 19.
880 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
881 symbols to the position of the debugging information.
885 * expr.c (expr): Always use unsigned right shifts for >>.
889 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
890 slots when processing BL fixups.
892 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
893 on the first half of the instruction.
897 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
898 stabs expression, rather than giving an error.
900 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
905 * config/tc-mips.c (append_insn): Fill in the value for a constant
906 jump, rather than creating a reloc.
910 * config/tc-mips.c (append_insn): Don't swap an instruction which
911 sets a condition code with an instruction which uses a condition
913 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
916 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
917 (mips_ip): Give a better error message if the ISA level is wrong.
918 (md_parse_option): Recognize -mcpu=[v][r]5000.
922 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
924 (load_register): Ensure hi32 bits are not lost during lo32bit
925 processing. Fix shift offset that was overflowing into the next
926 instruction field. Add code to generate shorter sequences for
927 constants with a single contiguous seqeuence of ones.
932 * config/tc-d10v.c (d10v_dot_word): New function to support
933 "@word" with the word pseudo-op.
934 (md_apply_fix3): Cleanup and changes to support correct sizes
935 for 16 and 18-bit relocs.
940 * configure.in (sparc-*-aout): Set `em'.
941 * configure: Regenerated.
942 * config/te-sparcaout.h: New file.
943 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
944 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
945 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
946 (SPECIAL_CASE_{SETSW,SETX}): Define.
947 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
948 (md_begin): Delete setting of `target_big_endian'.
949 (output_insn): New function.
950 (md_assemble): Rewrite. Add `setx' support.
951 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
953 (md_atof): Add little endian support.
954 (md_number_to_chars): Likewise.
955 (md_apply_fix): Likewise.
956 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
957 (md_parse_option): Likewise.
958 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
962 * ecoff.c (ecoff_new_file): New function.
963 * ecoff.h (ecoff_new_file): Declare.
964 * config/obj-ecoff.h (obj_app_file): Define.
968 * config/tc-alpha.c (load_expression): Bias the .lit8 section
969 symbol by 32k so that our 16-bit signed offset can address the
974 * config/tc-mips.c (load_register): Remove unused variable tmp.
978 * config/tc-mips.c (load_register): Remove unnecessary code that
979 was causing the high 32bits of 64bit constants to be lost.
984 * config/tc-d10v.c: Added changes to support function
985 pointers and "@word" syntax.
991 * config/tc-v850.c: Remove commented out and #if 0'd code.
992 (v850_reloc_prefix): Provide prototype.
993 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
995 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
996 (md_assemble): Add default to case statement.
1000 * config/tc-v850.c (md_assemble): Compute size of the instrction
1003 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
1008 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
1009 routines to fetch/store the updated instruction from/to memory.
1010 (v850_insert_operand): If the operand has a specialized insert
1015 * config/tc-v850.c (reg_name_search): Align calling convention to
1016 be like identical function found in tc-ppc.c.
1017 (get_reloc): Removed.
1018 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
1019 (md_assemble): emit fixups.
1020 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
1022 (md_apply_fix3): handle fixups/relocs.
1023 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
1029 * configure.in (sh-*-elf*): New target.
1030 * config/tc-sh.h (TARGET_ARCH): Define.
1031 (WORKING_DOT_WORD): Define.
1032 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
1033 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
1034 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
1035 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
1036 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
1037 (SUB_SEGMENT_ALIGN): Likewise.
1038 (RELOC_32): Don't define.
1039 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
1040 (target_big_endian): Declare if OBJ_ELF.
1041 (TARGET_FORMAT): Define if OBJ_ELF.
1042 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
1044 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
1045 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
1046 (struct sh_count_relocs): Define.
1047 (sh_count_relocs): New static function, broken out of
1048 sh_frob_file. Add BFD_ASSEMBLER code.
1049 (sh_frob_section): Likewise.
1050 (sh_frob_file): Call sh_frob_section.
1051 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
1052 call section_symbol rather than seg_info (seg)->dot.
1053 (md_section_align): Add OBJ_ELF version.
1054 (SWITCH_TABLE_CONS): Define.
1055 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
1056 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
1057 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
1059 (struct reloc_map): Define if not BFD_ASSEMBLER.
1060 (coff_reloc_map): Likewise.
1061 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
1062 (tc_gen_reloc): New function if BFD_ASSEMBLER.
1063 * write.c (write_relocs): Ifdef out fx_where test which triggers
1064 inappropriately for SH ELF.
1065 (write_object_file): Call tc_frob_file_before_adjust and
1066 obj_frob_file_before_adjust if they are defined.
1068 * write.c (write_object_file): Use BFD_RELOC_16, not
1069 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
1071 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
1077 * config/tc-d10v.c (find_opcode): Fix problem with calculating
1078 branch sizes in across sections.
1084 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
1089 * config/tc-v850.c (md_begin): Deal with end of opcode
1096 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
1097 the wrong opcode for cases like st2w where there are many forms
1098 of the same instruction.
1103 * expr.c (operand): If md_parse_name is defined, call it before
1104 calling symbol_find_or_make.
1105 * config/tc-ppc.h (md_parse_name): Define.
1106 (ppc_parse_name): Declare.
1107 * config/tc-ppc.c (reg_name_search): Add regs and regcount
1109 (register_name): Update call to reg_name_search.
1110 (cr_operand): New static variable.
1111 (cr_names): New static const array.
1112 (ppc_parse_name): New function.
1113 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
1114 cr_operand before calling expression.
1118 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
1119 hppa_gen_reloc_type call.
1124 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
1125 on addresses. Improved error messages.
1126 * doc/c-d10v.texi: Added docs for register pairs.
1130 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
1136 * ecoff.c (init_file): Initialize fMerge to 1.
1137 (add_file): Restore old file merging code, but only merge files if
1139 (ecoff_directive_loc): Clear fMerge field of current file.
1140 (ecoff_generate_asm_lineno): Likewise.
1145 * doc/c-d10v.texi: Fix typo.
1151 * config/tc-v850.c (md_assemble): Correct bit masking for
1152 hi and lo expressions.
1154 * config/tc-v850.c (md_assemble): Rough cut at demanding
1155 "ep" or "r30" in sst and sld instructions.
1156 (md_apply_fix3): Don't abort. Just warn that we don't
1159 * config/tc-v850.c (CC_NAME_CNT): Define.
1160 (cc_name): New function.
1161 (md_assemble): Handle V850_OPERAND_CC correctly.
1163 * config/tc-v850.c (md_assemble): Don't forget to initialize
1166 * config/tc-v850.c (reg_name_search): Generalize to search
1167 any given register table.
1168 (register_name): Pass appropriate table and size to reg_name_search.
1169 (system_register_name): New function.
1170 (SYSREG_NAME_CNT): Define.
1171 (md_assemble): Handle operands which are system registers.
1173 * config/tc-v850.c (md_assemble): If we find a register, but the
1174 opcode doesn't want a register, then we don't have a match.
1175 (md_assemble): Get size of the instruction from the opcode table.
1180 * configure.in: Set and substitute HLDENV.
1181 * configure: Rebuild.
1182 * Makefile.in (HLDENV): New variable.
1183 (as.new): Use $(HLDENV).
1185 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
1186 if an array dimension is not known.
1191 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
1192 * doc/c-d10v.texi: Cleanup.
1198 * config/tc-v850.c: New file.
1199 * config/tc-v850.h: New file.
1200 * configure (v850-*-elf): New target.
1201 * configure.in (v850-*-elf): New target.
1207 * doc/c-d10v.texi: New file.
1208 * doc/all.texi: Added D10V stuff.
1209 * doc/as.texinfo: Added D10V stuff.
1213 * config/tc-d10v.c: All references to defined symbols should
1214 now use the optimal instruction. .float and .double now work.
1219 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
1220 section address for the i960 as is done for the i386.
1224 * mpw-config.in: Add wildcards for config matching, add mips-*-*
1225 case, forward-include bfd/elf-bfd.h.
1230 * config/tc-d10v.c: Add additional information to the opcode
1231 table to help determinine which instructions can be done
1237 * config/tc-arm.c: Major changes to add Thumb support, with lots
1239 Reverted to INSN_SIZE macro, rather than insn_size variable.
1240 (insns): Added ARM "bx" instruction support.
1241 (tinsns): Added Thumb instruction definition structure.
1242 (arm_tops_hsh): Added hash structure for Thumb opcodes.
1243 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
1244 (opcode_select,s_arm,s_thumb,s_code): Added.
1245 (decode_shift): Allow upper-case RRX.
1246 (do_ldst): Simpler halfword support.
1247 (do_ldmstm): Improved.
1248 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
1249 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
1250 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
1251 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
1252 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
1254 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
1255 (md_parse_option): Add support for -mthumb.
1256 (md_show_usage): Updated to reflect new command line option.
1257 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
1258 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
1259 fragment information to be held.
1263 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
1264 non-null, but is a constant.
1265 (fix_new_arm): Call make_expr_symbol to make the expression symbol
1266 so that error reporting will work correctly.
1270 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
1271 against weak symbols.
1275 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
1276 (ppc_force_relocation): Declare if OBJ_XCOFF.
1277 * config/tc-ppc.c (ppc_force_relocation): New function if
1282 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
1287 * config/tc-d10v.c: Fix problem with relocs.
1292 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
1293 with nops if not in data_section or bss_section.
1297 Add support for openVMS/Alpha.
1298 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
1299 * config/obj-evax.c: New file.
1300 * config/obj-evax.h: New file.
1301 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
1303 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
1304 defined. Add case for bfd_target_evax_flavour.
1305 * config/vms-a-conf.h: New file.
1306 * conf-a-gas.com: New file.
1307 * configure.in: Add target alpha-*-*vms*.
1308 * configure: Rebuild.
1309 * makefile.vms: New file.
1310 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
1311 Don't call ffs on openVMS/Alpha.
1315 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
1316 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
1318 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
1319 (NEED_DECLARATION_SBRK): New macro.
1320 * configure, conf.in: Rebuild.
1321 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
1322 (strstr): Declare if NEED_DECLARATION_STRSTR.
1323 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
1327 * symbols.c (resolve_symbol_value): Handle addition or subtraction
1328 by a constant before entering the main switch. Reject attempts to
1329 apply an arithmetic function to non-absolute symbols, except for
1330 the special case of subtraction of two symbols in the same
1333 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
1334 not to more than a 16 byte boundary.
1336 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
1338 (tc_gen_reloc): If out of memory call as_fatal rather than
1339 assert. If no howto found, call as_bad_where rather than
1340 as_fatal. Change the error message slightly. Set howto to a
1341 non-NULL value in order to keep going.
1346 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
1351 * config/tc-h8300.c (get_specific): New operand "size" derived
1352 from ".b", ".w" and ".l" extensions. All callers changed. If
1353 the base instruction has no operands, then use the size to
1354 determine which specific instruction to use.
1358 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
1362 * config/tc-mips.c (md_section_align): Don't change addr if
1367 * config/tc-hppa.c: Revert yesterday's changes.
1372 * config/tc-d10v.c: Disable range checking on 16-bit values.
1377 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
1378 not just the ones that don't match.
1383 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
1388 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
1389 pre-cursor to adding Thumb support. Also added cpu_variant flag
1390 information to each of the asm_flg structures.
1391 (md_parse_option): Updated ARM7 parsing to allow 't' for
1392 thumb/halfword support, aswell as 'm' for long multiply.
1393 (md_show_usage): Updated help message.
1394 (md_assemble): Check that instruction flags are applicated to the
1395 current cpu variant.
1396 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
1397 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
1398 signextension instructions.
1399 (do_ldst): Generate halfword and signextension variants if
1400 mnemonic flags match.
1401 (ldst_extend): Do not allow shifts in the offset field of halfword
1402 or signextension instructions.
1403 (validate_offset_imm): Provide check on halfword and signextension
1405 (add_to_lit_pool): Merge identical literal pool values.
1409 * config/tc-hppa.c (selector_table): Add 'E' selector.
1410 (cons_fix_new_hppa): Don't coke on e_esel.
1411 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
1412 to help generate exception handling tables.
1413 (md_apply_fix): Don't try to apply fixups with an e_esel
1415 (hppa_fix_adjustable): Fixups with e_esel selectors
1420 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
1426 * config/tc-d10v.c: Added lots of error checking. Added hacks
1427 to support accumulator shifts.
1432 * symbols.c (S_SET_EXTERNAL): Let .weak override.
1433 (S_CLEAR_EXTERNAL): Likewise.
1434 (S_SET_WEAK): Remove error; just let .weak override.
1439 * config/tc-d10v.c (md_assemble): Now handles multiline
1444 * config/tc-d10v.c: Fix packaging bug. Added range checking.
1445 Added kludge for divs instruction. Fixed minor problem with
1446 multiple text sections.
1447 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1451 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
1452 addresses to be right-shifted by 2.
1456 * config/tc-d10v.c: Many changes to get relocs working.
1457 (register_name): No longer creates a symbol for register names.
1458 (pre_defined_registers): moved to opcodes/d10v-opc.c.
1459 (d10v_insert_operand): Now works correctly for either container.
1460 * config/tc-d10v.h (d10v_cleanup): Declare.
1465 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
1466 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
1470 * tc-alpha.c: Patches to track current minimum alignment to reduce
1471 the number of fragments created with frag_align.
1472 (alpha_current_align): New static variable.
1473 (s_alpha_text): Reset alignment to 0.
1474 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
1475 (s_alpha_stringer, s_alpha_space): New functions.
1476 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
1477 (alpha_cons_align): New function to replace both of them.
1478 (emit_insn): Only align if alpha_current_align is less than 2;
1479 reset alpha_current_align to 2.
1480 (s_alpha_gprel32): Likewise.
1481 (s_alpha_section): New function. Basically duplicate the other
1482 alpha section change hooks. Only define for ELF.
1483 (s_alpha_float_cons): Simplify alignment handling.
1484 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
1485 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
1486 Don't define the s_alpha_cons pseudo-ops. Do define
1487 s_alpha_stringer and s_alpha_space pseudo-ops.
1488 (alpha_align): Skip if less than current default alignment. Set
1490 * tc-alpha.h (md_flush_pending_output): Remove.
1491 (md_cons_align): Add.
1493 * tc-alpha.c: Add oodles of function description comments.
1494 (md_bignum_to_chars): Remove; there are no callers.
1495 (md_show_usage): Mention some more variants.
1500 * configure.in (sparc-*-sysv4*): New target.
1501 * configure: Rebuild.
1503 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
1504 uaxword to use s_uacons.
1505 (sparc_no_align_cons): New static variable.
1506 (s_uacons): New static function.
1507 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
1510 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
1512 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
1513 5 with regard to shared libraries.
1514 * configure: Rebuild.
1516 * config/tc-m68k.c (m68k_ip): Use the correct length when
1517 allocating space for the unsupported architecture error message.
1522 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
1526 * config/tc-d10v.c: New file.
1527 * config/tc-d10v.h: New file.
1528 * configure (d10v-*-elf): New target.
1529 * configure.in (d10v-*-elf): New target.
1534 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
1538 * config/tc-alpha.c (alpha_align): Change fill parameter
1539 to a pointer. Take NULL as 0 or nop depending on section. Change
1541 (s_alpha_align): Rename local variables.
1543 * doc/as.texinfo (.align): Document action of omitted
1548 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
1549 when an unsupported PC relative reloc is seen, rather than calling
1552 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
1553 Instead, check state when '#' comment is seen.
1557 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
1559 (tc_gen_reloc): If fixup was changed to be PC relative, change
1560 reloc type accordingly. Use name of reloc in error message.
1562 * as.h: Don't define const or volatile.
1563 * flonum.h: Don't define const.
1565 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
1566 if fx_pcrel is set. Correct setting the addend case in the
1567 OBJ_ELF case (from Andreas Schwab
1569 (md_show_usage): Correct -mfc5200 to -m5200.
1573 * doc/c-m68k.texi: Document -m5200 flag.
1574 * doc/as.texinfo: Likewise.
1576 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
1581 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
1582 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
1586 * Makefile.in (VERSION): Set to cygnus-2.7.1.
1588 * Released binutils 2.7.
1592 * config/tc-mips.c (mips_ip): Only perform range check when
1593 dealing with O_constant expressions.
1597 * m68k-parse.h (m68k_register): Add new coldfile control
1600 * config/tc-m68k.c (mcf5200_control_regs): New variable,
1601 array of control registers for the coldfire.
1602 (cpu_of_arch): Added mcf5200.
1603 (archs): Added mcf5200.
1604 (init_table): Add new control registers.
1605 (m68k_ip): Added support for new control registers.
1606 (m68k_init_after_args): Likewise.
1608 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
1612 * read.h (is_it_end_of_statement): Declare.
1613 * read.c (is_it_end_of_statement): Remove declaration.
1615 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
1617 (md_assemble): Fix handling of @l with an unsigned constant. Add
1618 default case to reloc switch.
1620 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
1623 * ecoff.c (last_lineno): New static variable.
1624 (add_procedure): Set last_lineno.
1625 (ecoff_directive_loc): Likewise.
1626 (ecoff_generate_asm_lineno): Likewise.
1627 (ecoff_fix_loc): New function.
1628 * ecoff.h (ecoff_fix_loc): Declare.
1629 * config/tc-mips.c (append_insn): When inserting nops, and using
1630 ECOFF debugging, call ecoff_fix_loc.
1634 * config/tc-h8300.c (build_bytes): If an operand type is
1635 marked as SRC_IN_DST retrieve it from the "destination" op.
1639 * configure.in (arm-*-riscix*): Set emulation to riscix.
1640 * configure: Rebuild.
1641 * config/te-riscix.h: New file to define TE_RISCIX.
1643 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
1647 * Makefile.in (config.status): Just run config.status as other
1652 * mpw-config.in (TARGET_OS): Add definition to conf.
1656 * config/tc-mips.c (append_insn): Parenthesize
1657 cop_interlocks expressions.
1661 * listing.c (listing_print): Close the listing file if it is not
1662 stdout. Close the other files opened for the listing.
1664 * config/tc-sparc.h (md_cons_align): Define.
1665 (sparc_cons_align): Declare.
1666 (HANDLE_ALIGN): Define.
1667 (sparc_handle_align): Declare.
1668 * config/tc-sparc.c (sparc_cons_align): New function.
1669 (sparc_handle_align): New function.
1670 * read.c (cons_worker): Call md_cons_align if it is defined.
1672 * as.h (struct frag): Add fr_file and fr_line fields.
1673 * frags.c (frag_new): Set fr_file and fr_line.
1674 (frag_var): Likewise.
1675 (frag_variant): Likewise.
1677 * as.h (struct frag): Remove unused align_mask and align_offset
1680 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
1685 * configure.in (mips-*-osf*): New target.
1686 * configure: Rebuild.
1688 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
1692 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
1693 between co-processor comparisons and branches for the VR4300.
1697 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
1698 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
1700 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
1701 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
1702 includedir): Use autoconf set values.
1707 * listing.c (listing_eject): Don't do anything if listing is 0.
1708 (listing_list): Likewise.
1709 (listing_source_line): Likewise.
1710 (listing_title): Don't save title if listing is 0.
1711 (listing_source_file): Check listing rather than listing_tail.
1713 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
1714 using shared libraries.
1715 * configure: Rebuild.
1719 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
1720 absolute expression if a relocation type was specified.
1724 * configure.in: Add support for *-*-rtems* configurations.
1725 * configure: Rebuild.
1729 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
1730 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
1731 opcodes library for alpha.
1732 * configure: Rebuild with autoconf 2.10.
1733 * config/tc-alpha.c: Substantial rewrite to add ELF support and
1734 use new opcode table.
1735 * config/tc-alpha.h (md_undefined_symbol): Don't define.
1736 (LOCAL_LABEL): Define differently if OBJ_ELF.
1737 (FAKE_LABEL_NAME): Define if OBJ_ELF.
1738 * config/alpha-opcode.h: Remove.
1739 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
1740 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
1741 include/opcode/alpha.h rather than config/alpha-opcode.h.
1745 * config/obj-aout.c (obj_emit_relocations): Give an error if the
1746 relocation symbol was not resolved.
1747 * config/obj-coff.c (do_relocs_for): Likewise.
1749 * write.c (adjust_reloc_syms): Refetch the symbol section after
1750 calling S_GET_VALUE, since it may have changed.
1752 * expr.c (struct expr_symbol_line): Define.
1753 (expr_symbol_lines): New static variable.
1754 (make_expr_symbol): Add entry to expr_symbol_lines.
1755 (expr_symbol_where): New function.
1756 * expr.h: Use extern on function declarations.
1757 (expr_symbol_where): Declare.
1758 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
1759 rather than printing the meaningless name of an expression
1764 * config/tc-i386.c (md_number_to_chars): Deleted.
1765 * config/tc-i386.h (md_number_to_chars): New macro.
1767 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
1769 (load_symbol_address, load_expression): Use build_mem.
1770 (build_operate): New function.
1771 (emit_addq_r): Use it.
1775 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
1779 * config/tc-vax.h (NOP_OPCODE): Define.
1783 * config/obj-vms.h (S_IS_COMMON): Define.
1784 (S_IS_LOCAL): Check for \002 as well as \001.
1785 (LONGWORD_ALIGNMENT): New macro.
1786 (SUB_SEGMENT_ALIGN): Use it.
1790 * config/vms-conf.h: Reconcile with conf.in.
1794 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
1797 * doc/as.texinfo: Document that any number of hex digits can
1800 * as.c (struct defsym_list): Define.
1801 (defsyms): New static variable.
1802 (parse_args): Just put --defsym arguments on defsyms list, rather
1804 (main): Define defsyms after output file is created.
1806 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
1809 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
1810 insn_error rather than calling check_absolute_expr.
1812 * as.c (emulation_name): Remove unused static variable.
1813 (default_emul_bfd_name): Add return NULL to avoid warning.
1814 * ecoff.c (ecoff_stab): Remove unused variables name and
1816 * frags.c (frag_new): Remove unused variable tmp.
1817 * hash.c (hash_grow): Parenthesize + within <<.
1818 (hash_print_statistics): Use %lu, not %d, to print unsigned
1820 * messages.c: Include "libiberty.h".
1821 (fprint_value): Add cast to avoid printf warning.
1822 (sprint_value): Likewise.
1823 * read.c: Include "ecoff.h".
1824 (emit_expr): Add casts to avoid printf warnings.
1825 * read.h: Use extern for function declarations.
1826 (pop_insert): Declare.
1827 * stabs.c: Include "ecoff.h".
1828 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
1829 former_last_fragP, and new_fragP.
1830 * subsegs.h (subsegs_print_statistics): Declare.
1831 * symbols.c (debug_verify_symchain): Change macro to discard
1833 * write.c (dump_section_relocs): Likewise.
1834 * write.h: Use extern for function declarations.
1835 (write_print_statistics): Declare.
1836 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
1838 * config/e-mipself.c (mipself_bfd_name): Likewise.
1839 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
1841 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
1842 emit relocations against external symbols.
1844 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
1845 message if bfd_reloc_type_lookup fails, rather than calling
1848 * config/tc-alpha.c (alpha_force_relocation): Add
1849 BFD_RELOC_12_PCREL to switch.
1853 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
1854 TE_PE (Lfoo, not .Lfoo).
1858 * read.c (s_fill): Don't warn about a zero repeat count.
1860 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
1861 coprocessor register.
1863 * config/tc-i386.c (md_assemble): When checking the size of a
1864 register to set the size of an instruction, do a bitwise and with
1865 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
1870 * config/tc-h8300.c (parse_reg): Tweak error messages.
1871 (build_bytes): Likewise.
1872 (skip_colonthing): Handle :32 suffix.
1873 (get_specific): Promote L_24 to L_32 if it makes a match.
1874 Don't always promote L_8 to L_16.
1875 (do_a_fix_imm): Clean up L_32 and L_24 handling.
1877 * config/tc-h8300.c (Smode): New variable.
1878 (h8300hmode): Turn off Hmode.
1879 (h8300smode): New function. Turn on Smode and Hmode.
1880 (md_pseudo_table): New ".h8300s" pseudo-op.
1881 (parse_reg): Handle "exr" register.
1882 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
1883 Handle "mach" and "machl" operands for ldmac.
1884 (get_specific): Handle "stm.l" and "ldm.l".
1885 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
1886 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
1891 * doc/as.texinfo: Reorder chapter of machine dependent options so
1892 that it is sorted by chip name.
1894 * doc/as.texinfo: Use consistant spelling of Vax.
1895 * doc/c-vax.texi: Likewise.
1899 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
1901 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
1902 (md_apply_fix): Likewise.
1903 (pa_try): New function.
1904 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
1908 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
1909 (pa_level): New function.
1913 * listing.c (listing_newline): Don't do anything if listing is 0.
1917 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
1918 symbol_create, not symbol_new.
1922 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
1923 absolute symbol unless TC_FORCE_RELOCATION returns true.
1925 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
1927 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
1928 the symbol to a pointer. Don't set previous_file_symbol.
1929 Simplify symbol list rearrangement.
1930 (coff_frob_symbol): Don't do anything with C_FILE symbols.
1931 (coff_adjust_symtab): Don't check previous_file_symbol.
1935 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
1937 (md_pseudo_table): Add ppc_elf_lcomm.
1941 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
1942 d0{24:1},d0' works without an immediate prefix on the bit numbers.
1943 (md_begin): Add digits to alt_notend_table.
1944 (md_parse_option): Make s a const pointer.
1946 * config/tc-sparc.c (md_pseudo_table): Add "empty".
1947 (s_empty): New static function.
1949 * config/obj-coff.c (struct filename_list): Only define if not
1951 (filename_list_head, filename_list_tail): Likewise.
1952 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
1953 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
1954 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
1955 have a section of N_ABS rather than N_DEBUG. If we do a merge,
1956 remove the new symbol from the list.
1957 (obj_coff_endef, both versions): Call tag_insert even if there is
1958 an old symbol with the same name, if the old symbol does not
1960 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
1961 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
1962 symbol that will be moved to the end of the symbol list.
1963 (coff_adjust_section_syms): Always call section_symbol for .text,
1965 (coff_frob_section): Likewise. Also, remove unused variable
1968 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
1969 manipulating frags directly.
1970 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
1973 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
1974 whether the next character is '*' before checking whether we are
1975 at the start of a line. Permit LINE_COMMENT_START to start a
1976 comment in state 1 (seen some whitespace) as well, to match the
1979 * gasp.c (do_align): Permit a fill value for .align.
1983 * read.c (next_char_of_string): Warn if a newline is seen in the
1984 middle of a string. Call bump_line_counters when appropriate.
1988 * symbols.c (colon): Use LOCAL_LABEL.
1992 * Makefile.in (install): Don't check to see if tooldir exists.
1993 Make $(tooldir) and $(tooldir)/bin.
1997 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
2002 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
2003 insn_error. Don't put the string "ERROR" in insn_error. Set
2004 insn_error rather than calling as_warn for an unsupported opcode.
2008 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
2009 before permitting -64.
2010 * output-file.c (output_file_create): Remove duplicate
2015 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
2016 section rather than a .reginfo section.
2017 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
2020 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
2021 bit value with the high bit set.
2025 * read.c (s_lcomm): Set section flags for .sbss section.
2027 * config/tc-mips.c (mips_64): New static variable.
2028 (mips_target_format): If mips_64, return elf64 targets rather than
2030 (md_longopts): Add "32" and "64".
2031 (md_parse_option): Handle -32 and -64.
2032 (md_show_usage): Mention -32 and -64.
2033 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
2038 * config/tc-ppc.c (comment_chars): Make '!' a comment character
2039 for Solaris compatibility.
2041 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
2042 .stabd with 4 arguments into a .stabn.
2046 * config/tc-mips.c (macro): When passing X_add_number to
2047 macro_build, cast it to int first.
2051 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
2052 created by emit_expr.
2054 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
2055 as_perror rather than assert.
2059 * config/tc-mips.c (mips_ip): Mark sections created to hold
2060 floating point information as read only.
2064 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
2068 * read.c (potable): Add .skip as a synonym for .space.
2070 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
2071 4 arguments, providing the 4th argument is 0, to allow
2072 compatibility with the Solaris assembler.
2076 * config/tc-sh.h (struct sh_segment_info_type): Define.
2077 (TC_SEGMENT_INFO_TYPE): Define.
2078 (sh_frob_label): Declare.
2079 (tc_frob_label): Define.
2080 (sh_flush_pending_output): Declare.
2081 (md_flush_pending_output): Define.
2082 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
2083 reloc before the instruction if necessary.
2084 (sh_frob_label): New function.
2085 (sh_flush_pending_output): New function.
2086 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
2087 when looking for the reloc for the target of .uses.
2088 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
2089 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
2091 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
2092 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
2093 the absolute symbol.
2095 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
2096 TC_SEGMENT_INFO_TYPE is defined.
2100 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
2101 matches the register size.
2105 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
2107 (do_relocs_for): Likewise.
2108 (fixup_segment): Don't just quit if linkrelax is set. Try to
2109 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
2114 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
2115 relative relocations with -mrelocatable. Also allow unfixed
2116 relocs in .ex_shared.
2120 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
2121 non-zero before assuming this is a long file name.
2122 (w_strings): Likewise.
2123 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
2125 * config/obj-coff.c (w_strings): Move declaration of i inside
2126 #ifdef block which uses it.
2130 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
2131 "relaxmode". Output relocs which identify various relaxing
2132 possibilities for mov.[bwl] instructions.
2133 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
2137 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
2138 (MD_APPLY_FIX3): Define.
2139 * config/tc-arm.c (my_get_expression): Only watch for bad segments
2141 (md_apply_fix3): Renamed from md_apply_fix.
2142 If pcrel reloc and symbol is in different section, undo effects
2147 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
2148 any reloc with an LR% or RR% field selector for SOM.
2152 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
2157 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
2159 (yank_symbols): Likewise.
2163 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
2164 BFD_ASSEMBLER, add name field.
2165 * config/obj-coff.c: Include "libiberty.h".
2166 (coff_header_append): Handle long section names.
2167 (crawl_symbols): Just use the name field for the symbol name,
2168 without worrying about null byte termination.
2169 (w_strings): Handle long section names.
2170 (write_object_file): Likewise. Also, use the name field, rather
2172 (obj_coff_add_segment): Permit long section names.
2173 (obj_coff_init_stab_section): Use the name field, rather than
2175 (adjust_stab_section): Likewise.
2176 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
2178 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
2179 (mem_fmt): Likewise.
2180 (md_apply_fix): Don't check fx_im_disp field.
2184 * configure.in: Add * after sparc*-*-vxworks.
2185 * configure: Rebuild.
2187 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
2188 define it to comment_chars. Use tc_comment_chars rather than
2190 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
2191 * config/tc-m68k.h (tc_comment_chars): Define.
2192 (m68k_comment_chars): Declare.
2193 * config/tc-m68k.c (m68k_comment_chars): Rename from
2194 comment_chars. Change into a pointer rather than an array.
2195 (md_longopts): Add "bitwise-or".
2196 (md_parse_option): Handle OPTION_BITWISE_OR.
2197 (md_show_usage): Mention --bitwise-or.
2198 * doc/c-m68k.texi: Document --bitwise-or.
2202 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
2205 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
2206 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
2207 correct handling of input line pointer, and call
2208 demand_empty_rest_of_line.
2212 * config/tc-sparc.c (in_bitfield_range): New static function.
2213 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
2214 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
2215 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
2216 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
2217 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
2221 * config/obj-coff.c: BFD_ASSEMBLER:
2222 (coff_last_bf): New static variable.
2223 (coff_frob_symbol): Set endndx of a .bf symbol.
2225 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
2226 (last_bfP): New static variable.
2227 (yank_symbols): Set endndx of a .bf symbol.
2231 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
2232 and -s options. Add -be for good measure.
2236 * read.c (s_space): Support non-constant fill value. Handle fill
2237 value correctly for a size other than 1.
2241 * config/tc-arm.c (my_get_float_expression): Update call to
2242 gen_to_words, X_PRECISION changed from 6 to 5.
2246 * config/tc-ppc.c (register_name,reg_name_search): Move register
2247 name lookup from PE specific code to all targets. Add support for
2248 -mregnames/-mno-regnames to control whether register names are
2250 (md_assemble): Call register_name for all platforms.
2251 (md_parse_option): Add support for -mregnames/-mno-regnames.
2253 * configure.in (powerpcle*-*-solaris): Add support.
2254 (powerpc*-*-linux): Ditto.
2255 * configure: Regenerate.
2257 * config/ppc-sol.mt: New config file for PowerPC Solaris.
2261 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
2262 be associated with a single %lo reloc.
2264 * config/tc-mips.c (load_address): Cast X_add_number to valueT
2265 before comparing against MAX_GPREL_OFFSET, so that negative
2266 numbers are handled correctly.
2271 * config/tc-sparc.c (last_insn): New static variable.
2272 (md_assemble): Warn about putting floating point branches in a
2273 delay slot. If architecture is less than v9, insert NOP
2274 instructions between floating point instructions and floating
2275 point branches. (The SunOS assembler does both these operations.)
2276 Save the last instruction opcode.
2277 (sparc_ip): Add pinsn parameter. Change caller.
2279 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
2280 for byte jump to next instruction to skip empty frags.
2284 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
2285 for addq with a 16 bit signed value, just emit a lda instruction
2290 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
2295 * configure.in: Permit --enable-shared to specify a list of
2297 * configure: Rebuild.
2301 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
2302 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
2306 * config/atof-ieee.c: Fix handling of denormalized extended
2307 precision numbers and overflow/underflow detection.
2308 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
2309 not include the 16 bit gap in the m68k extended precision format.
2313 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
2315 * configure: Rebuild.
2319 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
2320 be accessed using the $gp register.
2321 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
2325 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
2326 bfd_mach_sparc_sparclet if current_architecture is sparclet.
2330 * read.c (get_line_sb): Bump line counters based on
2331 input_line_pointer[-1], not *input_line_pointer. Don't bother to
2332 call LISTING_NEWLINE.
2333 (s_macro): Don't call demand_empty_rest_of_line.
2334 * app.c (do_scrub_chars): When handling C style comments, unget
2339 * read.h (enum linkonce_type): Define.
2340 (s_linkonce): Declare.
2341 * read.c (potable): Add "linkonce".
2342 (s_linkonce): New function.
2343 * subsegs.h (segment_info_type): Add linkonce field to
2344 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
2345 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
2346 (obj_coff_pe_handle_link_once): Declare if TE_PE.
2347 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
2349 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
2350 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
2351 entry based on the linkonce field in segment_info.
2352 * doc/as.texinfo: Document .linkonce.
2356 * doc/as.1: Changed to be recognized by catman -w on Solaris.
2360 * stabs.c (s_stab_generic): Call the listing functions before
2361 doing the rest of the processing, which may involve freeing the
2362 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
2363 SEPARATE_STAB_SECTIONS case.
2365 * config/tc-hppa.c: Remove nested comment.
2366 (tc_gen_reloc): Move label done inside the ifdef in which it is
2368 (md_apply_fix): Pass pointers to correct types to libhppa.h
2369 functions. Always return a value.
2371 * config/tc-mips.h (tc_frob_file): Define.
2372 (mips_frob_file): Declare.
2373 * config/tc-mips.c (struct mips_hi_fixup): Define.
2374 (mips_hi_fixup_list): New static variable.
2375 (imm_unmatched_hi): New static variable.
2376 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
2377 offset_reloc. Pass imm_unmatched_hi to append_insn.
2378 (append_insn): Add unmatched_hi parameter. If it is set, add the
2379 new fixup to mips_hi_fixup_list. Change all callers.
2380 (mips_ip): Set imm_unmatched_hi when appropriate.
2381 (mips_frob_file): New function.
2385 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
2386 * configure: Regenerated.
2390 * as.c (main): Call bfd_set_error_program_name.
2394 * as.h (strdup): Don't declare.
2395 * stabs.c: Include libiberty.h
2396 (get_stab_string_offset): Use xstrdup rather than strdup.
2397 (s_stab_generic): Likewise.
2398 * as.c (parse_args): Likewise.
2399 * read.c (s_mri_sect): Likewise.
2401 * gasp.c (change_base): Recognize \(...) construct documented to
2402 pass through enclosed characters literally through to the output.
2403 (process_assigns): Likewise. Also, be more careful to avoid
2404 looking past the end of the buffer.
2408 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
2413 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
2417 * config/alpha-opcode.h: Added cvtst instruction.
2421 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
2422 string. Extract the addend from the result of expression.
2426 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
2427 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
2428 mode, change the state to 3 rather than 1.
2432 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
2433 correspond to 11 March change.
2437 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
2438 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
2442 * configure: Rebuild with autoconf 2.8.
2446 * config/atof-ieee.c (gen_to_words): Improve handling of
2447 X_PRECISION numbers. Based on patches from Andreas Schwab
2452 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
2454 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
2455 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
2456 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
2457 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
2462 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
2466 * config/tc-hppa.c (SUBSEG_MILLI): Define.
2467 (pa_def_subspaces): Add $MILLICODE$.
2468 (pa_spaces_begin): Set section flags for $MILLICODE$.
2472 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
2473 not SHT_NOBITS. Don't tamper with flags based on type if a
2474 special section was found (revert Feb 29 change).
2476 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
2477 if aligning to a longword boundary or greater.
2481 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
2485 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
2486 * configure: Regenerated.
2490 * config/tc-hppa.c: Add default definitions for R_N0SEL and
2491 R_N1SEL since they're not defined for old versions of hpux.
2493 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
2494 Set "sym_ptr_ptr" and "addend" fields to dummy values for
2495 R_N0SEL and R_N1SEL.
2499 * flonum-konst.c: Add two more constants for 1e+-2048 and
2500 1e+-4096, and correct the other constants.
2502 * symbols.c (resolve_symbol_value): Handle O_logical_not.
2506 * config/obj-elf.c (obj_elf_section): Allow predefined section
2507 types to set the nobits type. Avoid a shadowed declaration.
2511 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
2512 reduce relocs using e_nlrsel field selectors.
2514 * write.c (fix_new_exp): Don't use #elif. Some compilers
2517 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
2519 (pa_chk_field_selector): Handle new field selectors for SOM.
2523 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
2524 * configure: Rebuild.
2525 * config/te-linux.h (LOCAL_LABELS_FB): Define.
2526 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
2528 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
2533 * mpw-make.sed: Update to handle shared library support.
2537 * config/tc-sparc.c (sparc_ip): Print all architectures that support
2538 the insn on mismatch.
2542 * configure.in: Add support for a29-coff.
2543 * configure: Rebuild.
2547 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
2548 of the section when looking for the R_SH_USES fixup, because the
2549 frag addresses have not yet been adjusted.
2551 * gdbinit.in: Set a breakpoint on as_warn_where.
2553 * config/tc-mips.c (macro): Add missing arguments to macro_build
2558 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
2559 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
2563 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
2567 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
2568 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
2569 Recognize [uU] format args as sparclet cpregs.
2573 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
2578 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
2579 well as S_IS_EXTERNAL.
2580 (tc_fix_adjustable): Likewise.
2581 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
2582 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
2584 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
2585 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
2589 * config/tc-sparc.c (max_architecture): Change to sparclite for
2591 (default_compatible): Delete.
2592 (sparc_ffs): New function.
2593 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
2594 (sparc_ip): Rewrite architecture match and bump logic.
2598 * configure.in: Check for 'do not mix' from native linker before
2599 trying to use -rpath.
2600 * configure: Rebuild.
2604 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2605 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2606 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2607 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
2609 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
2610 to the symbol lnnoptr field, to get the correct file offset.
2614 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
2615 symbols that are not function, file, or section symbols to be
2620 * configure: Set and substitute RPATH_ENVVAR.
2621 * configure: Rebuild.
2622 * Makefile.in (RPATH_ENVVAR): New variable.
2623 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
2626 doesn't need opcodes. If configuring shared, opcodes needs bfd.
2627 * configure: Rebuild.
2631 * read.c (s_mri_sect): Don't return '\0' in type. Set all
2632 appropriate flags in BFD section.
2634 * configure.in (m68k-*-psos*): New target.
2635 * configure: Rebuild.
2636 * config/te-psos.h: New file.
2637 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
2643 * configure.in: Remove duplicate setting of cpu_type. Check
2644 whether opcodes library is required for on all targets, not just
2646 * configure: Rebuild.
2648 * config/tc-mips.c (mips_big_got): New static variable.
2649 (s_extern): Don't declare.
2650 (reg_needs_delay): New static function.
2651 (macro_build): Permit GOT/CALL_HI/LO relocs.
2652 (macro_build_lui): If place is not NULL, use the number in the
2654 (load_address): Handle mips_big_got case.
2655 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
2657 (OPTION_XGOT): Define.
2658 (md_longopts): Add "xgot" if OBJ_ELF.
2659 (md_parse_option): Handle -xgot.
2660 (md_show_usage): Mention -xgot.
2661 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
2662 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
2666 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
2667 the byte relocation, point it to the low byte of the word.
2671 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
2672 * configure: Rebuild.
2676 * configure.in: Recognize any sparc* cpu.
2677 * configure: Regenerated.
2681 * read.c (potable): Add "mri" and ".mri".
2682 (s_mri): New function.
2683 * read.h (s_mri): Declare.
2684 * app.c (scrub_m68k_mri): New static variable.
2685 (mri_pseudo): New static variable.
2686 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
2687 flag_m68k_mri. Initialize scrub_m68k_mri.
2688 (mri_state, mri_last_ch): New static variables.
2689 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
2691 (app_push): Save new fields.
2692 (app_pop): Restore new fields.
2693 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
2694 If TC_M68K, use a trivial state machine to look for occurrences of
2695 the .mri pseudo-op, and change the mode appropriately.
2696 * as.h (do_scrub_begin): Update prototype.
2697 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
2699 * config/tc-m68k.c (reg_prefix_optional_seen): New static
2701 (m68k_mri_mode_change): New function.
2702 (md_parse_option): Set reg_prefix_optional_seen.
2703 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
2704 (MRI_MODE_CHANGE): Define.
2705 * doc/as.texinfo: Document .mri pseudo-op.
2707 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
2712 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
2713 * config/m68k-parse.h (enum pic_relocation): Define.
2714 (struct m68k_exp): Add pic_reloc field.
2715 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
2716 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
2717 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
2719 * config/tc-m68k.c: Delete definition of NO_RELOC.
2720 (struct m68k_it): Add pic_reloc field.
2721 (add_fix): Copy over pic_reloc field.
2722 (md_pseudo_table): Interpret .align parameter as byte count.
2723 (mote_pseudo_table): Likewise.
2724 (tc_m68k_fix_adjustable): New function.
2725 (get_reloc_code): New function.
2726 (md_assemble): Use it as last argument to fix_new_exp.
2727 (md_apply_fix_2): For a relocation against a symbol don't put the
2728 addend into the data.
2729 (tc_gen_reloc): Different addend computation for OBJ_ELF.
2730 (m68k_ip): Don't relax an operand that requires pic relocation.
2731 (md_begin): Align .text, .data and .bss on 4 byte boundary by
2733 * write.c (fixup_segment): Don't add symbol value to addend if
2734 TC_M68K and OBJ_ELF.
2735 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
2736 (motorola_operand): Add rule for `(zapc, EXPR)'.
2740 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
2741 *really* is permissible before the comma.
2745 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
2746 inserting nop instructions.
2750 * config/te-aux.h: Change include of aux.h to aux-coff.h.
2754 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
2755 class setting for a CALLNAME symbol in COFF.
2757 * read.c (potable): Pass negative numbers for new .balign[wl] and
2758 .p2align[wl] pseudo-ops.
2759 (s_align_bytes): Treat a negative argument as specifying the fill
2761 (s_align_ptwo): Likewise.
2765 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
2766 (do_align): Take new len parameter. Change all callers. Pass it
2768 (s_align_bytes): Arg now indicates the length of the fill pattern.
2769 (s_align_ptwo): Likewise.
2770 * config/obj-coff.c (write_object_file): Pass length to
2772 * config/tc-i386.h (md_do_align): Take new len parameter.
2773 * config/tc-m88k.h (md_do_align): Likewise.
2774 * config/tc-m88k.c (m88k_do_align): Likewise.
2775 * config/tc-sh.h (md_do_align): Likewise.
2776 * config/tc-sh.c (sh_do_align): Likewise.
2777 * doc/as.texinfo: Document new pseudo-ops.
2779 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
2780 done in cvt_frag_to_fill.
2782 * config/tc-sh.h (sh_do_align): Declare.
2783 (md_do_align): Define.
2784 * config/tc-sh.c (sh_do_align): New function.
2786 * ecoff.c (ecoff_build_lineno): Don't try to store the address
2787 difference if the next address is before the current one.
2789 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
2790 (archs): Initialize new field.
2791 (m68k_ip): Don't list alias names when listing CPUs which support
2794 * as.c (main): Call parse_args before read_begin.
2795 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
2796 front of generated pseudo-ops.
2797 * read.c (potable): Ignore "name".
2798 (s_app_file): Permit a single quote after the string, since one
2799 may appear in m68k MRI mode.
2801 * configure.in: Check for --enable-shared. If linking against
2802 shared BFD and opcodes, fix library name on SunOS, and try to set
2804 * configure: Rebuild.
2808 * as.h (flag_m68k_mri): Declare.
2809 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
2810 * Many files: For MRI syntax that is specific to the m68k MRI
2811 assembler, check flag_m68k_mri rather than flag_mri or
2812 MRI_MODE_NEEDS_PSEUDO_DOT.
2816 * config/tc-i960.c (ARCH_HX): Define.
2818 (targ_has_sfr): Handle ARCH_HX.
2819 (targ_has_iclass): Handle ARCH_HX.
2820 (tc_coff_fix2rtype): Add return 0 to avoid warning.
2821 (tc_headers_hook): If the architecture was specified explicitly,
2822 use it when setting the flags. Set the extern variable coff_flags
2823 rather than headers->filehdr.f_flags, since the latter is set
2824 unconditionally in obj-coff.c.
2825 (i960_handle_align): Remove unused variable fixp.
2827 Support for building bfd and opcodes as shared libraries, based on
2829 * configure.in: Set OPCODES and BFD to search directories.
2830 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
2831 * configure: Rebuild.
2832 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
2833 (LIBDEPS): New variable.
2834 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
2835 $(HLDFLAGS) in link.
2836 (check): Set LD_LIBRARY_PATH in the environment.
2840 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
2841 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
2842 read-only, not read/write.
2846 * Makefile.in (INSTALL_XFORM): Remove -e.
2850 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
2851 than comparing S_GET_SEGMENT to undefined_section.
2852 (write_object_file): Skip symbols which were equated to an
2853 undefined or common symbol.
2854 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
2855 S_IS_COMMON rather than comparing S_GET_SEGMENT to
2857 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
2858 or S_IS_COMMON call S_GET_VALUE.
2859 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
2860 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
2861 to an undefined or common symbol, convert the reloc to be against
2863 (obj_crawl_symbol_chain): Skip symbols which were equated to an
2864 undefined or common symbol.
2865 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
2866 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
2867 to an undefined or common symbol, convert the reloc to be against
2869 (obj_crawl_symbol_chain): Skip symbols which were equated to an
2870 undefined or common symbol.
2871 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
2872 S_IS_COMMON rather than comparing S_GET_SEGMENT to
2874 (yank_symbols): Skip symbols which were equated to an undefined or
2879 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
2880 * config/obj-bout.h (S_IS_LOCAL): Likewise.
2882 * configure.in: Make sure we only add m68k-parse.o to
2883 ${extra_objects} once, no matter how many m68k targets have been
2885 * configure: Rebuild.
2889 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
2890 * configure: Rebuild.
2894 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
2895 * doc/c-m68k.texi: Document .extend and .ldouble.
2897 * configure.in (m68*-apple-aux*): New target.
2898 * config/te-aux.h: New file.
2899 * config/obj-coff.c (compare_external_relocs): New static function
2901 (do_relocs_for): Sort relocs if TE_AUX.
2902 (fixup_segment): If TE_AUX, store common symbol value in segment.
2903 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
2907 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
2909 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
2911 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
2912 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
2914 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
2915 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
2916 an undefined symbol and the size of a function symbol.
2917 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
2918 for all common symbols.
2922 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
2925 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
2930 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
2933 * ecoff.c (localsym_t): Add addend field.
2934 (add_ecoff_symbol): Add addend argument. Change all callers.
2935 (coff_sym_value): Make static.
2936 (coff_sym_addend): New static variable.
2937 (ecoff_directive_def): Initialize coff_sym_addend.
2938 (ecoff_directive_val): Accept symbol + constant.
2939 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
2940 (ecoff_build_symbols): Include the addend in the symbol value.
2944 * config/tc-ppc.c (md_assemble): Ignore overflow on
2945 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
2949 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
2950 bytes in size, so offset appropriately in big endian mode when
2951 writing the bottom 2 bytes.
2955 * config/tc-sparc.c (default_compatible): New static local.
2956 (md_begin): Initialize it. Rewrite warn_on_bump handling.
2957 (sparc_ip): If no architecture or -bump specified, don't mark as
2958 mismatched those in default_compatible.
2963 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
2964 * configure: Rebuild.
2965 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
2966 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
2967 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
2968 (sco_id): Declare if SCO_ELF.
2972 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
2973 ({max,warn_after}_architecture): New static locals.
2974 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
2975 If both architecture and -bump requested, set max_architecture to max.
2976 (sparc_md_end): Simplify.
2977 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
2978 Rewrite code to bump architecture and check for conflicts.
2979 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
2980 with Solaris assembler.
2981 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
2982 (md_show_usage): Update.
2986 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
2990 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
2991 #ifdef sparcv9 when choosing value.
2994 (sparc_md_end): Declare.
2995 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
2996 (initial_architecture): New static local.
2997 (can_bump_v9_p): Likewise.
2998 (NO_V9): Delete all occurrences.
2999 (sparc_md_end): New function.
3000 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
3001 Don't bump architecture to v9 unless can_bump_v9_p set.
3002 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
3003 * configure.in (sparc64 target cpu): Don't set obj_format here.
3004 (SPARC_V9): Renamed from sparcv9.
3005 (sparc64-*-elf*): Define SPARC_ARCH64.
3006 * configure: Regenerated.
3007 * acconfig.h (SPARC_V9): Renamed from sparcv9.
3008 (SPARC_ARCH64): Add.
3009 * conf.in: Regenerated.
3010 * config/vmsconf.h: Update.
3014 * config/tc-mips.c (load_register): Optimise "dli" loads.
3015 (md_show_usage): add "-mcpu=vr4100" to help text.
3019 * symbols.c (resolve_symbol_value): If a symbol is equated to an
3020 undefined symbol, preserve the X_op of O_symbol.
3021 (S_GET_VALUE): Fix check to permit this case.
3022 * write.c (write_relocs): If a reloc is against an undefined
3023 symbol equated to another symbol, change the reloc to be against
3025 * config/obj-coff.c (do_relocs_for): Likewise.
3027 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
3032 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
3033 computer is PowerPC.
3035 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
3037 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
3038 mentioned in the eabi.
3042 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
3043 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
3047 * config/tc-ppc.c (mapping): Add more relocation suffixes.
3051 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
3052 as a section it is ok to have unadorned -mrelocatable pointers in.
3056 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
3061 * config/obj-elf.c (obj_elf_section): Add hooks so machine
3062 dependent section attributes can be handled.
3064 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
3065 macros to add support for exclude section flag and ordered section
3068 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
3069 functions to add support for exclude section flag and ordered
3074 * subsegs.c (section_symbol): Don't try to look up the section
3075 symbol in the hash table. It should be possible to have a symbol
3076 with the same name as a section, but no connection to it.
3078 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
3081 * expr.c (operand): Skip whitespace after a close parenthesis.
3086 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
3087 as_warn_where for relocation overflow.
3088 (parse_reg): Accept register name only if next character is
3093 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
3094 addresses for static variables.
3098 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
3099 and xgot give the old toc16 relocation.
3100 (md_apply_fix3): Support all GOT relocations.
3104 * config/tc-mips.c (load_address): Correctly handle a constant in
3109 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
3110 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
3111 (build_Mbytes): Handle REG_B.
3115 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
3119 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
3123 * config/obj-coff.c (write_object_file): Set the s_align field to
3124 the number of bytes, rather than to the power of 2.
3128 * Makefile.in (DISTCLEAN_HERE): New variable.
3129 (distclean): Use it.
3130 (maintainer-clean): Depend upon clean-here rather than clean,
3131 distclean, and clean-info. Run make maintainer-clean in doc.
3132 Remove files listed in DISTCLEAN_HERE.
3133 * doc/Makefile.in (maintainer-clean realclean): Split out from
3134 distclean. Depend upon clean-info and distclean.
3138 * mac-as.r: Fix copyright and version strings.
3139 (cfrg): Use PROG_NAME instead of literal name.
3143 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
3145 * config/tc-a29k.h (tc_unrecognized_line): Define.
3146 * config/tc-a29k.c (a29k_unrecognized_line): New function.
3147 (md_operand): Handle a29k style local dollar labels.
3151 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
3155 * read.c (s_fill): If md_flush_pending_output is defined, call
3160 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
3161 Treat rs_align_code like rs_align.
3165 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
3166 converting an absolute address into a PC-relative one.
3170 * mpw-config.in: Don't always use te-generic.h for emulation.
3171 (powerpc-apple-macos): Use emulation te-macos.h.
3173 (mips-gm-magic): New configuration.
3175 * mpw-make.sed (install, install-only): Edit in Mac-specific
3180 * configure.in: Improve message about unsupported ELF targets.
3181 * configure: Rebuild.
3183 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
3184 pattern is zero. From Manfred Hollstein.
3188 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
3189 the various idata sections, we check the name on the .section pseudo.
3193 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
3194 add section's paddr to add_number; compatibility to native as and
3199 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
3201 * hash.c (struct hash_entry): Moved here...
3202 * hash.h (struct hash_entry): ...from here.
3204 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
3205 if it's already known to be null.
3209 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
3210 bit relocs converted to PC relative relocs. This turned out to
3211 add the offset from the beginning of .text twice.
3215 * stabs.c (s_stab_generic): In 's' case, free string from
3218 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
3220 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
3223 * subsegs.h (segment_info_type): Make bitfields unsigned.
3225 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
3226 and move them together. On most systems this will reduce the
3227 structure size by one word.
3228 (operatorT): Define O_max.
3229 * expr.c (expr_begin): Verify that X_op is wide enough to hold
3232 * read.c (pop_insert): Print error returned by hash table
3235 * as.c (dump_statistics): Split out from main; dump some hash
3236 table stats and target-specific stats.
3237 (start_time): No longer automatic to main.
3238 (main): Set file-level start_time and call dump_statistics at
3239 exit. Exit by calling xexit.
3240 (show_usage): Make --statistics description less specific.
3241 * subsegs.c (subsegs_print_statistics): New function.
3242 * write.c (write_print_statistics): New function.
3243 (n_fixups): New static variable.
3244 (fix_new_internal): Increment it.
3245 * read.c (read_print_statistics): New function.
3246 * read.h (read_print_statistics): Declare.
3247 * symbol.c (symbol_print_statistics): New function.
3248 * symbol.h (symbol_print_statistics): Declare.
3249 * hash.c (hash_print_statistics): New function.
3250 * hash.h (hash_print_statistics): Declare.
3251 * config/tc-i386.c (i386_print_statistics): New function.
3252 * config/tc-i386.h (i386_print_statistics): Declare.
3253 (tc_print_statistics): New macro.
3254 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
3257 * hash.c (DELETED): Rewrite to use a valid but unique address.
3258 (START_POWER): Reduce to 10.
3259 (enum stat_enum): New enumerator, replacing STAT_* index macros.
3260 Add new values for counting strcmp calls.
3261 (GROW_FACTOR): New macro.
3262 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
3264 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
3265 (hash_new): Use FULL_VALUE.
3266 (struct hash_control): Definition moved here.
3267 (hash_code): Don't mask to low bits.
3268 (hash_ask): Mask returned hash code here. Check hash value before
3269 calling strcmp; count strcmp calls.
3270 * hash.h (struct hash_control): Declare, don't define, here.
3271 (HASH_STATLENGTH): Deleted.
3272 (struct hash_entry): Add field for hash code.
3273 (hash_say, hash_apply): Don't declare.
3275 * hash.c (destroy): Return void.
3276 (applicatee): Ditto.
3277 (main): Fix declarations.
3278 (hash_apply): Return void. Argument `function' returns void. Put
3279 inside "#ifdef TEST".
3280 (hash_say): Define only if TEST is defined.
3281 * hash.h (hash_apply, hash_say): Declarations deleted.
3285 * configure: Rebuild with autoconf 2.7.
3289 * aclocal.m4 (AC_PROG_CC): Remove local definition.
3290 * configure: Rebuild with autoconf 2.6.
3294 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
3295 (ppc_stabx): Don't increment ppc_debug_name_section_size.
3297 (ppc_frob_file): Remove.
3298 * config/tc-ppc.h (tc_frob_file): Don't define.
3299 (ppc_frob_file): Don't declare.
3303 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
3304 * config/alpha-opcode.h: Include one-operand variants of jmp and
3307 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
3308 with te-delta.h in 8.3 file systems.
3309 * configure.in: Adjusted.
3313 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
3315 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
3317 * expr.c (operand): Handle MRI suffixes after unadorned 0.
3321 Version 2.6 released.
3322 * Makefile.in (VERSION): Updated to 2.6.
3324 * config/obj-coff.c (write_object_file): Change use of md_do_align
3325 to pass a pointer rather than a fill value, to match other uses.
3329 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
3331 * as.c (main): Move md_end call to just after call to
3332 perform_an_assembly_pass. Delete cpu-specific code here.
3333 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
3334 * config/tc-arm.c (md_end): Unused function deleted.
3335 * config/tc-ns32k.c (md_end): Ditto.
3337 * config/tc-i386.c (i386_align_code): New function, moved here
3338 from HANDLE_ALIGN macro.
3339 * config/tc-i386.h (HANDLE_ALIGN): Call it.
3343 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
3348 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
3349 multi-byte nops to avoid changing gcc. The necessary gcc change
3350 might break old assemblers.
3354 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
3355 generate optimal multi-byte nop instructions for ".align n"
3356 ".align n,0x90", and aligns requiring more than 15 bytes of
3357 padding still generate multiple 0x90's as before.
3361 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
3365 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
3366 * configure: Update.
3367 * mpw-make.sed: Reorder commands to make sed happier.
3368 * config/te-macos.h: New file.
3369 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
3373 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
3374 conditional branches.
3375 (md_apply_fix): Likewise.
3379 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
3380 generate a reloc which the linker will not handle correctly. Fix
3381 overflow checking--R_IREL is 18 bits, not 17.
3385 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
3386 from a PC relative reloc if TC_A29K.
3388 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
3389 code depends upon it.
3391 * config/tc-a29k.c (md_operand): Handle $float, $double, and
3392 $extend. Based on code from Eric Freudenthal
3394 * config/tc-a29k.h (LEX_DOLLAR): Define.
3395 * read.c (LEX_DOLLAR): Define if not defined.
3396 (lex_type): Use LEX_DOLLAR.
3400 * configure.in (a29k-nyu-sym1): New target, just like other a29k
3405 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
3409 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
3410 used only by "dlltool.c".
3414 * config/tc-alpha.c: Undefine inline if not __GNUC__.
3415 (md_pseudo_table): Don't define "extern".
3419 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
3421 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
3425 * macro.c (macro_expand_body): Don't warn about == with a
3426 nonexistent parameter, in case it is in a comment field.
3428 * as.c (main): On TC_A29K, call macro_init with macro_alternate
3430 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
3431 set, even if macro_alternate is set.
3432 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
3433 (sub_actual): If macro_strip_at, and kind is '@', don't look up
3434 the token unless it ended in '@'.
3435 * config/tc-a29k.c (line_separator_chars): Remove '@'.
3436 * doc/c-a29k.texi: Document macro usage on A29K.
3440 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
3442 (install_operand): Likewise.
3444 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
3448 * configure.in (m88k-motorola-sysv*): New target.
3449 * configure: Rebuild.
3450 * config/te-delta88.h: New file.
3451 * config/obj-coff.c (write_object_file): Use md_do_align if it is
3453 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
3454 (md_do_align): Define.
3455 * config/tc-m88k.c: Include "subsegs.h".
3456 (m88k_do_align): New function.
3458 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
3459 (COFF_NOLOAD_PROBLEM): Define.
3460 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
3464 * config/tc-i386.c (md_assemble): For a jump instruction with
3465 non-constant target, require 7 available bytes in the current
3470 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
3475 * configure.in (alpha-*-linux*): Use ecoff.
3476 * configure: Rebuild.
3477 * ecoff.c (ecoff_directive_extern): New function.
3478 (ecoff_directive_weakext): New function.
3479 (ecoff_build_symbols): Handle weak symbols.
3480 (ecoff_setup_ext): Likewise.
3481 (ecoff_frob_symbol): Warn about weak common symbols.
3482 * ecoff.h (ecoff_directive_extern): Declare.
3483 (ecoff_directive_weakext): Declare.
3484 * symbols.c (S_IS_WEAK): New function.
3485 * symbols.h (S_IS_WEAK): Declare.
3486 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
3488 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
3493 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
3495 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
3496 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
3497 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
3498 * config/obj-coff.c (coff_frob_symbol): Don't call
3499 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
3500 one. If OBJ_XCOFF, try to figure out whether the symbol is going
3503 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
3504 (ppc_stab_symbol): New static variable.
3505 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
3506 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
3507 sy_tc.real_name to the stab string.
3508 (ppc_bc, ppc_ec): New static functions.
3509 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
3511 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
3513 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
3516 * input-scrub.c (input_scrub_push): Reset sb_index.
3520 * listing.c (listing_newline): Don't create a frag in the absolute
3525 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
3526 (ppc_csect): Move most of the code to ppc_change_csect, and call
3528 (ppc_change_csect): New static function, taken from ppc_csect.
3529 (ppc_section): New static function.
3530 (ppc_saw_abs): New static varable.
3531 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
3532 a symbol has no csect.
3533 (ppc_adjust_symtab): New function.
3534 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
3535 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
3537 * write.c (write_object_file): If tc_adjust_symtab is defined,
3538 call it just before the call to obj_adjust_symtab.
3540 * symbols.c (symbol_find_or_make): Change name to be const.
3541 * symbols.h (symbol_find_or_make): Update declaration.
3545 * doc/as.texinfo (Align): Mention SH.
3546 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
3548 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
3549 (Z8000 Directives, even): Don't give numeric align value, instead
3554 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
3559 * Makefile.in (diststuff): Also make info.
3560 (maintainer-clean realclean): Also make clean-info.
3564 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
3565 (pa_subspace): For ".nsubspa", always create a new subspace
3566 with the given attributes, even if one already exists with the
3571 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
3572 TC_FORCE_RELOCATION, taking an additional section argument. If
3573 the section of the target symbol is not the same as the current
3574 section, always force the relocation to be used.
3575 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
3577 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
3578 md_pcrel_from function, taking an additional section argument.
3579 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
3581 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
3582 older TC_FORCE_RELOCATION if not defined.
3583 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
3584 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
3585 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
3586 TC_FORCE_RELOCATION.
3590 * input-scrub.c (as_where): Set name to null pointer if we don't
3592 * messages.c (identify): Only print filename if non-null.
3593 (as_show_where): Ditto, for line number too.
3594 (as_warn_internal, as_bad_internal): Ditto.
3596 * input-file.c (input_file_open): If the input file can't be
3597 opened, consider it an error.
3601 * config/tc-mips.c: Added mips_4100 control, and support for
3602 accepting the 4100 as a MIPS architecture variant (md_begin,
3603 macro_build, mips_ip, md_parse_option). Adding suitable
3604 command-line OPTIONs, and updating the help text (md_show_usage).
3608 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
3609 BFD_ASSEMBLER is defined.
3611 * Use one active frag and one obstack per frag chain:
3612 * frags.c (frags): Variable deleted.
3613 (frag_alloc): New function.
3614 (frag_grow, frag_more, frag_variant, frag_now_fix,
3615 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
3617 (frag_new): Ditto. Verify that frch_last and frag_now match on
3618 entry and exit, and that old frag_now has non-zero type. Replace
3619 "know" uses with "assert". Use frag_alloc instead of mucking with
3621 * frags.h (frags): Declaration deleted.
3622 * subsegs.h (struct frchain): Add new field frch_frag_now.
3623 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
3625 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
3626 give it any stricter alignment than frchainS structures need. Do
3627 not initialize frags obstack. Set frag_now to point to
3628 dummy_obstack. Initialize absolute_frchain.
3629 (subseg_set_rest): Save and restore frag_now in frch_frag_now
3630 field of frchainS. Don't create new frags on section switch, and
3631 use frag_alloc when creating a new frag chain. For absolute
3632 section, set frchain_now to absolute_frchain. Verify that
3633 frch_last and frag_now match on entry and exit. Initialize
3634 per-chain obstack, and under gcc, set required alignment to that
3635 needed by fragS structure.
3637 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
3639 * stabs.c (get_stab_string_offset): Only copy input string if a
3640 fresh copy is needed, not if the section already exists.
3641 (s_stab_generic): Cache stab section name to bypass lookups, since
3642 usually it will match. Could be made faster still by changing the
3643 memory allocation rules.
3644 (s_xstab): Cache section name to bypass repeated string
3647 * frags.c (frag_new): Deleted register declarations.
3649 * listing.c (frag_now): Don't declare.
3651 * as.c (chunksize): New variable.
3652 (debug_memory): New variable.
3653 (main): If debug_memory is set, reduce chunksize and
3655 * as.h (chunksize): Declare it.
3656 * read.c (read_begin): Use it.
3658 * config/tc-alpha.c (md_shortopts): Include 'g'.
3659 (md_parse_option): Ignore it.
3661 * Makefile.in (distclean): Remove Makefile and config.status from
3662 testsuite directory.
3663 (clean-here): Don't delete testsuite. Instead, delete only the
3664 files within it that would be generated by running tests.
3666 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
3667 frag_now_fix instead of accessing obstack info directly.
3668 * config/tc-arm.c (s_ltorg): Ditto.
3669 (md_assemble): Ditto.
3671 * config/tc-i386.c (md_assemble): Call frag_grow instead of
3676 * stabs.c (aout_process_stab): Insert debug symbol into symbol
3677 chain after parsing value expression, if any, to avoid separating
3678 continued .stabs lines.
3682 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
3683 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
3687 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
3688 -mcom support and PPC_OPCODE_ANY for -many.
3689 (md_parse_option): Ditto.
3691 (md_begin): For duplicate instructions, print all duplicates
3696 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
3697 common mode operation.
3698 (md_show_usage): Add -mcom to usage message.
3702 * expr.c (op_rank): Add O_symbol_rva.
3703 * expr.h (operatorT): Add O_symbol_rva.
3704 * read.c (cons_worker): Set O_symbol_rva when necessary.
3705 * write.c (fix_new_exp): Understand O_symbol_rva.
3709 * config/tc-mips.c: Correct s_cons arguments. From Michael
3714 * config/tc-ppc.c (ppc_macro): Make count unsigned.
3715 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
3716 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
3718 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
3719 make non-static. Change all users.
3720 * config/obj-coff.h (coff_n_line_nos): Declare.
3726 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
3728 * config/obj-elf.c (obj_elf_common): Convert specified byte
3729 alignment to power of two. Set size of local bss symbol.
3731 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
3735 * sb.c, macro.c: Decide whether to include <string.h> or
3736 <strings.h> just as as.h does.
3740 * Makefile.in (site.exp): Fix setting of $srcdir.
3742 * config/tc-arm.c (md_atof): Fix little-endian output.
3743 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
3748 * doc/as.texinfo: Split out the NS32k family documentation,
3749 despite its being commented out for now.
3750 * doc/c-ns32k.texi: New file.
3752 * sb.c, macro.c: Include string.h.
3754 * Makefile.in (comparison): Only check *.o; we don't care if
3755 timestamps inserted by the native linker differ.
3757 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
3758 if alignment stricter than 4 bytes is requested; in that case,
3759 align to a 4-byte boundary first.
3763 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
3764 uninitialized" warning for `Max_Source_Offset'.
3768 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
3769 qualifiers and detect errors if present.
3770 (md_assemble): Add call to parse_toc_entry. Also added some support
3771 for the [tocv] qualifier.
3772 (ppc_pe_tocd): New function to support data in the toc section.
3776 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
3781 * config/tc-m68k.c (isword): Accept all values from -65536 to
3782 +65535, so ~VAL will not be rejected.
3784 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
3785 skip characters after the pseudo-op.
3787 * read.c (get_line_sb): Don't look past buffer_limit.
3788 (s_include): In MRI mode, skip characters after the file name.
3792 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
3793 periods in register names.
3795 For older changes see ChangeLog.1.