3 * write.c (adjust_reloc_syms): If the fixup symbol has been
4 equated to an undefined symbol, convert the fixup to being against
5 the target symbol. Remove obsolete code handling a special case
10 * mpw-make.sed: Use NewFolderRecursive for installation.
15 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
19 * config/tc-d10v.c (md_assemble): Fix bug which caused
20 second instruction in a line to be case sensitize. PR11312
25 * config/tc-mn10200.c (mn10200_insert_operand): Don't
26 range check operands with MN10200_OPERAND_NOCHECK set.
27 (check_operand): Likewise.
31 * config/tc-mips.c: Undo part of last Friday's alignment changes.
32 (md_begin): Always align the text section to a four byte
34 (append_insn): Remove call to record_align.
36 * config/tc-mips.c (insn_label): Remove.
37 (struct insn_label_list): Define.
38 (insn_labels, free_insn_labels): New static variables.
39 (mips_clear_insn_labels): New static function.
40 (append_insn): Mark all mips16 text labels, and make them odd.
41 Handle all labels after emitting a nop, not just one. Call
42 mips_clear_insn_labels rather than just clearing insn_label.
43 (mips_emit_delays): Add insns parameter, and use it to decide
44 whether to mark mips16 labels. Handle all labels, not just one.
45 Force mips16 labels to be odd. Change all callers.
46 (mips16_immed): Don't check for an odd branch target.
47 (md_apply_fix): Don't check mips16 mode for a branch reloc.
48 (mips16_extended_frag): Ignore the low bit in a branch target.
49 (md_convert_frag): Likewise.
50 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
52 (mips_align, mips_flush_pending_output, s_cons): Likewise.
53 (s_float_cons, s_gpword): Likewise.
54 (s_align): Use insn_labels rather than insn_label.
55 (s_cons, s_float_cons, s_gpword): Likewise.
56 (mips_frob_file_after_relocs): New function.
57 (mips_define_label): Rewrite to add to insn_labels list.
58 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
59 * ecoff.c (ecoff_build_symbols): If the size of a function comes
60 out odd, increment it.
62 * config/tc-mips.c (append_insn): Only update prev_insn when not
63 reordering if place is NULL.
65 * config/tc-mips.c (mips16_ip): Check for a missing expression
66 when using the register indirect addressing mode.
70 * config/tc-mn10200.c (mn10200_insert_operand): Don't
71 check 24bit operands for overflow.
72 (check_operand): Likewise.
76 * doc/as.texinfo (Section): Document how to use the .section
77 pseudo-op for COFF and ELF.
81 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
85 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
86 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
87 store them. Adjust other RELAX_MIPS16 macros.
88 (RELAX_MIPS16_DSLOT): Define.
89 (RELAX_MIPS16_JAL_DSLOT): Define.
90 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
91 handling of whether previous instruction has a fixup. Set
93 (mips_no_prev_insn): Clear prev_insn_reloc_type.
94 (mips16_extended_frag): Use the right base address for a PC
96 (md_convert_frag): Likewise. If a PC relative add or load is
97 used, record the alignment for the section.
101 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
102 linkonce section into a reloc against the section symbol.
104 * config/tc-mips.c (mips16_macro): Remove nop instructions after
107 * config/tc-mips.c (md_begin): If configured for an embedded ELF
108 system, don't set the section alignment to 2**4.
109 (s_change_sec): Likewise.
110 (append_insn): Call record_alignment for the section.
111 (md_section_align): Don't align the section size for an embedded
116 * write.c (adjust_reloc_syms): Make sure that symbols are
117 resolved; expression symbols may have been skipped.
118 * config/obj-coff.c (fixup_segment): Likewise.
122 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
123 BFD_RELOC_24_PLT_PCREL relocation.
124 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
129 * config/tc-d10v.c (write_2_short): Remove code that called
130 parallel_ok() when the programmer specified parallel instructions.
135 * config/tc-mn10300.c (md_assemble): Update to handle endianness
138 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
139 * config/tc-mn10300.c (md_assemble): Likewise.
143 * config/tc-mips.c (append_insn): Make sure there is enough room
144 in a frag after a mips16 instruction to switch it with a jump
147 * config/tc-mips.c (mips16_extended_frag): Give an error for an
148 attempt to use a non absolute symbol in an extending frag.
152 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
153 * config/tc-mn10200.h: Likewise.
157 * app.c (do_scrub_chars): At the end of a C comment, pass space to
158 UNGET rather than PUT. Set old_state before setting state to -2.
160 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
161 when extending because the value is exactly maxtiny + 1.
163 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
164 arguments, and store them. Adjust other RELAX_MIPS16 macros.
165 (RELAX_MIPS16_USER_SMALL): Define.
166 (RELAX_MIPS16_USER_EXT): Define.
167 (mips16_small, mips16_ext): New static variables.
168 (append_insn): Pass mips16_small and mips16_ext to
170 (mips16_ip): Set mips16_small and mips16_ext.
171 (mips16_immed): Don't check mips16_autoextend.
172 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
174 * write.c (write_relocs): Print an error for an out of range
175 fixup, rather than calling abort.
177 * as.c (main): Unlink the output file if there are errors while
178 generating the fixups.
182 * config/tc-mips.c (mips16_extended_frag): Don't call
184 (md_convert_frag): Call resolve_symbol_value before calling
185 S_GET_VALUE, and don't add in the frag address.
187 * config/tc-mips.c (mips16_immed): Add file and line parameters,
188 and use them when reporting errors. Change all callers.
192 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
193 Remove '$' prefixing for registers.
198 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
199 is initialized before using it.
200 (d10v_cleanup): No longer uses its argument, so make it void.
202 * config/tc-d10v.h (d10v_cleanup): Change prototype.
207 * write.c (fixup_segment): Don't discard the symbol for a PC
208 relative fixup to an absolute symbol.
213 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
214 with multiple sections.
219 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
220 no-mips-16 to no-mips16.
221 (s_mipsset): Accept .set mips16 and .set nomips16.
225 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
226 argument, and check for +/- constant following the suffix, folding
227 it into the expression.
228 (ppc_elf_cons): Change ppc_elf_suffix calls.
229 (md_assemble): Ditto.
230 (shlib): Replace boolean mrelocatable with enumeration shlib.
231 (md_parse_option): Discriminate between PIC style shared libraries
233 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
238 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
239 change the comment characters.
241 * config/tc-ppc.c (comment_chars): Delete in favor of
243 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
245 (ppc_comment_chars): Select appropriate comment chars by default.
246 (msolaris): New flag for -m{,no-}solaris.
247 (md_parse_option): Recognize -K pic. Add support for
249 (md_show_usage): Update.
250 (md_begin): Do not set ELF flags if Solaris.
251 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
252 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
256 * as.c (main): Correct handling of flag_always_generate_output.
260 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
261 fx_offset, not fx_addnumber.
263 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
268 * config/tc-i386.c: Correct misspelling: balenced to balanced.
272 * config/tc-mips.c (md_section_align): Check for an alignment of
273 4, not an alignment of 16. Corrects August 7 patch.
277 * configure, conf.in: Rebuild with autoconf 2.12.
279 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
280 alignment is specified.
282 Add support for mips16 (16 bit MIPS implementation):
283 * config/tc-mips.c: Extensive additions for mips16 support, not
285 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
286 (insn_uses_reg): Change last parameter to an enum.
287 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
288 (md_relax_frag): Define as mips_relax_frag.
289 (mips_relax_frag): Declare.
290 (struct mips_cl_insn): Add use_extend and extend fields.
291 (tc_fix_adjustable): Define.
292 * config/obj-elf.h (S_GET_OTHER): Define.
293 (S_SET_OTHER): Define.
297 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
298 o and p operand specifiers.
302 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
303 (relax_segment, case rs_org): Move code inside braces. Move locals
304 target,after inside too.
305 (relax_segment, case rs_machine_dependent): Guts moved to ...
306 (relax_frag): New function.
307 Call md_prepare_relax_scan if defined.
308 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
313 * config/tc-mn10300.c (address_registers): Use '$' as register
314 prefix instead of '%'.
315 (data_registers, other_registers, md_assemble): Likewise.
317 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
318 (data_registers, other_registers, md_assemble): Likewise.
320 * config/tc-mn10300.c (md_assemble): Correctly determine the
321 correct location and type for each relocation.
322 (md_pcrel_from): Simplify.
326 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
327 converted into branch around branch.
331 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
332 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
334 * config/tc-mn10300.c (md_begin): Set linkrelax.
335 (md_assemble): Create fixups as needed.
336 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
341 * config/tc-d10v.c (parallel_ok): When automatically converting
342 serial ops to parallel, do not consider a branch as the first
348 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
352 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
354 (check_operand): Likewise.
359 * config/tc-d10v.c (parallel_ok): Branch and link instructions
361 (write_2_short): Call parallel_ok to check whether two short
362 instructions the user requested execute in parallel, can be
367 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
368 allow a branch and link in parallel with an exe instruction.
372 * doc/c-d10v.texi: Add info on @word modifier.
377 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
378 operands are assumed to be 32bits. Use "bits" field to hold the
379 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
380 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
383 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
384 of a MN10300_OPERAND_SPLIT operand by operand->shift.
386 * config/tc-mn10300.c (mn10300_insert_operand): Handle
387 MN10300_OPERAND_SPLIT.
391 * config/tc-mn10300.c (md_assemble): Insert operands into
392 the extension part of the instruction if necessary.
393 (mn10300_insert_operand): Accept pointer to extension word
394 argument. Make insn a pointer argument too. Return type
395 is now void. All callers changed.
399 * config/tc-mn10300.c (mn10300_insert_operand): Handle
400 repeated register operands.
404 * doc/as.texinfo: Added section on reporting bugs.
406 * config/tc-alpha.c: Change uses of void * to PTR. Change the
407 alpha_macro emit field to expect a const argument, and change the
408 arg field to be const. Fix some spacing to follow the GNU
413 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
414 (pca56) and 21264 (ev6) cpus.
415 (md_apply_fix): Private relocation types are now negative.
416 (alpha_force_relocation): Likewise.
417 (tc_gen_reloc): Likewise.
418 (emit_insn): Likewise.
419 (emit_ldXu): Do the right thing when the hardware can do byte insns.
420 (emit_stX): Likewise.
421 (emit_sextX): Likewise.
425 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
426 a symbol found in a reloc.
428 * symbols.c (resolve_symbol_value): Improve the error message if
429 an undefined symbol is used in an expression.
433 * doc/internals.texi: Rewrite, and add a lot of documentation.
434 * doc/Makefile.in (internals.info): New target.
439 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
440 against weak symbols.
444 * config/tc-v850.c (md_assemble): Don't lose for relaxable
447 * config/tc-v850.c (md_convert_frag): Make sure we insert the
448 fixup at the right address within the frag.
450 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
451 to an absolute value, instead increment it as needed.
453 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
454 * config/tc-v850.c: Fix some indention problems.
455 (md_relax_table): Define for D9->D99 branch displacement
457 (md_convert_frag): Do something useful instead of aborting.
458 (md_estimate_size_before_relax): Likewise.
459 (md_assemble): Note if the matching instruction has a relaxable
460 operand. If it does, allocate frag with frag_var and don't
467 * config/tc-d10v.h (md_cleanup): New function. This is needed to
468 write out any buffered instructions when a ".end" is found.
473 * read.c (read_a_source_file): New hook md_cleanup().
477 * write.c (fix_new_exp): Use make_expr_symbol to build an
478 expression symbol for a complex fixup.
483 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
484 hacks to improve parsing of complex hi, lo, zda, etc
486 (md_assemble): Don't demand and eat a trailing ')' after finding
487 a v850 relocation prefix. Sign extend the constant in a
488 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
490 (parse_cons_expression_v850): Don't eat a trailing ')' after
491 finding a v850 relocation prefix.
493 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
494 (TC_CONS_FIX_NEW): Likewise.
495 * config/tc-v850.c (parse_cons_expression_v850): New function.
496 (cons_fix_new_v850): Likewise.
498 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
503 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
504 for an unknown relocation type.
509 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
514 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
515 differently for movea & sst/sld insns.
521 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
527 * obj.h (struct format_ops): Add frob_file_after_relocs field.
528 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
529 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
530 frob_file_after_relocs field.
531 * config/obj-elf.c (elf_format_ops): Likewise.
532 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
537 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
538 specially if not ELF.
539 (md_apply_fix): Handle BFD_RELOC_64.
540 (tc_gen_reloc): Handle BFD_RELOC_64.
542 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
543 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
546 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
547 Don't check S_GET_OTHER.
551 * config/tc-mips.c (mips_ip): Accept an odd floating point
552 register with l.s or s.s.
554 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
556 (obj_aout_type): New static function.
560 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
565 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
566 and sdaoff expressions.
568 * write.c (fixup_segment): Don't add symbol value to addend if
570 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
573 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
574 (md_pcrel_from_section): Likewise.
575 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
579 * config/tc-v850.c (md_pcrel_from): Delete unused function.
580 (md_pcrel_from_section): New function.
581 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
586 * config/tc-mips.c (load_register): Add cast to offsetT when using
591 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
592 (elf_frob_file_after_relocs): ... here. New function.
593 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
594 * write.c (write_object_file): Call *frob_after_relocs after the
595 call to write_relocs.
597 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
599 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
600 symbol, as this messes up linking. Instead, expand the recursive call
601 inline and change up the appropriate bits to get the 0x8000 offset
606 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
607 two symbols in the same segment to be adjusted.
609 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
610 * configure: Rebuild.
614 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
616 (ppc_elf_validate_fix): Ditto.
617 (md_assemble): Ditto.
619 (ppc_pe_section): Ditto.
620 (ppc_frob_symbol): Ditto.
624 * config/mn10300.c (md_assemble): Pass an extra shift count
625 to mn10300_insert_operand based on the opcode format.
626 (mn10300_insert_operand): Accept and use extra shift count
629 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
630 formats rather than hard-coded constants.
632 * config/tc-mn10300.c (md_assemble): Format D5 instructions
633 are 7 bytes long. Write out instructions in big-endian format.
637 * config/tc-mn10300.c (md_assemble): Tweak further so
638 that all instructions are parsed correctly.
642 * as.h: Include libiberty.h.
643 (xmalloc, xrealloc): Don't declare.
644 * as.c: Don't include libiberty.h.
645 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
646 * config/tc-mips.c: Likewise.
647 * messages.c: Likewise.
648 (xstrerror): Don't declare.
653 * config/tc-mn10300.h (pre_defined_registers) Remove.
654 (system_registers, cc_names): Likewise.
655 (address_registers, data_registers, other_registers): New register
657 (register_name, system_register_name, cc_name): Remove.
658 (mn10300_reloc_prefix): Likewise.
659 (data_register_name): New function.
660 (address_register_name, other_register_name): Likewise.
661 (md_assemble): Rough cut at parsing operands. Remove lots of
663 (md_apply_fix3): Disable for now.
667 * config/tc-m68k.c (select_control_regs): New function, extracted
668 out of m68k_init_after_args.
669 (m68k_init_after_args): Use it.
670 (mri_chip): Use it here as well to update set of allowed control
675 * config/obj-elf.c (elf_begin): New function.
676 (obj_elf_section): Add the section symbol to the symbol table.
677 * config/obj-elf.h (obj_begin): Define.
678 (elf_begin): Declare.
679 * as.c (perform_an_assembly_pass): Call obj_begin if it is
684 * config/obj-coff.c (fixup_segment): Subtract the section address
685 from a PC relative reloc if TC_M68K.
689 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
690 .uaxword available even if not OBJ_ELF.
691 (md_atof): Remove unused local variable wordP.
695 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
696 for Matsushita MN10x00 support.
697 * configure.in: Recognize mn10x00-*-*
698 * configure: Rebuilt.
702 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
703 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
704 (load_expression): track clobbering of base reg before jmp/jsr.
705 (s_alpha_file): pass case_hack flags and source filename via
707 * tc-alpha.h (TC_CONS_FIX_NEW): define
711 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
712 * configure: Rebuild.
716 * read.c (s_macro): Warn if a macro has the same name as a
718 (s_space): In m68k MRI mode, align to a word boundary.
719 * macro.c (define_macro): Add namep parameter. Change all
721 * macro.h (define_macro): Update declaration.
723 * as.c (show_usage): Print bug report address.
724 (parse_args): Change version printing to match current GNU
726 * gasp.c (show_usage): Print bug report address.
727 (main): Change version printing to match current GNU standards.
729 * config/tc-m68k.c (init_table): Correct access control unit
732 * config/tc-alpha.c: Add some static function prototypes.
733 (alpha_macros): Move to top of file. Make static.
734 (alpha_num_macros): Move to top of file.
739 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
745 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
746 and just call sprintf_vma.
750 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
754 * read.c (float_cons): Call md_flush_pending_output if it is
760 * config/tc-d10v.c (md_operand): Created. Allows operands to
762 * config/tc-d10v.h (md_operand): Undefined.
767 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
768 (md_assemble): A fixup width of '3' means a 1 byte reloc.
772 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
773 reloc for the i960 for a reloc in the same section. This undoes
774 one of the two changes made Aug 19.
778 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
779 symbols to the position of the debugging information.
783 * expr.c (expr): Always use unsigned right shifts for >>.
787 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
788 slots when processing BL fixups.
790 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
791 on the first half of the instruction.
795 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
796 stabs expression, rather than giving an error.
798 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
803 * config/tc-mips.c (append_insn): Fill in the value for a constant
804 jump, rather than creating a reloc.
808 * config/tc-mips.c (append_insn): Don't swap an instruction which
809 sets a condition code with an instruction which uses a condition
811 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
814 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
815 (mips_ip): Give a better error message if the ISA level is wrong.
816 (md_parse_option): Recognize -mcpu=[v][r]5000.
820 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
822 (load_register): Ensure hi32 bits are not lost during lo32bit
823 processing. Fix shift offset that was overflowing into the next
824 instruction field. Add code to generate shorter sequences for
825 constants with a single contiguous seqeuence of ones.
830 * config/tc-d10v.c (d10v_dot_word): New function to support
831 "@word" with the word pseudo-op.
832 (md_apply_fix3): Cleanup and changes to support correct sizes
833 for 16 and 18-bit relocs.
838 * configure.in (sparc-*-aout): Set `em'.
839 * configure: Regenerated.
840 * config/te-sparcaout.h: New file.
841 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
842 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
843 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
844 (SPECIAL_CASE_{SETSW,SETX}): Define.
845 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
846 (md_begin): Delete setting of `target_big_endian'.
847 (output_insn): New function.
848 (md_assemble): Rewrite. Add `setx' support.
849 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
851 (md_atof): Add little endian support.
852 (md_number_to_chars): Likewise.
853 (md_apply_fix): Likewise.
854 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
855 (md_parse_option): Likewise.
856 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
860 * ecoff.c (ecoff_new_file): New function.
861 * ecoff.h (ecoff_new_file): Declare.
862 * config/obj-ecoff.h (obj_app_file): Define.
866 * config/tc-alpha.c (load_expression): Bias the .lit8 section
867 symbol by 32k so that our 16-bit signed offset can address the
872 * config/tc-mips.c (load_register): Remove unused variable tmp.
876 * config/tc-mips.c (load_register): Remove unnecessary code that
877 was causing the high 32bits of 64bit constants to be lost.
882 * config/tc-d10v.c: Added changes to support function
883 pointers and "@word" syntax.
889 * config/tc-v850.c: Remove commented out and #if 0'd code.
890 (v850_reloc_prefix): Provide prototype.
891 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
893 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
894 (md_assemble): Add default to case statement.
898 * config/tc-v850.c (md_assemble): Compute size of the instrction
901 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
906 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
907 routines to fetch/store the updated instruction from/to memory.
908 (v850_insert_operand): If the operand has a specialized insert
913 * config/tc-v850.c (reg_name_search): Align calling convention to
914 be like identical function found in tc-ppc.c.
915 (get_reloc): Removed.
916 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
917 (md_assemble): emit fixups.
918 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
920 (md_apply_fix3): handle fixups/relocs.
921 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
927 * configure.in (sh-*-elf*): New target.
928 * config/tc-sh.h (TARGET_ARCH): Define.
929 (WORKING_DOT_WORD): Define.
930 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
931 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
932 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
933 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
934 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
935 (SUB_SEGMENT_ALIGN): Likewise.
936 (RELOC_32): Don't define.
937 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
938 (target_big_endian): Declare if OBJ_ELF.
939 (TARGET_FORMAT): Define if OBJ_ELF.
940 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
942 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
943 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
944 (struct sh_count_relocs): Define.
945 (sh_count_relocs): New static function, broken out of
946 sh_frob_file. Add BFD_ASSEMBLER code.
947 (sh_frob_section): Likewise.
948 (sh_frob_file): Call sh_frob_section.
949 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
950 call section_symbol rather than seg_info (seg)->dot.
951 (md_section_align): Add OBJ_ELF version.
952 (SWITCH_TABLE_CONS): Define.
953 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
954 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
955 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
957 (struct reloc_map): Define if not BFD_ASSEMBLER.
958 (coff_reloc_map): Likewise.
959 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
960 (tc_gen_reloc): New function if BFD_ASSEMBLER.
961 * write.c (write_relocs): Ifdef out fx_where test which triggers
962 inappropriately for SH ELF.
963 (write_object_file): Call tc_frob_file_before_adjust and
964 obj_frob_file_before_adjust if they are defined.
966 * write.c (write_object_file): Use BFD_RELOC_16, not
967 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
969 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
975 * config/tc-d10v.c (find_opcode): Fix problem with calculating
976 branch sizes in across sections.
982 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
987 * config/tc-v850.c (md_begin): Deal with end of opcode
994 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
995 the wrong opcode for cases like st2w where there are many forms
996 of the same instruction.
1001 * expr.c (operand): If md_parse_name is defined, call it before
1002 calling symbol_find_or_make.
1003 * config/tc-ppc.h (md_parse_name): Define.
1004 (ppc_parse_name): Declare.
1005 * config/tc-ppc.c (reg_name_search): Add regs and regcount
1007 (register_name): Update call to reg_name_search.
1008 (cr_operand): New static variable.
1009 (cr_names): New static const array.
1010 (ppc_parse_name): New function.
1011 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
1012 cr_operand before calling expression.
1016 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
1017 hppa_gen_reloc_type call.
1022 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
1023 on addresses. Improved error messages.
1024 * doc/c-d10v.texi: Added docs for register pairs.
1028 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
1034 * ecoff.c (init_file): Initialize fMerge to 1.
1035 (add_file): Restore old file merging code, but only merge files if
1037 (ecoff_directive_loc): Clear fMerge field of current file.
1038 (ecoff_generate_asm_lineno): Likewise.
1043 * doc/c-d10v.texi: Fix typo.
1049 * config/tc-v850.c (md_assemble): Correct bit masking for
1050 hi and lo expressions.
1052 * config/tc-v850.c (md_assemble): Rough cut at demanding
1053 "ep" or "r30" in sst and sld instructions.
1054 (md_apply_fix3): Don't abort. Just warn that we don't
1057 * config/tc-v850.c (CC_NAME_CNT): Define.
1058 (cc_name): New function.
1059 (md_assemble): Handle V850_OPERAND_CC correctly.
1061 * config/tc-v850.c (md_assemble): Don't forget to initialize
1064 * config/tc-v850.c (reg_name_search): Generalize to search
1065 any given register table.
1066 (register_name): Pass appropriate table and size to reg_name_search.
1067 (system_register_name): New function.
1068 (SYSREG_NAME_CNT): Define.
1069 (md_assemble): Handle operands which are system registers.
1071 * config/tc-v850.c (md_assemble): If we find a register, but the
1072 opcode doesn't want a register, then we don't have a match.
1073 (md_assemble): Get size of the instruction from the opcode table.
1078 * configure.in: Set and substitute HLDENV.
1079 * configure: Rebuild.
1080 * Makefile.in (HLDENV): New variable.
1081 (as.new): Use $(HLDENV).
1083 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
1084 if an array dimension is not known.
1089 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
1090 * doc/c-d10v.texi: Cleanup.
1096 * config/tc-v850.c: New file.
1097 * config/tc-v850.h: New file.
1098 * configure (v850-*-elf): New target.
1099 * configure.in (v850-*-elf): New target.
1105 * doc/c-d10v.texi: New file.
1106 * doc/all.texi: Added D10V stuff.
1107 * doc/as.texinfo: Added D10V stuff.
1111 * config/tc-d10v.c: All references to defined symbols should
1112 now use the optimal instruction. .float and .double now work.
1117 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
1118 section address for the i960 as is done for the i386.
1122 * mpw-config.in: Add wildcards for config matching, add mips-*-*
1123 case, forward-include bfd/elf-bfd.h.
1128 * config/tc-d10v.c: Add additional information to the opcode
1129 table to help determinine which instructions can be done
1135 * config/tc-arm.c: Major changes to add Thumb support, with lots
1137 Reverted to INSN_SIZE macro, rather than insn_size variable.
1138 (insns): Added ARM "bx" instruction support.
1139 (tinsns): Added Thumb instruction definition structure.
1140 (arm_tops_hsh): Added hash structure for Thumb opcodes.
1141 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
1142 (opcode_select,s_arm,s_thumb,s_code): Added.
1143 (decode_shift): Allow upper-case RRX.
1144 (do_ldst): Simpler halfword support.
1145 (do_ldmstm): Improved.
1146 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
1147 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
1148 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
1149 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
1150 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
1152 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
1153 (md_parse_option): Add support for -mthumb.
1154 (md_show_usage): Updated to reflect new command line option.
1155 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
1156 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
1157 fragment information to be held.
1161 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
1162 non-null, but is a constant.
1163 (fix_new_arm): Call make_expr_symbol to make the expression symbol
1164 so that error reporting will work correctly.
1168 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
1169 against weak symbols.
1173 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
1174 (ppc_force_relocation): Declare if OBJ_XCOFF.
1175 * config/tc-ppc.c (ppc_force_relocation): New function if
1180 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
1185 * config/tc-d10v.c: Fix problem with relocs.
1190 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
1191 with nops if not in data_section or bss_section.
1195 Add support for openVMS/Alpha.
1196 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
1197 * config/obj-evax.c: New file.
1198 * config/obj-evax.h: New file.
1199 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
1201 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
1202 defined. Add case for bfd_target_evax_flavour.
1203 * config/vms-a-conf.h: New file.
1204 * conf-a-gas.com: New file.
1205 * configure.in: Add target alpha-*-*vms*.
1206 * configure: Rebuild.
1207 * makefile.vms: New file.
1208 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
1209 Don't call ffs on openVMS/Alpha.
1213 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
1214 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
1216 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
1217 (NEED_DECLARATION_SBRK): New macro.
1218 * configure, conf.in: Rebuild.
1219 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
1220 (strstr): Declare if NEED_DECLARATION_STRSTR.
1221 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
1225 * symbols.c (resolve_symbol_value): Handle addition or subtraction
1226 by a constant before entering the main switch. Reject attempts to
1227 apply an arithmetic function to non-absolute symbols, except for
1228 the special case of subtraction of two symbols in the same
1231 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
1232 not to more than a 16 byte boundary.
1234 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
1236 (tc_gen_reloc): If out of memory call as_fatal rather than
1237 assert. If no howto found, call as_bad_where rather than
1238 as_fatal. Change the error message slightly. Set howto to a
1239 non-NULL value in order to keep going.
1244 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
1249 * config/tc-h8300.c (get_specific): New operand "size" derived
1250 from ".b", ".w" and ".l" extensions. All callers changed. If
1251 the base instruction has no operands, then use the size to
1252 determine which specific instruction to use.
1256 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
1260 * config/tc-mips.c (md_section_align): Don't change addr if
1265 * config/tc-hppa.c: Revert yesterday's changes.
1270 * config/tc-d10v.c: Disable range checking on 16-bit values.
1275 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
1276 not just the ones that don't match.
1281 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
1286 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
1287 pre-cursor to adding Thumb support. Also added cpu_variant flag
1288 information to each of the asm_flg structures.
1289 (md_parse_option): Updated ARM7 parsing to allow 't' for
1290 thumb/halfword support, aswell as 'm' for long multiply.
1291 (md_show_usage): Updated help message.
1292 (md_assemble): Check that instruction flags are applicated to the
1293 current cpu variant.
1294 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
1295 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
1296 signextension instructions.
1297 (do_ldst): Generate halfword and signextension variants if
1298 mnemonic flags match.
1299 (ldst_extend): Do not allow shifts in the offset field of halfword
1300 or signextension instructions.
1301 (validate_offset_imm): Provide check on halfword and signextension
1303 (add_to_lit_pool): Merge identical literal pool values.
1307 * config/tc-hppa.c (selector_table): Add 'E' selector.
1308 (cons_fix_new_hppa): Don't coke on e_esel.
1309 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
1310 to help generate exception handling tables.
1311 (md_apply_fix): Don't try to apply fixups with an e_esel
1313 (hppa_fix_adjustable): Fixups with e_esel selectors
1318 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
1324 * config/tc-d10v.c: Added lots of error checking. Added hacks
1325 to support accumulator shifts.
1330 * symbols.c (S_SET_EXTERNAL): Let .weak override.
1331 (S_CLEAR_EXTERNAL): Likewise.
1332 (S_SET_WEAK): Remove error; just let .weak override.
1337 * config/tc-d10v.c (md_assemble): Now handles multiline
1342 * config/tc-d10v.c: Fix packaging bug. Added range checking.
1343 Added kludge for divs instruction. Fixed minor problem with
1344 multiple text sections.
1345 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1349 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
1350 addresses to be right-shifted by 2.
1354 * config/tc-d10v.c: Many changes to get relocs working.
1355 (register_name): No longer creates a symbol for register names.
1356 (pre_defined_registers): moved to opcodes/d10v-opc.c.
1357 (d10v_insert_operand): Now works correctly for either container.
1358 * config/tc-d10v.h (d10v_cleanup): Declare.
1363 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
1364 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
1368 * tc-alpha.c: Patches to track current minimum alignment to reduce
1369 the number of fragments created with frag_align.
1370 (alpha_current_align): New static variable.
1371 (s_alpha_text): Reset alignment to 0.
1372 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
1373 (s_alpha_stringer, s_alpha_space): New functions.
1374 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
1375 (alpha_cons_align): New function to replace both of them.
1376 (emit_insn): Only align if alpha_current_align is less than 2;
1377 reset alpha_current_align to 2.
1378 (s_alpha_gprel32): Likewise.
1379 (s_alpha_section): New function. Basically duplicate the other
1380 alpha section change hooks. Only define for ELF.
1381 (s_alpha_float_cons): Simplify alignment handling.
1382 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
1383 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
1384 Don't define the s_alpha_cons pseudo-ops. Do define
1385 s_alpha_stringer and s_alpha_space pseudo-ops.
1386 (alpha_align): Skip if less than current default alignment. Set
1388 * tc-alpha.h (md_flush_pending_output): Remove.
1389 (md_cons_align): Add.
1391 * tc-alpha.c: Add oodles of function description comments.
1392 (md_bignum_to_chars): Remove; there are no callers.
1393 (md_show_usage): Mention some more variants.
1398 * configure.in (sparc-*-sysv4*): New target.
1399 * configure: Rebuild.
1401 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
1402 uaxword to use s_uacons.
1403 (sparc_no_align_cons): New static variable.
1404 (s_uacons): New static function.
1405 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
1408 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
1410 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
1411 5 with regard to shared libraries.
1412 * configure: Rebuild.
1414 * config/tc-m68k.c (m68k_ip): Use the correct length when
1415 allocating space for the unsupported architecture error message.
1420 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
1424 * config/tc-d10v.c: New file.
1425 * config/tc-d10v.h: New file.
1426 * configure (d10v-*-elf): New target.
1427 * configure.in (d10v-*-elf): New target.
1432 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
1436 * config/tc-alpha.c (alpha_align): Change fill parameter
1437 to a pointer. Take NULL as 0 or nop depending on section. Change
1439 (s_alpha_align): Rename local variables.
1441 * doc/as.texinfo (.align): Document action of omitted
1446 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
1447 when an unsupported PC relative reloc is seen, rather than calling
1450 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
1451 Instead, check state when '#' comment is seen.
1455 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
1457 (tc_gen_reloc): If fixup was changed to be PC relative, change
1458 reloc type accordingly. Use name of reloc in error message.
1460 * as.h: Don't define const or volatile.
1461 * flonum.h: Don't define const.
1463 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
1464 if fx_pcrel is set. Correct setting the addend case in the
1465 OBJ_ELF case (from Andreas Schwab
1467 (md_show_usage): Correct -mfc5200 to -m5200.
1471 * doc/c-m68k.texi: Document -m5200 flag.
1472 * doc/as.texinfo: Likewise.
1474 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
1479 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
1480 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
1484 * Makefile.in (VERSION): Set to cygnus-2.7.1.
1486 * Released binutils 2.7.
1490 * config/tc-mips.c (mips_ip): Only perform range check when
1491 dealing with O_constant expressions.
1495 * m68k-parse.h (m68k_register): Add new coldfile control
1498 * config/tc-m68k.c (mcf5200_control_regs): New variable,
1499 array of control registers for the coldfire.
1500 (cpu_of_arch): Added mcf5200.
1501 (archs): Added mcf5200.
1502 (init_table): Add new control registers.
1503 (m68k_ip): Added support for new control registers.
1504 (m68k_init_after_args): Likewise.
1506 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
1510 * read.h (is_it_end_of_statement): Declare.
1511 * read.c (is_it_end_of_statement): Remove declaration.
1513 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
1515 (md_assemble): Fix handling of @l with an unsigned constant. Add
1516 default case to reloc switch.
1518 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
1521 * ecoff.c (last_lineno): New static variable.
1522 (add_procedure): Set last_lineno.
1523 (ecoff_directive_loc): Likewise.
1524 (ecoff_generate_asm_lineno): Likewise.
1525 (ecoff_fix_loc): New function.
1526 * ecoff.h (ecoff_fix_loc): Declare.
1527 * config/tc-mips.c (append_insn): When inserting nops, and using
1528 ECOFF debugging, call ecoff_fix_loc.
1532 * config/tc-h8300.c (build_bytes): If an operand type is
1533 marked as SRC_IN_DST retrieve it from the "destination" op.
1537 * configure.in (arm-*-riscix*): Set emulation to riscix.
1538 * configure: Rebuild.
1539 * config/te-riscix.h: New file to define TE_RISCIX.
1541 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
1545 * Makefile.in (config.status): Just run config.status as other
1550 * mpw-config.in (TARGET_OS): Add definition to conf.
1554 * config/tc-mips.c (append_insn): Parenthesize
1555 cop_interlocks expressions.
1559 * listing.c (listing_print): Close the listing file if it is not
1560 stdout. Close the other files opened for the listing.
1562 * config/tc-sparc.h (md_cons_align): Define.
1563 (sparc_cons_align): Declare.
1564 (HANDLE_ALIGN): Define.
1565 (sparc_handle_align): Declare.
1566 * config/tc-sparc.c (sparc_cons_align): New function.
1567 (sparc_handle_align): New function.
1568 * read.c (cons_worker): Call md_cons_align if it is defined.
1570 * as.h (struct frag): Add fr_file and fr_line fields.
1571 * frags.c (frag_new): Set fr_file and fr_line.
1572 (frag_var): Likewise.
1573 (frag_variant): Likewise.
1575 * as.h (struct frag): Remove unused align_mask and align_offset
1578 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
1583 * configure.in (mips-*-osf*): New target.
1584 * configure: Rebuild.
1586 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
1590 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
1591 between co-processor comparisons and branches for the VR4300.
1595 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
1596 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
1598 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
1599 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
1600 includedir): Use autoconf set values.
1605 * listing.c (listing_eject): Don't do anything if listing is 0.
1606 (listing_list): Likewise.
1607 (listing_source_line): Likewise.
1608 (listing_title): Don't save title if listing is 0.
1609 (listing_source_file): Check listing rather than listing_tail.
1611 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
1612 using shared libraries.
1613 * configure: Rebuild.
1617 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
1618 absolute expression if a relocation type was specified.
1622 * configure.in: Add support for *-*-rtems* configurations.
1623 * configure: Rebuild.
1627 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
1628 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
1629 opcodes library for alpha.
1630 * configure: Rebuild with autoconf 2.10.
1631 * config/tc-alpha.c: Substantial rewrite to add ELF support and
1632 use new opcode table.
1633 * config/tc-alpha.h (md_undefined_symbol): Don't define.
1634 (LOCAL_LABEL): Define differently if OBJ_ELF.
1635 (FAKE_LABEL_NAME): Define if OBJ_ELF.
1636 * config/alpha-opcode.h: Remove.
1637 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
1638 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
1639 include/opcode/alpha.h rather than config/alpha-opcode.h.
1643 * config/obj-aout.c (obj_emit_relocations): Give an error if the
1644 relocation symbol was not resolved.
1645 * config/obj-coff.c (do_relocs_for): Likewise.
1647 * write.c (adjust_reloc_syms): Refetch the symbol section after
1648 calling S_GET_VALUE, since it may have changed.
1650 * expr.c (struct expr_symbol_line): Define.
1651 (expr_symbol_lines): New static variable.
1652 (make_expr_symbol): Add entry to expr_symbol_lines.
1653 (expr_symbol_where): New function.
1654 * expr.h: Use extern on function declarations.
1655 (expr_symbol_where): Declare.
1656 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
1657 rather than printing the meaningless name of an expression
1662 * config/tc-i386.c (md_number_to_chars): Deleted.
1663 * config/tc-i386.h (md_number_to_chars): New macro.
1665 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
1667 (load_symbol_address, load_expression): Use build_mem.
1668 (build_operate): New function.
1669 (emit_addq_r): Use it.
1673 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
1677 * config/tc-vax.h (NOP_OPCODE): Define.
1681 * config/obj-vms.h (S_IS_COMMON): Define.
1682 (S_IS_LOCAL): Check for \002 as well as \001.
1683 (LONGWORD_ALIGNMENT): New macro.
1684 (SUB_SEGMENT_ALIGN): Use it.
1688 * config/vms-conf.h: Reconcile with conf.in.
1692 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
1695 * doc/as.texinfo: Document that any number of hex digits can
1698 * as.c (struct defsym_list): Define.
1699 (defsyms): New static variable.
1700 (parse_args): Just put --defsym arguments on defsyms list, rather
1702 (main): Define defsyms after output file is created.
1704 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
1707 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
1708 insn_error rather than calling check_absolute_expr.
1710 * as.c (emulation_name): Remove unused static variable.
1711 (default_emul_bfd_name): Add return NULL to avoid warning.
1712 * ecoff.c (ecoff_stab): Remove unused variables name and
1714 * frags.c (frag_new): Remove unused variable tmp.
1715 * hash.c (hash_grow): Parenthesize + within <<.
1716 (hash_print_statistics): Use %lu, not %d, to print unsigned
1718 * messages.c: Include "libiberty.h".
1719 (fprint_value): Add cast to avoid printf warning.
1720 (sprint_value): Likewise.
1721 * read.c: Include "ecoff.h".
1722 (emit_expr): Add casts to avoid printf warnings.
1723 * read.h: Use extern for function declarations.
1724 (pop_insert): Declare.
1725 * stabs.c: Include "ecoff.h".
1726 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
1727 former_last_fragP, and new_fragP.
1728 * subsegs.h (subsegs_print_statistics): Declare.
1729 * symbols.c (debug_verify_symchain): Change macro to discard
1731 * write.c (dump_section_relocs): Likewise.
1732 * write.h: Use extern for function declarations.
1733 (write_print_statistics): Declare.
1734 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
1736 * config/e-mipself.c (mipself_bfd_name): Likewise.
1737 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
1739 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
1740 emit relocations against external symbols.
1742 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
1743 message if bfd_reloc_type_lookup fails, rather than calling
1746 * config/tc-alpha.c (alpha_force_relocation): Add
1747 BFD_RELOC_12_PCREL to switch.
1751 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
1752 TE_PE (Lfoo, not .Lfoo).
1756 * read.c (s_fill): Don't warn about a zero repeat count.
1758 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
1759 coprocessor register.
1761 * config/tc-i386.c (md_assemble): When checking the size of a
1762 register to set the size of an instruction, do a bitwise and with
1763 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
1768 * config/tc-h8300.c (parse_reg): Tweak error messages.
1769 (build_bytes): Likewise.
1770 (skip_colonthing): Handle :32 suffix.
1771 (get_specific): Promote L_24 to L_32 if it makes a match.
1772 Don't always promote L_8 to L_16.
1773 (do_a_fix_imm): Clean up L_32 and L_24 handling.
1775 * config/tc-h8300.c (Smode): New variable.
1776 (h8300hmode): Turn off Hmode.
1777 (h8300smode): New function. Turn on Smode and Hmode.
1778 (md_pseudo_table): New ".h8300s" pseudo-op.
1779 (parse_reg): Handle "exr" register.
1780 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
1781 Handle "mach" and "machl" operands for ldmac.
1782 (get_specific): Handle "stm.l" and "ldm.l".
1783 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
1784 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
1789 * doc/as.texinfo: Reorder chapter of machine dependent options so
1790 that it is sorted by chip name.
1792 * doc/as.texinfo: Use consistant spelling of Vax.
1793 * doc/c-vax.texi: Likewise.
1797 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
1799 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
1800 (md_apply_fix): Likewise.
1801 (pa_try): New function.
1802 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
1806 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
1807 (pa_level): New function.
1811 * listing.c (listing_newline): Don't do anything if listing is 0.
1815 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
1816 symbol_create, not symbol_new.
1820 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
1821 absolute symbol unless TC_FORCE_RELOCATION returns true.
1823 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
1825 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
1826 the symbol to a pointer. Don't set previous_file_symbol.
1827 Simplify symbol list rearrangement.
1828 (coff_frob_symbol): Don't do anything with C_FILE symbols.
1829 (coff_adjust_symtab): Don't check previous_file_symbol.
1833 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
1835 (md_pseudo_table): Add ppc_elf_lcomm.
1839 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
1840 d0{24:1},d0' works without an immediate prefix on the bit numbers.
1841 (md_begin): Add digits to alt_notend_table.
1842 (md_parse_option): Make s a const pointer.
1844 * config/tc-sparc.c (md_pseudo_table): Add "empty".
1845 (s_empty): New static function.
1847 * config/obj-coff.c (struct filename_list): Only define if not
1849 (filename_list_head, filename_list_tail): Likewise.
1850 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
1851 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
1852 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
1853 have a section of N_ABS rather than N_DEBUG. If we do a merge,
1854 remove the new symbol from the list.
1855 (obj_coff_endef, both versions): Call tag_insert even if there is
1856 an old symbol with the same name, if the old symbol does not
1858 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
1859 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
1860 symbol that will be moved to the end of the symbol list.
1861 (coff_adjust_section_syms): Always call section_symbol for .text,
1863 (coff_frob_section): Likewise. Also, remove unused variable
1866 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
1867 manipulating frags directly.
1868 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
1871 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
1872 whether the next character is '*' before checking whether we are
1873 at the start of a line. Permit LINE_COMMENT_START to start a
1874 comment in state 1 (seen some whitespace) as well, to match the
1877 * gasp.c (do_align): Permit a fill value for .align.
1881 * read.c (next_char_of_string): Warn if a newline is seen in the
1882 middle of a string. Call bump_line_counters when appropriate.
1886 * symbols.c (colon): Use LOCAL_LABEL.
1890 * Makefile.in (install): Don't check to see if tooldir exists.
1891 Make $(tooldir) and $(tooldir)/bin.
1895 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
1900 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
1901 insn_error. Don't put the string "ERROR" in insn_error. Set
1902 insn_error rather than calling as_warn for an unsupported opcode.
1906 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
1907 before permitting -64.
1908 * output-file.c (output_file_create): Remove duplicate
1913 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
1914 section rather than a .reginfo section.
1915 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
1918 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
1919 bit value with the high bit set.
1923 * read.c (s_lcomm): Set section flags for .sbss section.
1925 * config/tc-mips.c (mips_64): New static variable.
1926 (mips_target_format): If mips_64, return elf64 targets rather than
1928 (md_longopts): Add "32" and "64".
1929 (md_parse_option): Handle -32 and -64.
1930 (md_show_usage): Mention -32 and -64.
1931 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
1936 * config/tc-ppc.c (comment_chars): Make '!' a comment character
1937 for Solaris compatibility.
1939 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
1940 .stabd with 4 arguments into a .stabn.
1944 * config/tc-mips.c (macro): When passing X_add_number to
1945 macro_build, cast it to int first.
1949 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
1950 created by emit_expr.
1952 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
1953 as_perror rather than assert.
1957 * config/tc-mips.c (mips_ip): Mark sections created to hold
1958 floating point information as read only.
1962 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
1966 * read.c (potable): Add .skip as a synonym for .space.
1968 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
1969 4 arguments, providing the 4th argument is 0, to allow
1970 compatibility with the Solaris assembler.
1974 * config/tc-sh.h (struct sh_segment_info_type): Define.
1975 (TC_SEGMENT_INFO_TYPE): Define.
1976 (sh_frob_label): Declare.
1977 (tc_frob_label): Define.
1978 (sh_flush_pending_output): Declare.
1979 (md_flush_pending_output): Define.
1980 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
1981 reloc before the instruction if necessary.
1982 (sh_frob_label): New function.
1983 (sh_flush_pending_output): New function.
1984 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
1985 when looking for the reloc for the target of .uses.
1986 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
1987 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
1989 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
1990 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
1991 the absolute symbol.
1993 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
1994 TC_SEGMENT_INFO_TYPE is defined.
1998 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
1999 matches the register size.
2003 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
2005 (do_relocs_for): Likewise.
2006 (fixup_segment): Don't just quit if linkrelax is set. Try to
2007 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
2012 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
2013 relative relocations with -mrelocatable. Also allow unfixed
2014 relocs in .ex_shared.
2018 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
2019 non-zero before assuming this is a long file name.
2020 (w_strings): Likewise.
2021 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
2023 * config/obj-coff.c (w_strings): Move declaration of i inside
2024 #ifdef block which uses it.
2028 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
2029 "relaxmode". Output relocs which identify various relaxing
2030 possibilities for mov.[bwl] instructions.
2031 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
2035 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
2036 (MD_APPLY_FIX3): Define.
2037 * config/tc-arm.c (my_get_expression): Only watch for bad segments
2039 (md_apply_fix3): Renamed from md_apply_fix.
2040 If pcrel reloc and symbol is in different section, undo effects
2045 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
2046 any reloc with an LR% or RR% field selector for SOM.
2050 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
2055 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
2057 (yank_symbols): Likewise.
2061 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
2062 BFD_ASSEMBLER, add name field.
2063 * config/obj-coff.c: Include "libiberty.h".
2064 (coff_header_append): Handle long section names.
2065 (crawl_symbols): Just use the name field for the symbol name,
2066 without worrying about null byte termination.
2067 (w_strings): Handle long section names.
2068 (write_object_file): Likewise. Also, use the name field, rather
2070 (obj_coff_add_segment): Permit long section names.
2071 (obj_coff_init_stab_section): Use the name field, rather than
2073 (adjust_stab_section): Likewise.
2074 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
2076 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
2077 (mem_fmt): Likewise.
2078 (md_apply_fix): Don't check fx_im_disp field.
2082 * configure.in: Add * after sparc*-*-vxworks.
2083 * configure: Rebuild.
2085 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
2086 define it to comment_chars. Use tc_comment_chars rather than
2088 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
2089 * config/tc-m68k.h (tc_comment_chars): Define.
2090 (m68k_comment_chars): Declare.
2091 * config/tc-m68k.c (m68k_comment_chars): Rename from
2092 comment_chars. Change into a pointer rather than an array.
2093 (md_longopts): Add "bitwise-or".
2094 (md_parse_option): Handle OPTION_BITWISE_OR.
2095 (md_show_usage): Mention --bitwise-or.
2096 * doc/c-m68k.texi: Document --bitwise-or.
2100 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
2103 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
2104 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
2105 correct handling of input line pointer, and call
2106 demand_empty_rest_of_line.
2110 * config/tc-sparc.c (in_bitfield_range): New static function.
2111 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
2112 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
2113 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
2114 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
2115 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
2119 * config/obj-coff.c: BFD_ASSEMBLER:
2120 (coff_last_bf): New static variable.
2121 (coff_frob_symbol): Set endndx of a .bf symbol.
2123 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
2124 (last_bfP): New static variable.
2125 (yank_symbols): Set endndx of a .bf symbol.
2129 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
2130 and -s options. Add -be for good measure.
2134 * read.c (s_space): Support non-constant fill value. Handle fill
2135 value correctly for a size other than 1.
2139 * config/tc-arm.c (my_get_float_expression): Update call to
2140 gen_to_words, X_PRECISION changed from 6 to 5.
2144 * config/tc-ppc.c (register_name,reg_name_search): Move register
2145 name lookup from PE specific code to all targets. Add support for
2146 -mregnames/-mno-regnames to control whether register names are
2148 (md_assemble): Call register_name for all platforms.
2149 (md_parse_option): Add support for -mregnames/-mno-regnames.
2151 * configure.in (powerpcle*-*-solaris): Add support.
2152 (powerpc*-*-linux): Ditto.
2153 * configure: Regenerate.
2155 * config/ppc-sol.mt: New config file for PowerPC Solaris.
2159 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
2160 be associated with a single %lo reloc.
2162 * config/tc-mips.c (load_address): Cast X_add_number to valueT
2163 before comparing against MAX_GPREL_OFFSET, so that negative
2164 numbers are handled correctly.
2169 * config/tc-sparc.c (last_insn): New static variable.
2170 (md_assemble): Warn about putting floating point branches in a
2171 delay slot. If architecture is less than v9, insert NOP
2172 instructions between floating point instructions and floating
2173 point branches. (The SunOS assembler does both these operations.)
2174 Save the last instruction opcode.
2175 (sparc_ip): Add pinsn parameter. Change caller.
2177 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
2178 for byte jump to next instruction to skip empty frags.
2182 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
2183 for addq with a 16 bit signed value, just emit a lda instruction
2188 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
2193 * configure.in: Permit --enable-shared to specify a list of
2195 * configure: Rebuild.
2199 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
2200 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
2204 * config/atof-ieee.c: Fix handling of denormalized extended
2205 precision numbers and overflow/underflow detection.
2206 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
2207 not include the 16 bit gap in the m68k extended precision format.
2211 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
2213 * configure: Rebuild.
2217 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
2218 be accessed using the $gp register.
2219 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
2223 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
2224 bfd_mach_sparc_sparclet if current_architecture is sparclet.
2228 * read.c (get_line_sb): Bump line counters based on
2229 input_line_pointer[-1], not *input_line_pointer. Don't bother to
2230 call LISTING_NEWLINE.
2231 (s_macro): Don't call demand_empty_rest_of_line.
2232 * app.c (do_scrub_chars): When handling C style comments, unget
2237 * read.h (enum linkonce_type): Define.
2238 (s_linkonce): Declare.
2239 * read.c (potable): Add "linkonce".
2240 (s_linkonce): New function.
2241 * subsegs.h (segment_info_type): Add linkonce field to
2242 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
2243 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
2244 (obj_coff_pe_handle_link_once): Declare if TE_PE.
2245 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
2247 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
2248 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
2249 entry based on the linkonce field in segment_info.
2250 * doc/as.texinfo: Document .linkonce.
2254 * doc/as.1: Changed to be recognized by catman -w on Solaris.
2258 * stabs.c (s_stab_generic): Call the listing functions before
2259 doing the rest of the processing, which may involve freeing the
2260 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
2261 SEPARATE_STAB_SECTIONS case.
2263 * config/tc-hppa.c: Remove nested comment.
2264 (tc_gen_reloc): Move label done inside the ifdef in which it is
2266 (md_apply_fix): Pass pointers to correct types to libhppa.h
2267 functions. Always return a value.
2269 * config/tc-mips.h (tc_frob_file): Define.
2270 (mips_frob_file): Declare.
2271 * config/tc-mips.c (struct mips_hi_fixup): Define.
2272 (mips_hi_fixup_list): New static variable.
2273 (imm_unmatched_hi): New static variable.
2274 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
2275 offset_reloc. Pass imm_unmatched_hi to append_insn.
2276 (append_insn): Add unmatched_hi parameter. If it is set, add the
2277 new fixup to mips_hi_fixup_list. Change all callers.
2278 (mips_ip): Set imm_unmatched_hi when appropriate.
2279 (mips_frob_file): New function.
2283 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
2284 * configure: Regenerated.
2288 * as.c (main): Call bfd_set_error_program_name.
2292 * as.h (strdup): Don't declare.
2293 * stabs.c: Include libiberty.h
2294 (get_stab_string_offset): Use xstrdup rather than strdup.
2295 (s_stab_generic): Likewise.
2296 * as.c (parse_args): Likewise.
2297 * read.c (s_mri_sect): Likewise.
2299 * gasp.c (change_base): Recognize \(...) construct documented to
2300 pass through enclosed characters literally through to the output.
2301 (process_assigns): Likewise. Also, be more careful to avoid
2302 looking past the end of the buffer.
2306 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
2311 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
2315 * config/alpha-opcode.h: Added cvtst instruction.
2319 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
2320 string. Extract the addend from the result of expression.
2324 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
2325 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
2326 mode, change the state to 3 rather than 1.
2330 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
2331 correspond to 11 March change.
2335 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
2336 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
2340 * configure: Rebuild with autoconf 2.8.
2344 * config/atof-ieee.c (gen_to_words): Improve handling of
2345 X_PRECISION numbers. Based on patches from Andreas Schwab
2350 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
2352 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
2353 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
2354 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
2355 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
2360 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
2364 * config/tc-hppa.c (SUBSEG_MILLI): Define.
2365 (pa_def_subspaces): Add $MILLICODE$.
2366 (pa_spaces_begin): Set section flags for $MILLICODE$.
2370 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
2371 not SHT_NOBITS. Don't tamper with flags based on type if a
2372 special section was found (revert Feb 29 change).
2374 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
2375 if aligning to a longword boundary or greater.
2379 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
2383 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
2384 * configure: Regenerated.
2388 * config/tc-hppa.c: Add default definitions for R_N0SEL and
2389 R_N1SEL since they're not defined for old versions of hpux.
2391 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
2392 Set "sym_ptr_ptr" and "addend" fields to dummy values for
2393 R_N0SEL and R_N1SEL.
2397 * flonum-konst.c: Add two more constants for 1e+-2048 and
2398 1e+-4096, and correct the other constants.
2400 * symbols.c (resolve_symbol_value): Handle O_logical_not.
2404 * config/obj-elf.c (obj_elf_section): Allow predefined section
2405 types to set the nobits type. Avoid a shadowed declaration.
2409 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
2410 reduce relocs using e_nlrsel field selectors.
2412 * write.c (fix_new_exp): Don't use #elif. Some compilers
2415 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
2417 (pa_chk_field_selector): Handle new field selectors for SOM.
2421 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
2422 * configure: Rebuild.
2423 * config/te-linux.h (LOCAL_LABELS_FB): Define.
2424 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
2426 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
2431 * mpw-make.sed: Update to handle shared library support.
2435 * config/tc-sparc.c (sparc_ip): Print all architectures that support
2436 the insn on mismatch.
2440 * configure.in: Add support for a29-coff.
2441 * configure: Rebuild.
2445 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
2446 of the section when looking for the R_SH_USES fixup, because the
2447 frag addresses have not yet been adjusted.
2449 * gdbinit.in: Set a breakpoint on as_warn_where.
2451 * config/tc-mips.c (macro): Add missing arguments to macro_build
2456 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
2457 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
2461 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
2465 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
2466 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
2467 Recognize [uU] format args as sparclet cpregs.
2471 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
2476 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
2477 well as S_IS_EXTERNAL.
2478 (tc_fix_adjustable): Likewise.
2479 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
2480 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
2482 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
2483 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
2487 * config/tc-sparc.c (max_architecture): Change to sparclite for
2489 (default_compatible): Delete.
2490 (sparc_ffs): New function.
2491 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
2492 (sparc_ip): Rewrite architecture match and bump logic.
2496 * configure.in: Check for 'do not mix' from native linker before
2497 trying to use -rpath.
2498 * configure: Rebuild.
2502 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2503 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2504 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
2505 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
2507 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
2508 to the symbol lnnoptr field, to get the correct file offset.
2512 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
2513 symbols that are not function, file, or section symbols to be
2518 * configure: Set and substitute RPATH_ENVVAR.
2519 * configure: Rebuild.
2520 * Makefile.in (RPATH_ENVVAR): New variable.
2521 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
2524 doesn't need opcodes. If configuring shared, opcodes needs bfd.
2525 * configure: Rebuild.
2529 * read.c (s_mri_sect): Don't return '\0' in type. Set all
2530 appropriate flags in BFD section.
2532 * configure.in (m68k-*-psos*): New target.
2533 * configure: Rebuild.
2534 * config/te-psos.h: New file.
2535 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
2541 * configure.in: Remove duplicate setting of cpu_type. Check
2542 whether opcodes library is required for on all targets, not just
2544 * configure: Rebuild.
2546 * config/tc-mips.c (mips_big_got): New static variable.
2547 (s_extern): Don't declare.
2548 (reg_needs_delay): New static function.
2549 (macro_build): Permit GOT/CALL_HI/LO relocs.
2550 (macro_build_lui): If place is not NULL, use the number in the
2552 (load_address): Handle mips_big_got case.
2553 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
2555 (OPTION_XGOT): Define.
2556 (md_longopts): Add "xgot" if OBJ_ELF.
2557 (md_parse_option): Handle -xgot.
2558 (md_show_usage): Mention -xgot.
2559 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
2560 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
2564 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
2565 the byte relocation, point it to the low byte of the word.
2569 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
2570 * configure: Rebuild.
2574 * configure.in: Recognize any sparc* cpu.
2575 * configure: Regenerated.
2579 * read.c (potable): Add "mri" and ".mri".
2580 (s_mri): New function.
2581 * read.h (s_mri): Declare.
2582 * app.c (scrub_m68k_mri): New static variable.
2583 (mri_pseudo): New static variable.
2584 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
2585 flag_m68k_mri. Initialize scrub_m68k_mri.
2586 (mri_state, mri_last_ch): New static variables.
2587 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
2589 (app_push): Save new fields.
2590 (app_pop): Restore new fields.
2591 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
2592 If TC_M68K, use a trivial state machine to look for occurrences of
2593 the .mri pseudo-op, and change the mode appropriately.
2594 * as.h (do_scrub_begin): Update prototype.
2595 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
2597 * config/tc-m68k.c (reg_prefix_optional_seen): New static
2599 (m68k_mri_mode_change): New function.
2600 (md_parse_option): Set reg_prefix_optional_seen.
2601 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
2602 (MRI_MODE_CHANGE): Define.
2603 * doc/as.texinfo: Document .mri pseudo-op.
2605 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
2610 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
2611 * config/m68k-parse.h (enum pic_relocation): Define.
2612 (struct m68k_exp): Add pic_reloc field.
2613 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
2614 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
2615 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
2617 * config/tc-m68k.c: Delete definition of NO_RELOC.
2618 (struct m68k_it): Add pic_reloc field.
2619 (add_fix): Copy over pic_reloc field.
2620 (md_pseudo_table): Interpret .align parameter as byte count.
2621 (mote_pseudo_table): Likewise.
2622 (tc_m68k_fix_adjustable): New function.
2623 (get_reloc_code): New function.
2624 (md_assemble): Use it as last argument to fix_new_exp.
2625 (md_apply_fix_2): For a relocation against a symbol don't put the
2626 addend into the data.
2627 (tc_gen_reloc): Different addend computation for OBJ_ELF.
2628 (m68k_ip): Don't relax an operand that requires pic relocation.
2629 (md_begin): Align .text, .data and .bss on 4 byte boundary by
2631 * write.c (fixup_segment): Don't add symbol value to addend if
2632 TC_M68K and OBJ_ELF.
2633 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
2634 (motorola_operand): Add rule for `(zapc, EXPR)'.
2638 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
2639 *really* is permissible before the comma.
2643 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
2644 inserting nop instructions.
2648 * config/te-aux.h: Change include of aux.h to aux-coff.h.
2652 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
2653 class setting for a CALLNAME symbol in COFF.
2655 * read.c (potable): Pass negative numbers for new .balign[wl] and
2656 .p2align[wl] pseudo-ops.
2657 (s_align_bytes): Treat a negative argument as specifying the fill
2659 (s_align_ptwo): Likewise.
2663 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
2664 (do_align): Take new len parameter. Change all callers. Pass it
2666 (s_align_bytes): Arg now indicates the length of the fill pattern.
2667 (s_align_ptwo): Likewise.
2668 * config/obj-coff.c (write_object_file): Pass length to
2670 * config/tc-i386.h (md_do_align): Take new len parameter.
2671 * config/tc-m88k.h (md_do_align): Likewise.
2672 * config/tc-m88k.c (m88k_do_align): Likewise.
2673 * config/tc-sh.h (md_do_align): Likewise.
2674 * config/tc-sh.c (sh_do_align): Likewise.
2675 * doc/as.texinfo: Document new pseudo-ops.
2677 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
2678 done in cvt_frag_to_fill.
2680 * config/tc-sh.h (sh_do_align): Declare.
2681 (md_do_align): Define.
2682 * config/tc-sh.c (sh_do_align): New function.
2684 * ecoff.c (ecoff_build_lineno): Don't try to store the address
2685 difference if the next address is before the current one.
2687 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
2688 (archs): Initialize new field.
2689 (m68k_ip): Don't list alias names when listing CPUs which support
2692 * as.c (main): Call parse_args before read_begin.
2693 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
2694 front of generated pseudo-ops.
2695 * read.c (potable): Ignore "name".
2696 (s_app_file): Permit a single quote after the string, since one
2697 may appear in m68k MRI mode.
2699 * configure.in: Check for --enable-shared. If linking against
2700 shared BFD and opcodes, fix library name on SunOS, and try to set
2702 * configure: Rebuild.
2706 * as.h (flag_m68k_mri): Declare.
2707 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
2708 * Many files: For MRI syntax that is specific to the m68k MRI
2709 assembler, check flag_m68k_mri rather than flag_mri or
2710 MRI_MODE_NEEDS_PSEUDO_DOT.
2714 * config/tc-i960.c (ARCH_HX): Define.
2716 (targ_has_sfr): Handle ARCH_HX.
2717 (targ_has_iclass): Handle ARCH_HX.
2718 (tc_coff_fix2rtype): Add return 0 to avoid warning.
2719 (tc_headers_hook): If the architecture was specified explicitly,
2720 use it when setting the flags. Set the extern variable coff_flags
2721 rather than headers->filehdr.f_flags, since the latter is set
2722 unconditionally in obj-coff.c.
2723 (i960_handle_align): Remove unused variable fixp.
2725 Support for building bfd and opcodes as shared libraries, based on
2727 * configure.in: Set OPCODES and BFD to search directories.
2728 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
2729 * configure: Rebuild.
2730 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
2731 (LIBDEPS): New variable.
2732 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
2733 $(HLDFLAGS) in link.
2734 (check): Set LD_LIBRARY_PATH in the environment.
2738 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
2739 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
2740 read-only, not read/write.
2744 * Makefile.in (INSTALL_XFORM): Remove -e.
2748 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
2749 than comparing S_GET_SEGMENT to undefined_section.
2750 (write_object_file): Skip symbols which were equated to an
2751 undefined or common symbol.
2752 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
2753 S_IS_COMMON rather than comparing S_GET_SEGMENT to
2755 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
2756 or S_IS_COMMON call S_GET_VALUE.
2757 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
2758 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
2759 to an undefined or common symbol, convert the reloc to be against
2761 (obj_crawl_symbol_chain): Skip symbols which were equated to an
2762 undefined or common symbol.
2763 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
2764 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
2765 to an undefined or common symbol, convert the reloc to be against
2767 (obj_crawl_symbol_chain): Skip symbols which were equated to an
2768 undefined or common symbol.
2769 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
2770 S_IS_COMMON rather than comparing S_GET_SEGMENT to
2772 (yank_symbols): Skip symbols which were equated to an undefined or
2777 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
2778 * config/obj-bout.h (S_IS_LOCAL): Likewise.
2780 * configure.in: Make sure we only add m68k-parse.o to
2781 ${extra_objects} once, no matter how many m68k targets have been
2783 * configure: Rebuild.
2787 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
2788 * configure: Rebuild.
2792 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
2793 * doc/c-m68k.texi: Document .extend and .ldouble.
2795 * configure.in (m68*-apple-aux*): New target.
2796 * config/te-aux.h: New file.
2797 * config/obj-coff.c (compare_external_relocs): New static function
2799 (do_relocs_for): Sort relocs if TE_AUX.
2800 (fixup_segment): If TE_AUX, store common symbol value in segment.
2801 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
2805 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
2807 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
2809 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
2810 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
2812 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
2813 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
2814 an undefined symbol and the size of a function symbol.
2815 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
2816 for all common symbols.
2820 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
2823 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
2828 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
2831 * ecoff.c (localsym_t): Add addend field.
2832 (add_ecoff_symbol): Add addend argument. Change all callers.
2833 (coff_sym_value): Make static.
2834 (coff_sym_addend): New static variable.
2835 (ecoff_directive_def): Initialize coff_sym_addend.
2836 (ecoff_directive_val): Accept symbol + constant.
2837 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
2838 (ecoff_build_symbols): Include the addend in the symbol value.
2842 * config/tc-ppc.c (md_assemble): Ignore overflow on
2843 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
2847 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
2848 bytes in size, so offset appropriately in big endian mode when
2849 writing the bottom 2 bytes.
2853 * config/tc-sparc.c (default_compatible): New static local.
2854 (md_begin): Initialize it. Rewrite warn_on_bump handling.
2855 (sparc_ip): If no architecture or -bump specified, don't mark as
2856 mismatched those in default_compatible.
2861 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
2862 * configure: Rebuild.
2863 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
2864 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
2865 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
2866 (sco_id): Declare if SCO_ELF.
2870 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
2871 ({max,warn_after}_architecture): New static locals.
2872 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
2873 If both architecture and -bump requested, set max_architecture to max.
2874 (sparc_md_end): Simplify.
2875 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
2876 Rewrite code to bump architecture and check for conflicts.
2877 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
2878 with Solaris assembler.
2879 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
2880 (md_show_usage): Update.
2884 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
2888 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
2889 #ifdef sparcv9 when choosing value.
2892 (sparc_md_end): Declare.
2893 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
2894 (initial_architecture): New static local.
2895 (can_bump_v9_p): Likewise.
2896 (NO_V9): Delete all occurrences.
2897 (sparc_md_end): New function.
2898 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
2899 Don't bump architecture to v9 unless can_bump_v9_p set.
2900 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
2901 * configure.in (sparc64 target cpu): Don't set obj_format here.
2902 (SPARC_V9): Renamed from sparcv9.
2903 (sparc64-*-elf*): Define SPARC_ARCH64.
2904 * configure: Regenerated.
2905 * acconfig.h (SPARC_V9): Renamed from sparcv9.
2906 (SPARC_ARCH64): Add.
2907 * conf.in: Regenerated.
2908 * config/vmsconf.h: Update.
2912 * config/tc-mips.c (load_register): Optimise "dli" loads.
2913 (md_show_usage): add "-mcpu=vr4100" to help text.
2917 * symbols.c (resolve_symbol_value): If a symbol is equated to an
2918 undefined symbol, preserve the X_op of O_symbol.
2919 (S_GET_VALUE): Fix check to permit this case.
2920 * write.c (write_relocs): If a reloc is against an undefined
2921 symbol equated to another symbol, change the reloc to be against
2923 * config/obj-coff.c (do_relocs_for): Likewise.
2925 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
2930 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
2931 computer is PowerPC.
2933 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
2935 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
2936 mentioned in the eabi.
2940 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
2941 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
2945 * config/tc-ppc.c (mapping): Add more relocation suffixes.
2949 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
2950 as a section it is ok to have unadorned -mrelocatable pointers in.
2954 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
2959 * config/obj-elf.c (obj_elf_section): Add hooks so machine
2960 dependent section attributes can be handled.
2962 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
2963 macros to add support for exclude section flag and ordered section
2966 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
2967 functions to add support for exclude section flag and ordered
2972 * subsegs.c (section_symbol): Don't try to look up the section
2973 symbol in the hash table. It should be possible to have a symbol
2974 with the same name as a section, but no connection to it.
2976 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
2979 * expr.c (operand): Skip whitespace after a close parenthesis.
2984 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
2985 as_warn_where for relocation overflow.
2986 (parse_reg): Accept register name only if next character is
2991 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
2992 addresses for static variables.
2996 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
2997 and xgot give the old toc16 relocation.
2998 (md_apply_fix3): Support all GOT relocations.
3002 * config/tc-mips.c (load_address): Correctly handle a constant in
3007 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
3008 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
3009 (build_Mbytes): Handle REG_B.
3013 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
3017 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
3021 * config/obj-coff.c (write_object_file): Set the s_align field to
3022 the number of bytes, rather than to the power of 2.
3026 * Makefile.in (DISTCLEAN_HERE): New variable.
3027 (distclean): Use it.
3028 (maintainer-clean): Depend upon clean-here rather than clean,
3029 distclean, and clean-info. Run make maintainer-clean in doc.
3030 Remove files listed in DISTCLEAN_HERE.
3031 * doc/Makefile.in (maintainer-clean realclean): Split out from
3032 distclean. Depend upon clean-info and distclean.
3036 * mac-as.r: Fix copyright and version strings.
3037 (cfrg): Use PROG_NAME instead of literal name.
3041 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
3043 * config/tc-a29k.h (tc_unrecognized_line): Define.
3044 * config/tc-a29k.c (a29k_unrecognized_line): New function.
3045 (md_operand): Handle a29k style local dollar labels.
3049 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
3053 * read.c (s_fill): If md_flush_pending_output is defined, call
3058 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
3059 Treat rs_align_code like rs_align.
3063 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
3064 converting an absolute address into a PC-relative one.
3068 * mpw-config.in: Don't always use te-generic.h for emulation.
3069 (powerpc-apple-macos): Use emulation te-macos.h.
3071 (mips-gm-magic): New configuration.
3073 * mpw-make.sed (install, install-only): Edit in Mac-specific
3078 * configure.in: Improve message about unsupported ELF targets.
3079 * configure: Rebuild.
3081 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
3082 pattern is zero. From Manfred Hollstein.
3086 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
3087 the various idata sections, we check the name on the .section pseudo.
3091 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
3092 add section's paddr to add_number; compatibility to native as and
3097 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
3099 * hash.c (struct hash_entry): Moved here...
3100 * hash.h (struct hash_entry): ...from here.
3102 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
3103 if it's already known to be null.
3107 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
3108 bit relocs converted to PC relative relocs. This turned out to
3109 add the offset from the beginning of .text twice.
3113 * stabs.c (s_stab_generic): In 's' case, free string from
3116 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
3118 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
3121 * subsegs.h (segment_info_type): Make bitfields unsigned.
3123 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
3124 and move them together. On most systems this will reduce the
3125 structure size by one word.
3126 (operatorT): Define O_max.
3127 * expr.c (expr_begin): Verify that X_op is wide enough to hold
3130 * read.c (pop_insert): Print error returned by hash table
3133 * as.c (dump_statistics): Split out from main; dump some hash
3134 table stats and target-specific stats.
3135 (start_time): No longer automatic to main.
3136 (main): Set file-level start_time and call dump_statistics at
3137 exit. Exit by calling xexit.
3138 (show_usage): Make --statistics description less specific.
3139 * subsegs.c (subsegs_print_statistics): New function.
3140 * write.c (write_print_statistics): New function.
3141 (n_fixups): New static variable.
3142 (fix_new_internal): Increment it.
3143 * read.c (read_print_statistics): New function.
3144 * read.h (read_print_statistics): Declare.
3145 * symbol.c (symbol_print_statistics): New function.
3146 * symbol.h (symbol_print_statistics): Declare.
3147 * hash.c (hash_print_statistics): New function.
3148 * hash.h (hash_print_statistics): Declare.
3149 * config/tc-i386.c (i386_print_statistics): New function.
3150 * config/tc-i386.h (i386_print_statistics): Declare.
3151 (tc_print_statistics): New macro.
3152 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
3155 * hash.c (DELETED): Rewrite to use a valid but unique address.
3156 (START_POWER): Reduce to 10.
3157 (enum stat_enum): New enumerator, replacing STAT_* index macros.
3158 Add new values for counting strcmp calls.
3159 (GROW_FACTOR): New macro.
3160 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
3162 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
3163 (hash_new): Use FULL_VALUE.
3164 (struct hash_control): Definition moved here.
3165 (hash_code): Don't mask to low bits.
3166 (hash_ask): Mask returned hash code here. Check hash value before
3167 calling strcmp; count strcmp calls.
3168 * hash.h (struct hash_control): Declare, don't define, here.
3169 (HASH_STATLENGTH): Deleted.
3170 (struct hash_entry): Add field for hash code.
3171 (hash_say, hash_apply): Don't declare.
3173 * hash.c (destroy): Return void.
3174 (applicatee): Ditto.
3175 (main): Fix declarations.
3176 (hash_apply): Return void. Argument `function' returns void. Put
3177 inside "#ifdef TEST".
3178 (hash_say): Define only if TEST is defined.
3179 * hash.h (hash_apply, hash_say): Declarations deleted.
3183 * configure: Rebuild with autoconf 2.7.
3187 * aclocal.m4 (AC_PROG_CC): Remove local definition.
3188 * configure: Rebuild with autoconf 2.6.
3192 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
3193 (ppc_stabx): Don't increment ppc_debug_name_section_size.
3195 (ppc_frob_file): Remove.
3196 * config/tc-ppc.h (tc_frob_file): Don't define.
3197 (ppc_frob_file): Don't declare.
3201 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
3202 * config/alpha-opcode.h: Include one-operand variants of jmp and
3205 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
3206 with te-delta.h in 8.3 file systems.
3207 * configure.in: Adjusted.
3211 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
3213 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
3215 * expr.c (operand): Handle MRI suffixes after unadorned 0.
3219 Version 2.6 released.
3220 * Makefile.in (VERSION): Updated to 2.6.
3222 * config/obj-coff.c (write_object_file): Change use of md_do_align
3223 to pass a pointer rather than a fill value, to match other uses.
3227 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
3229 * as.c (main): Move md_end call to just after call to
3230 perform_an_assembly_pass. Delete cpu-specific code here.
3231 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
3232 * config/tc-arm.c (md_end): Unused function deleted.
3233 * config/tc-ns32k.c (md_end): Ditto.
3235 * config/tc-i386.c (i386_align_code): New function, moved here
3236 from HANDLE_ALIGN macro.
3237 * config/tc-i386.h (HANDLE_ALIGN): Call it.
3241 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
3246 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
3247 multi-byte nops to avoid changing gcc. The necessary gcc change
3248 might break old assemblers.
3252 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
3253 generate optimal multi-byte nop instructions for ".align n"
3254 ".align n,0x90", and aligns requiring more than 15 bytes of
3255 padding still generate multiple 0x90's as before.
3259 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
3263 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
3264 * configure: Update.
3265 * mpw-make.sed: Reorder commands to make sed happier.
3266 * config/te-macos.h: New file.
3267 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
3271 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
3272 conditional branches.
3273 (md_apply_fix): Likewise.
3277 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
3278 generate a reloc which the linker will not handle correctly. Fix
3279 overflow checking--R_IREL is 18 bits, not 17.
3283 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
3284 from a PC relative reloc if TC_A29K.
3286 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
3287 code depends upon it.
3289 * config/tc-a29k.c (md_operand): Handle $float, $double, and
3290 $extend. Based on code from Eric Freudenthal
3292 * config/tc-a29k.h (LEX_DOLLAR): Define.
3293 * read.c (LEX_DOLLAR): Define if not defined.
3294 (lex_type): Use LEX_DOLLAR.
3298 * configure.in (a29k-nyu-sym1): New target, just like other a29k
3303 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
3307 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
3308 used only by "dlltool.c".
3312 * config/tc-alpha.c: Undefine inline if not __GNUC__.
3313 (md_pseudo_table): Don't define "extern".
3317 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
3319 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
3323 * macro.c (macro_expand_body): Don't warn about == with a
3324 nonexistent parameter, in case it is in a comment field.
3326 * as.c (main): On TC_A29K, call macro_init with macro_alternate
3328 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
3329 set, even if macro_alternate is set.
3330 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
3331 (sub_actual): If macro_strip_at, and kind is '@', don't look up
3332 the token unless it ended in '@'.
3333 * config/tc-a29k.c (line_separator_chars): Remove '@'.
3334 * doc/c-a29k.texi: Document macro usage on A29K.
3338 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
3340 (install_operand): Likewise.
3342 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
3346 * configure.in (m88k-motorola-sysv*): New target.
3347 * configure: Rebuild.
3348 * config/te-delta88.h: New file.
3349 * config/obj-coff.c (write_object_file): Use md_do_align if it is
3351 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
3352 (md_do_align): Define.
3353 * config/tc-m88k.c: Include "subsegs.h".
3354 (m88k_do_align): New function.
3356 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
3357 (COFF_NOLOAD_PROBLEM): Define.
3358 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
3362 * config/tc-i386.c (md_assemble): For a jump instruction with
3363 non-constant target, require 7 available bytes in the current
3368 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
3373 * configure.in (alpha-*-linux*): Use ecoff.
3374 * configure: Rebuild.
3375 * ecoff.c (ecoff_directive_extern): New function.
3376 (ecoff_directive_weakext): New function.
3377 (ecoff_build_symbols): Handle weak symbols.
3378 (ecoff_setup_ext): Likewise.
3379 (ecoff_frob_symbol): Warn about weak common symbols.
3380 * ecoff.h (ecoff_directive_extern): Declare.
3381 (ecoff_directive_weakext): Declare.
3382 * symbols.c (S_IS_WEAK): New function.
3383 * symbols.h (S_IS_WEAK): Declare.
3384 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
3386 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
3391 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
3393 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
3394 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
3395 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
3396 * config/obj-coff.c (coff_frob_symbol): Don't call
3397 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
3398 one. If OBJ_XCOFF, try to figure out whether the symbol is going
3401 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
3402 (ppc_stab_symbol): New static variable.
3403 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
3404 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
3405 sy_tc.real_name to the stab string.
3406 (ppc_bc, ppc_ec): New static functions.
3407 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
3409 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
3411 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
3414 * input-scrub.c (input_scrub_push): Reset sb_index.
3418 * listing.c (listing_newline): Don't create a frag in the absolute
3423 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
3424 (ppc_csect): Move most of the code to ppc_change_csect, and call
3426 (ppc_change_csect): New static function, taken from ppc_csect.
3427 (ppc_section): New static function.
3428 (ppc_saw_abs): New static varable.
3429 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
3430 a symbol has no csect.
3431 (ppc_adjust_symtab): New function.
3432 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
3433 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
3435 * write.c (write_object_file): If tc_adjust_symtab is defined,
3436 call it just before the call to obj_adjust_symtab.
3438 * symbols.c (symbol_find_or_make): Change name to be const.
3439 * symbols.h (symbol_find_or_make): Update declaration.
3443 * doc/as.texinfo (Align): Mention SH.
3444 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
3446 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
3447 (Z8000 Directives, even): Don't give numeric align value, instead
3452 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
3457 * Makefile.in (diststuff): Also make info.
3458 (maintainer-clean realclean): Also make clean-info.
3462 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
3463 (pa_subspace): For ".nsubspa", always create a new subspace
3464 with the given attributes, even if one already exists with the
3469 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
3470 TC_FORCE_RELOCATION, taking an additional section argument. If
3471 the section of the target symbol is not the same as the current
3472 section, always force the relocation to be used.
3473 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
3475 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
3476 md_pcrel_from function, taking an additional section argument.
3477 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
3479 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
3480 older TC_FORCE_RELOCATION if not defined.
3481 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
3482 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
3483 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
3484 TC_FORCE_RELOCATION.
3488 * input-scrub.c (as_where): Set name to null pointer if we don't
3490 * messages.c (identify): Only print filename if non-null.
3491 (as_show_where): Ditto, for line number too.
3492 (as_warn_internal, as_bad_internal): Ditto.
3494 * input-file.c (input_file_open): If the input file can't be
3495 opened, consider it an error.
3499 * config/tc-mips.c: Added mips_4100 control, and support for
3500 accepting the 4100 as a MIPS architecture variant (md_begin,
3501 macro_build, mips_ip, md_parse_option). Adding suitable
3502 command-line OPTIONs, and updating the help text (md_show_usage).
3506 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
3507 BFD_ASSEMBLER is defined.
3509 * Use one active frag and one obstack per frag chain:
3510 * frags.c (frags): Variable deleted.
3511 (frag_alloc): New function.
3512 (frag_grow, frag_more, frag_variant, frag_now_fix,
3513 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
3515 (frag_new): Ditto. Verify that frch_last and frag_now match on
3516 entry and exit, and that old frag_now has non-zero type. Replace
3517 "know" uses with "assert". Use frag_alloc instead of mucking with
3519 * frags.h (frags): Declaration deleted.
3520 * subsegs.h (struct frchain): Add new field frch_frag_now.
3521 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
3523 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
3524 give it any stricter alignment than frchainS structures need. Do
3525 not initialize frags obstack. Set frag_now to point to
3526 dummy_obstack. Initialize absolute_frchain.
3527 (subseg_set_rest): Save and restore frag_now in frch_frag_now
3528 field of frchainS. Don't create new frags on section switch, and
3529 use frag_alloc when creating a new frag chain. For absolute
3530 section, set frchain_now to absolute_frchain. Verify that
3531 frch_last and frag_now match on entry and exit. Initialize
3532 per-chain obstack, and under gcc, set required alignment to that
3533 needed by fragS structure.
3535 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
3537 * stabs.c (get_stab_string_offset): Only copy input string if a
3538 fresh copy is needed, not if the section already exists.
3539 (s_stab_generic): Cache stab section name to bypass lookups, since
3540 usually it will match. Could be made faster still by changing the
3541 memory allocation rules.
3542 (s_xstab): Cache section name to bypass repeated string
3545 * frags.c (frag_new): Deleted register declarations.
3547 * listing.c (frag_now): Don't declare.
3549 * as.c (chunksize): New variable.
3550 (debug_memory): New variable.
3551 (main): If debug_memory is set, reduce chunksize and
3553 * as.h (chunksize): Declare it.
3554 * read.c (read_begin): Use it.
3556 * config/tc-alpha.c (md_shortopts): Include 'g'.
3557 (md_parse_option): Ignore it.
3559 * Makefile.in (distclean): Remove Makefile and config.status from
3560 testsuite directory.
3561 (clean-here): Don't delete testsuite. Instead, delete only the
3562 files within it that would be generated by running tests.
3564 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
3565 frag_now_fix instead of accessing obstack info directly.
3566 * config/tc-arm.c (s_ltorg): Ditto.
3567 (md_assemble): Ditto.
3569 * config/tc-i386.c (md_assemble): Call frag_grow instead of
3574 * stabs.c (aout_process_stab): Insert debug symbol into symbol
3575 chain after parsing value expression, if any, to avoid separating
3576 continued .stabs lines.
3580 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
3581 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
3585 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
3586 -mcom support and PPC_OPCODE_ANY for -many.
3587 (md_parse_option): Ditto.
3589 (md_begin): For duplicate instructions, print all duplicates
3594 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
3595 common mode operation.
3596 (md_show_usage): Add -mcom to usage message.
3600 * expr.c (op_rank): Add O_symbol_rva.
3601 * expr.h (operatorT): Add O_symbol_rva.
3602 * read.c (cons_worker): Set O_symbol_rva when necessary.
3603 * write.c (fix_new_exp): Understand O_symbol_rva.
3607 * config/tc-mips.c: Correct s_cons arguments. From Michael
3612 * config/tc-ppc.c (ppc_macro): Make count unsigned.
3613 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
3614 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
3616 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
3617 make non-static. Change all users.
3618 * config/obj-coff.h (coff_n_line_nos): Declare.
3624 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
3626 * config/obj-elf.c (obj_elf_common): Convert specified byte
3627 alignment to power of two. Set size of local bss symbol.
3629 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
3633 * sb.c, macro.c: Decide whether to include <string.h> or
3634 <strings.h> just as as.h does.
3638 * Makefile.in (site.exp): Fix setting of $srcdir.
3640 * config/tc-arm.c (md_atof): Fix little-endian output.
3641 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
3646 * doc/as.texinfo: Split out the NS32k family documentation,
3647 despite its being commented out for now.
3648 * doc/c-ns32k.texi: New file.
3650 * sb.c, macro.c: Include string.h.
3652 * Makefile.in (comparison): Only check *.o; we don't care if
3653 timestamps inserted by the native linker differ.
3655 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
3656 if alignment stricter than 4 bytes is requested; in that case,
3657 align to a 4-byte boundary first.
3661 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
3662 uninitialized" warning for `Max_Source_Offset'.
3666 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
3667 qualifiers and detect errors if present.
3668 (md_assemble): Add call to parse_toc_entry. Also added some support
3669 for the [tocv] qualifier.
3670 (ppc_pe_tocd): New function to support data in the toc section.
3674 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
3679 * config/tc-m68k.c (isword): Accept all values from -65536 to
3680 +65535, so ~VAL will not be rejected.
3682 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
3683 skip characters after the pseudo-op.
3685 * read.c (get_line_sb): Don't look past buffer_limit.
3686 (s_include): In MRI mode, skip characters after the file name.
3690 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
3691 periods in register names.
3693 For older changes see ChangeLog.1.