3 * dep-in.sed: Remove libintl.h. Ignore include/fpoen-*.h
4 * Makefile.am (POTFILES.in): Unset LC_COLLATE.
6 * Makefile.in: Regenerate.
10 * config/tc-m68k.h (TARGET_USE_CFIPOP)
11 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT)
12 (tc_regname_to_dw2regnum, tc_cfi_frame_initial_instructions):
14 * config/tc-m68k.c: Include "dw2gencfi.h".
15 (tc_m68k_regname_to_dw2regnum)
16 (tc_m68k_frame_initial_instructions): New functions.
20 * config/tc-msp430.c: Add xW42 and xE42 parts. Sort MPU list
21 according to gcc order.
25 * config/tc-i860.c (i860_check_label): New function.
26 * config/tc-i860.h (i860_check_label): New prototype.
27 (tc_check_label): Define macro as i860_check_label.
31 * config/tc-i860.c (s_align_wrapper): New function and prototype.
32 (md_pseudo_table): Change s_align_bytes to s_align_wrapper, remove
33 surrounding OBJ_ELF ifdef, and re-format slightly.
34 * doc/c-i860.texi: Document the special .align syntax available
39 * config/tc-i860.c (i860_handle_align): New function.
40 * config/tc-i860.h (HANDLE_ALIGN): Define macro.
41 (MAX_MEM_FOR_RS_ALIGN_CODE): Define macro.
45 * config/tc-i860.c (i860_process_insn): Check that instructions
46 with their dual-bit set are 8-byte aligned.
50 * po/fr.po: Updated French translation.
54 * config/tc-i860.c (i860_process_insn): Don't handle dual-bit
55 setting during flop argument parsing. Instead, do it after
56 instruction is fully parsed.
60 * doc/c-i860.texi: Mention that .dual, .enddual, and .atmp
61 directives are only available in Intel syntax mode.
65 * config/tc-i860.c (s_dual): Accept .dual directive only in
66 the Intel syntax mode.
67 (s_enddual): Likewise for .enddual.
68 (s_atmp): Likewise for .atmp.
72 * config/tc-i386.c (i386_intel_operand): Always call i386_index_check
73 for memory operands. Pass the full operand_string to i386_index_check.
77 * config/tc-i860.c: Remove SYNTAX_SVR4 macro and occurrences.
78 (target_intel_syntax): Declare variable.
79 (OPTION_INTEL_SYNTAX): Declare macro.
80 (md_longopts): Add option -mintel-syntax.
81 (md_parse_option): Set target_intel_syntax.
82 (md_show_usage): Add -mintel-syntax usage.
83 (md_begin): Set reg_prefix based on target_intel_syntax.
84 (i860_process_insn): Skip register prefix only if there is one.
85 Parse relocatable expressions in either Intel or AT&T syntax based
86 on target_intel_syntax instead of the SYNTAX_SVR4 macro.
87 * doc/c-i860.texi: Document -mintel-syntax option and give blurb
88 about the differences in syntax.
92 * config/tc-msp430.c (msp430_srcoperand): Extend 'push' bug workaround
97 * config/tc-sh.c (md_assemble): For branches, check & update
102 * config/tc-i860.c: Convert to ISO C90.
106 * read.c (s_space): Revert 2003-07-28 change.
110 * config/obj-elf.c (obj_elf_change_section): Allow "x" for .note*.
114 * config/tc-sh.c (tc_gen_reloc): Test for R_SH_IND12W only when ELF.
118 * config/tc-ppc.c (md_apply_fix3): Fix check for lq insns.
122 * config/obj-elf.c (obj_elf_section_type): Also accept "note".
126 * read.c (s_space): Don't warn about .space 0.
130 * config/tc-mips.c (mips_flag_pdr): Define.
131 (md_begin) [OBJ_ELF]: Use it to control .pdr creation.
132 (s_mips_end) [OBJ_ELF]: Likewise.
133 (md_longopts) [OBJ_ELF]: Define OPTION_PDR, OPTION_NO_PDR.
134 (md_parse_option) [OBJ_ELF]: Handle them.
135 (md_show_usage) [OBJ_ELF]: Document -mpdr, -mno-pdr.
137 * doc/c-mips.texi (MIPS Opts): Document -mpdr, -mno-pdr.
138 * doc/as.texinfo (Overview) [MIPS]: Likewise.
142 * config/obj-elf.c (obj_elf_change_section): Update
143 elf_section_type and elf_section_flags only when they are
148 * config/obj-elf.c (obj_elf_change_section): Always set section
153 * config/obj-elf.c (special_sections): Removed.
154 (obj_elf_change_section): Call _bfd_elf_get_sec_type_attr. Set
155 elf_section_type and elf_section_flags.
156 (elf_frob_file): Set SHT_GROUP.
158 * config/obj-elf.h (obj_sec_set_private_data): New.
160 * config/tc-alpha.h (ELF_TC_SPECIAL_SECTIONS): Removed.
161 * config/tc-ia64.h: Likewise.
162 * config/tc-m32r.h: Likewise.
163 * config/tc-m68hc11.h: Likewise.
164 * config/tc-mcore.h: Likewise.
165 * config/tc-mips.h: Likewise.
166 * config/tc-ppc.h: Likewise.
167 * config/tc-sh64.h: Likewise.
168 * config/tc-v850.h: Likewise.
169 * config/tc-xtensa.h: Likewise.
171 * config/tc-v850.h (SHF_V850_GPREL): Removed.
172 (SHF_V850_EPREL): Likewise.
173 (SHF_V850_R0REL): Likewise.
175 * subsegs.c (subseg_get): Call obj_sec_set_private_data if it
180 * config/tc-m68hc11.h (DWARF2_ADDR_SIZE): Use 32-bit address for
181 debugging symbols so that we handle page memory correctly.
185 * po/fr.po: Updated French translation.
189 * config/tc-arm.c (arm_archs): Add iwmmxt.
193 * read.c (do_parse_cons_expression): Mark nbytes unused to
198 * config/tc-h8300.c (get_specific): No PCREL8 encoding for bsr/bc
201 * config/tc-h8300.c (md_assemble): Make sure characters after
202 slash and dot are lower-case.
206 * po/es.po: New Spanish translation.
207 * po/tr.po: New Turkish translation.
208 * po/opcodes.pot: Regenerate.
212 * dwarf2dbg.c (get_frag_fix): Revert 2001-11-15 change.
213 (generic_dwarf2_emit_offset): Don't define function when
214 TC__DWARF2_EMIT_OFFSET is defined.
218 * config/tc-mips.c (hilo_interlocks): True for CPU_RM7000.
219 (mips_cpu_info_table): Add rm7000 and rm9000 entries.
220 * doc/c-mips.texi: Document -march=rm9000.
224 * config/tc-v850.c (md_assemble): When no reloc, create pcrel fixups
225 only for V850_OPERAND_DISP operands.
229 * frags.c (frag_more): Move segment checks to..
230 (frag_alloc_check): ..here. New function.
231 (frag_append_1_char): Call frag_alloc_check.
235 * po/tr.po: Update with latest version.
236 * po/POTFILES.in: Regenerate.
237 * po/gas.pot: Regenerate.
238 * configure: Regenerate.
242 * config/tc-ip2k.c: Remove inclusion of <ctype.h>
243 * config/tc-tic4x.c: Replace inclusion of <ctype.h> with
244 "safe-ctype.h" and update use of macros.
245 * Makefile.am: Update dependencies.
246 * Makefile.in: Regenerate.
250 * po/gas.pot: Regenerate.
254 * config/tc-ppc.h (DWARF2_LINE_MIN_INSN_LENGTH): Define always.
255 (TARGET_USE_CFIPOP, tc_cfi_frame_initial_instructions,
256 tc_regname_to_dw2regnum, DWARF2_DEFAULT_RETURN_COLUMN,
257 DWARF2_CIE_DATA_ALIGNMENT): Define.
258 (ppc_cfi_frame_initial_instructions, tc_ppc_regname_to_dw2regnum): New
260 (ppc_cie_data_alignment): Declare.
261 * config/tc-ppc.c: Include dw2gencfi.h.
262 (ppc_cie_data_alignment): Define.
263 (md_begin): Initialize ppc_cie_data_alignment.
264 (ppc_cfi_frame_initial_instructions, tc_ppc_regname_to_dw2regnum): New
266 * config/tc-s390.h (DWARF2_LINE_MIN_INSN_LENGTH): Define always.
267 (TARGET_USE_CFIPOP, tc_cfi_frame_initial_instructions,
268 tc_regname_to_dw2regnum, DWARF2_DEFAULT_RETURN_COLUMN,
269 DWARF2_CIE_DATA_ALIGNMENT): Define.
270 (s390_cfi_frame_initial_instructions, tc_s390_regname_to_dw2regnum):
272 (s390_cie_data_alignment): Declare.
273 * config/tc-s390.c: Include dw2gencfi.h.
274 (s390_cie_data_alignment): Define.
275 (md_begin): Initialize s390_cie_data_alignment.
276 (s390_cfi_frame_initial_instructions, tc_s390_regname_to_dw2regnum):
282 * config/tc-mn10300.h (EXTERN_FORCE_RELOC): Don't define to zero.
284 * configure.in (am33_2.0, mn10300-*-linux*): Added.
285 * configure: Rebuilt.
286 * config/tc-mn10300.h (TARGET_FORMAT) [TE_LINUX]: Define to
288 * config/tc-mn10300.c (md_begin) [TE_LINUX]: Choose AM33/2.0
294 * config/tc-mn10300.c (mn10300_check_fixup): Set GOT_PCREL type
295 for subtracts from GLOBAL_OFFSET_TABLE that could not be
298 * config/tc-mn10300.c (mn10300_check_fixup): Accept subtracts that
299 could not be simplified.
300 (tc_gen_reloc): Turn an absolute fx_subsy into part of fx_offset.
302 * config/tc-mn10300.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't adjust
303 BDF_RELOC_MN10300_GOT32.
304 * config/tc-mn10300.c (mn10300_fix_adjustable): If
305 TC_RELOC_RTSYM_LOC_FIXUP doesn't hold, it's not adjustable.
307 * config/tc-mn10300.c (mn10300_parse_name): Don't return a
308 symbol if we know its value.
310 * config/tc-mn10300.h (GLOBAL_OFFSET_TABLE_NAME): Remove
311 duplicate underscore prefix.
313 * config/tc-mn10300.c (mn10300_parse_name): Store relocation
314 type in X_md, not X_add_number. Zero X_add_number.
315 (mn10300_check_fixup): Extract relocation type from X_md.
316 * config/tc-mn10300.h: Update comment.
318 * config/tc-mn10300.h (O_GOTOFF, O_PLT, O_GOT): Replace with...
320 * config/tc-mn10300.c (mn10300_PIC_related_p): Use it.
321 (mn10300_check_fixup): Likewise.
322 (mn10300_parse_name): Set X_add_number to relocation type.
323 * config/tc-mn10300.h (DIFF_EXPR_OK, GLOBAL_OFFSET_TABLE_NAME,
324 TC_RELOC_RTSYM_LOC_FIXUP, md_parse_name, TC_CONS_FIX_NEW,
325 O_GOTOFF, O_PLT, O_GOT): Define.
326 * config/tc-mn10300.c (mn10300_PIC_related_p): New fn.
327 (mn10300_check_fixup): New fn.
328 (md_assemble): Call it. Check for PIC-related relocs.
329 (mn10300_cons_fix_new): Likewise. New fn.
330 (mn10300_end_of_match): New fn.
331 (mn10300_md_parse_name_cont): New fn.
336 * config/tc-mn10300.c (mn10300_insert_operand): Negate negative
339 * config/tc-mn10300.c (md_relax_table, md_convert_frag,
340 md_assemble, md_estimate_size_before_relax): Handle fbCC.
342 * config/tc-mn10300.c (HAVE_AM33): Redefine in terms of
345 * config/tc-mn10300.c (md_pseudo_table): Use AM33_2 constant.
346 (HAVE_AM33): Match AM33_2 too.
347 (HAVE_AM33_2): New macro.
348 (md_assemble): Use it. Match 2.0 registers only if HAVE_AM33_2.
350 * config/tc-mn10300.c (md_pseudo_table): Added `am33_2'.
351 (float_registers, double_registers): New variables.
352 (float_register_name, double_register_name): New functions.
353 (md_assemble): Recognize FP registers. Implement FMT_D3.
354 (mn10300_insert_operand): Support FP registers.
358 * config/tc-mips.c (mips_validate_fix): Do not warn about branch
359 target being a global symbol if not compiling SVR4 PIC code.
363 * doc/c-m32r.texi (M32R-Directives): New node. Document the
364 .high, .shigh and .low directives.
368 * config/tc-h8300.c (h8300sxnmode): Add prototype.
370 (parse_exp): Replace expressionS argument with a h8_op. Parse the
371 operand size as well.
372 (skip_colonthing): Remove unused expression argument. Tighten checks
374 (colonmod24): Remove.
375 (get_mova_operands): Combine calls to parse_exp and skip_colonthing.
376 (get_operand): Likewise. Use the standard code to read the size of
377 pc-relative operands.
378 (fix_operand_size): Include the size-guessing logic that used to be
379 in colonmod24 and get_operand. Don't apply dd:2 optimizations to
380 offsets with a symbolic component.
384 * config/tc-i386.c (tc_x86_regname_to_dw2regnum): Use ARRAY_SIZE
385 macro to compute size of selected register name array.
389 * config/tc-s390.c (md_parse_option): Add cpu type z990.
390 (md_begin): Add minimal cpu type logic for instructions with different
391 binary format depending on the cpu.
392 (md_assemble): Remove check for minimal cpu.
393 (s390_insert_operand): Add support for long displacements.
394 (md_gather_operands): Likewise.
395 (tc_s390_fix_adjustable): Likewise.
396 (tc_s390_force_relocation): Likewise.
397 (md_apply_fix3): Likewise.
401 * config/tc-mips.c (s_mipsset): Implement -march= handling
406 * config/tc-mips.c: Convert to ISO C90 prototypes. Remove unnecessary
407 prototypes and casts. Replace PTR with void *. Reformat.
408 * config/tc-mips.h: Likewise.
412 * config/tc-ia64.c (note_register_values): Warning fix.
413 * config/tc-mips.c (append_insn): Likewise.
417 * config/tc-mips.c (mips_set_options,mips_opts): Support for
419 (file_mips_arch): Rename mips_arch.
420 (mips_arch_info,mips_tune_info): Remove.
421 (hilo_interlocks,gpr_interlocks,cop_interlocks): Use mips_opts.arch.
422 (mips_cpu_info_from_arch): New function.
423 (md_begin): Use file_mips_arch.
424 (macro_build,macro,mips_ip): Use mips_opts.arch.
425 (mips_set_architecture): Init file_mips_arch and mips_opts.arch.
426 (mips_after_parse_args): Remove mips_arch_info and mips_tune_info.
428 (s_mipsset): Support for .set arch=FOO.
429 (mips_cpu_info_table): Fix typo.
433 * config/tc-i386.c (md_assemble): Declare "exp" before "if".
437 * config/tc-h8300.c (get_specific): Allow ':8' to be used for
438 unsigned 8-bit operands.
442 * read.c (s_comm): Change error message to assume an unsigned size
443 has been passed to .comm.
444 * config/tc-sparc.c (s_common): Likewise.
445 * write.c (write_contents): Replace 'unsigned long' with
446 'addressT' and 'long' with offsetT in order to allow computations
447 with very large values to work for 64-bit addressed targets.
448 (relax_and_size_all_segments): Likewise.
449 (relax_frag): Likewise.
450 (relax_segment): Likewise.
454 * config/tc-ppc.c (ppc_cleanup): Use bytes to count APUinfo slots.
458 * gas/config/tc-i386.c (md_assemble): Support Intel Precott New
461 * gas/config/tc-i386.h (CpuPNI): New.
462 (CpuUnknownFlags): Add CpuPNI.
466 * config/tc-ia64.c (pseudo_func): Add ABI constants for linux,
467 freebsd, openvms, and nsk (non-stop kernel).
471 * config/tc-ns32k.c (md_begin): Initialize inst_hash_table after
472 all locals have been declared.
476 * config/tc-mips.c (ADDRESS_ADD_INSN,ADDRESS_ADDI_INSN): Remove
477 special handling for n32 ABI.
482 * config/tc-mips.c (s_cpsetup,s_cprestore,s_cpreturn): Revert
487 * config/tc-z8k.c (parse_reg): Invalid registers generate an error
488 now, not only a warning. Add some more checks to detect invalid
490 (get_operand): For CLASS_IR remember register size in mode struct.
491 (get_specific): Handle new CLASS_IRO type. Add register size
492 checks for CLASS_IR and CLASS_IRO.
493 (md_apply_fix3): Fix undefined usage of buf.
497 * config/tc-ppc.c (ppc_csect): Pass alignment to ppc_change_csect.
498 (ppc_change_csect): Add align param. Align frag at start of csect.
499 (ppc_section, ppc_named_section): Adjust ppc_change_csect calls.
500 (ppc_frob_section): Align vma.
504 * dw2gencfi.c (EH_FRAME_ALIGNMENT): Define if not defined.
505 (output_cie): Don't pad.
506 (output_fde): Add align argument. Pad to align if not 0.
507 (cfi_finish): Set .eh_frame alignment to EH_FRAME_ALIGNMENT.
508 Pad just last FDE to EH_FRAME_ALIGNMENT.
512 * config/tc-s390.c (init_default_arch): Make current_mode_mask
513 dependent on s390_arch_size and current_cpu dependent on
518 * configure.in: Add c4x as an architecture variant to tic4x.
519 * configure: Regenerate.
520 * Makefile.in: Regenerate.
521 * doc/Makefile.in: Regenerate.
525 * config/tc-ppc.c (ppc_elf_suffix): Don't remove symbols other than
526 ".TOC." from PPC64_TOC relocs.
530 * config/tc-ppc.c (md_apply_fix3): Allow BRTAKEN, BRNTAKEN relocs.
534 * config/tc-mips.c (ADDRESS_ADD_INSN,ADDRESS_ADDI_INSN,
535 ADDRESS_LOAD_INSN,ADDRESS_STORE_INSN): New macros.
536 (macro_build_ldst_constoffset,load_address,macro,s_cpsetup,
537 s_cprestore,s_cpadd): Use them.
541 * configure.in: Add specific case for cris-*-linux-gnu* with
543 * configure: Regenerate.
544 * config/tc-cris.c (DEFAULT_CRIS_AXIS_LINUX_GNU): New macro, TRUE
545 if TE_LINUX defined, else FALSE.
546 (bfd_boolean demand_register_prefix): Set default from
547 DEFAULT_CRIS_AXIS_LINUX_GNU.
548 (symbols_have_leading_underscore): Similar.
549 * config/tc-cris.h (LOCAL_LABELS_DOLLAR): Define to 1.
553 * tc-arm.c (FPU_DEFAULT, case TE_LINUX): Default to FPU_ARCH_FPA.
554 (FPU_DEFAULT, case TE_NetBSD): Default to FPU_ARCH_VFP for ELF,
555 FPU_ARCH_FPA for AOUT.
556 (md_begin): Don't try to guess the floating point architecture from
557 the CPU if the OS ABI (Linux, NetBSD) mandates a particular form.
561 * configure.in: Add i386-netbsd-gnu target.
562 * configure: Regenerate.
566 * doc/as.texinfo (Comm): Added @node. Moved before CFI
571 * config/tc-mips.c (append_insn): In a compound relocation, take the
572 field width from the final (outermost) operator.
576 * dw2gencfi.c (struct cfi_escape_data): New.
577 (cfi_add_CFA_nop): Remove.
578 (CFI_escape, dot_cfi_escape): New.
579 (dot_cfi): Remove nop.
580 (cfi_pseudo_table): Remove nop; add escape.
581 (output_cfi_insn): Likewise.
582 (select_cie_for_fde): Stop on escape.
583 * dw2gencfi.h (cfi_add_CFA_nop): Remove.
584 * read.c, read.h (do_parse_cons_expression): New.
585 * doc/as.texinfo (.cfi_escape): New.
589 * config/tc-mips.c (s_cpsetup): Use mips_frame_reg instead of SP.
590 (s_cprestore): Likewise.
591 (s_cpreturn): Likewise.
595 * config/tc-mips.c (tc_gen_reloc): Initialize retval amd reloc
600 * config/tc-mips.c (md_pcrel_from): Return actual pcrel address.
601 (md_apply_fix3): Ignore non-special relocations. Remove superfluous
602 exceptions from size assert. Remove most of the addend fixup
603 specialcasing. Remove value, use valP directly. simplify fx_addnumber
604 handling. Remove zero addend specialcases.
605 (tc_gen_reloc): Use appropriate value for reloc2 addend. Remove
606 the addend fixup specialcase.
607 * config/tc-mips.h (MD_APPLY_SYM_VALUE): Define as 0.
611 * write.c (write_relocs): Use xcalloc. Fix relocs initialization
612 in the RELOC_EXPANSION_POSSIBLE case.
616 * config/tc-i960.c (line_comment_chars): Add '#'.
617 * config/tc-mn10200.c (tc_gen_reloc): Don't ignore fx_subsy.
621 * po/Make-in (DESTDIR): New.
622 (install-data-yes): Support $(DESTDIR).
623 (uninstall): Likewise.
627 * Makefile.am: Run "make dep-am".
628 * Makefile.in: Regenerate.
629 * po/POTFILES.in: Regenerate.
633 * cgen.c (gas_cgen_finish_insn): CGEN_INSN_RELAX renamed to
635 * config/tc-fr30.c (md_estimate_size_before_relax): Ditto.
636 * config/tc-m32r.c (md_estimate_size_before_relax): Ditto.
637 * config/tc-openrisc.c (md_estimate_size_before_relax): Ditto.
642 * config/tc-ppc.c (md_assemble): Handle PPC_OPERAND_DQ.
643 (md_apply_fix3): Special case lq insn.
647 * config/tc-h8300.c (get_rtsl_operands): Accept unbracketed register
648 lists. Allow single-register ranges.
652 * config/tc-h8300.c (h8300sxnmode): New.
653 (md_pseudo_table): Add .h8300sxn entry.
657 * NEWS: Updated for the new -n option for the i386 assembler.
659 * config/tc-i386.c (optimize_align_code): New.
660 (md_shortopts): Add 'n'.
661 (md_parse_option): Handle 'n'.
662 (md_show_usage): Add '-n'.
664 * config/tc-i386.h (optimize_align_code): Declared.
665 (md_do_align): Optimize code alignment only if optimize_align_code
668 * doc/as.texinfo: Add the new -n option.
670 * doc/c-i386.texi: Document the new -n option.
674 * doc/as.texinfo: Document .cfi_rel_offset.
676 * dw2gencfi.c (struct cfa_save_data, cfa_save_stack): New.
677 (cfi_add_CFA_offset): Detect invalid offsets.
678 (cfi_add_CFA_remember_state): Save cur_cfa_offset.
679 (cfi_add_CFA_restore_state): Restore it.
680 (CFI_rel_offset): New.
681 (cfi_pseudo_table): Add it.
682 (dot_cfi): Handle it.
686 * app.c (do_scrub_chars): Add states 14 and 15 to handle
691 * config/tc-h8sx.c (get_specific): Distinguish h8h from h8s ops.
692 (build_bytes): Ditto.
696 * config/tc-h8sx.c (DMODE): Remove.
697 (colonmod24): Don't choose a default if the operand is a 16-bit
699 (fix_operand_size): New function.
700 (md_assemble): Use it to choose between @(d:2, ERn) and @(d:16,ERn).
701 Adjust @(d:2,ERn) operands before choosing the specific opcodes.
705 * dw2gencfi.c (cfi_add_CFA_insn, cfi_add_CFA_insn_reg)
706 (cfi_add_CFA_insn_reg_reg, cfi_add_CFA_insn_reg_offset): New.
707 (cfi_add_CFA_offset, cfi_add_CFA_def_cfa)
708 (cfi_add_CFA_register, cfi_add_CFA_def_cfa_register)
709 (cfi_add_CFA_def_cfa_offset): Use cfi_add_CFA_insn_*().
710 (cfi_add_CFA_restore, cfi_add_CFA_undefined)
711 (cfi_add_CFA_same_value, cfi_add_CFA_remember_state)
712 (cfi_add_CFA_restore_state, cfi_add_CFA_nop): New.
713 (cfi_pseudo_table): New directives .cfi_return_column,
714 .cfi_restore, .cfi_undefined, .cfi_same_value,
715 .cfi_remember_state, .cfi_restore_state, .cfi_nop.
716 (dot_cfi, output_cfi_insn): Handle new directives.
717 * dw2gencfi.h (cfi_add_CFA_restore, cfi_add_CFA_undefined)
718 (cfi_add_CFA_same_value, cfi_add_CFA_remember_state)
719 (cfi_add_CFA_restore_state, cfi_add_CFA_nop): New prototypes.
723 * dw2gencfi.c (output_cfi_insn): Fix typo for negative offsets.
725 * dw2gencfi.c (cfi_finish): Set .eh_frame read-only.
729 * config/tc-alpha.c (s_alpha_usepv): New.
730 (md_pseudo_table): Add it.
731 (alpha_cfi_frame_initial_instructions): New.
732 * config/tc-alpha.h (TARGET_USE_CFIPOP): New.
733 (tc_cfi_frame_initial_instructions): New.
734 * doc/c-alpha.texi: Document .usepv.
738 * as.c (show_usage): Document --execstack and --noexecstack.
739 (parse_args): Add --execstack and --noexecstack.
740 (main): Create .note.GNU-stack section if --execstack or
741 --noexecstack was given on comand line, set its SHF_EXECINSTR bit.
742 * as.h (flag_execstack, flag_noexecstack): New.
746 * config/tc-mips.c: (OPTION_ARCH_BASE, OPTION_ASE_BASE)
747 (OPTION_COMPAT_ARCH_BASE, OPTION_FIX_BASE)
748 (OPTION_MISC_BASE): New defines.
749 (OPTION_BREAK, OPTION_CONSTRUCT_FLOATS, OPTION_EB, OPTION_EL)
750 (OPTION_ELF_BASE, OPTION_FIX_VR4122, OPTION_FP32, OPTION_FP64)
751 (OPTION_GP32, OPTION_GP64, OPTION_M3900, OPTION_M4010, OPTION_M4100)
752 (OPTION_M4650, OPTION_M7000_HILO_FIX, OPTION_MARCH, OPTION_MDMX)
753 (OPTION_MEMBEDDED_PIC, OPTION_MIPS1, OPTION_MIPS16, OPTION_MIPS2)
754 (OPTION_MIPS3, OPTION_MIPS32, OPTION_MIPS32R2, OPTION_MIPS3D)
755 (OPTION_MIPS4, OPTION_MIPS5, OPTION_MIPS64)
756 (OPTION_MNO_7000_HILO_FIX, OPTION_MTUNE, OPTION_NO_CONSTRUCT_FLOATS)
757 (OPTION_NO_FIX_VR4122, OPTION_NO_M3900, OPTION_NO_M4010)
758 (OPTION_NO_M4100, OPTION_NO_M4650, OPTION_NO_MDMX, OPTION_NO_MIPS16)
759 (OPTION_NO_MIPS3D, OPTION_NO_RELAX_BRANCH, OPTION_RELAX_BRANCH)
760 (OPTION_TRAP): Redefine in terms of new defines.
761 (md_longopts): Reorder entries.
767 * config/tc-h8300.c: Add insns and addressing modes for h8300sx.
768 * config/tc-h8300.h: Ditto.
772 * tc-v850.c (tc-gen_reloc): Translate BFD_RELOC_32 into
773 BFD_RELOC_32_PCREL if the reloc is pc-relative. Do this
774 before calling bfd_reloc_type_lookup.
778 * read.c (emit_expr): Set dot_value.
779 * dw2gencfi.c (output_fde): Remove pcrel reloc hack.
783 * macro.c (sub_actual): Don't lose string if it turns out that
784 &string wasn't an arg.
788 * dw2gencfi.c (output_fde): Use fix_new to emit pc-relative reloc.
789 (cfi_finish): Set flag_traditional_format around .eh_frame data.
793 * config/tc-alpha.c (alpha_cur_ent_sym): Remove.
794 (all_frame_data, plast_frame_data, cur_frame_data): New.
795 (s_alpha_ent): Record data for dwarf2 cfi.
796 (s_alpha_end, s_alpha_mask, s_alpha_frame, s_alpha_prologue): Likewise.
797 (alpha_elf_md_end): Emit dwarf2 cfi for ecoff unwind directives.
798 * config/tc-alpha.h (md_end): New.
799 (DWARF2_DEFAULT_RETURN_COLUMN): New.
800 (DWARF2_CIE_DATA_ALIGNMENT): New.
804 * configure.in: Add i386-*-freebsd* entry.
805 * configure: Regenerate.
809 * config/tc-m32r.c: Amend comment to refer to Renesas
813 * expr.c (make_expr_symbol): Fold FAKE_LABEL_NAME use into the
815 (current_location): Use symbol_temp_new_now.
816 * stabs.c (s_stab_generic): Use symbol_temp_new.
817 * symbols.c (temp_label_name): Remove.
818 (symbol_temp_new, symbol_temp_make): Use FAKE_LABEL_NAME.
822 * dw2gencfi.c, dw2gencfi.h: Rewrite from scratch.
823 * as.c (main): Always call cfi_finish.
824 * config/tc-i386.c (x86_dwarf2_return_column): New.
825 (x86_cie_data_alignment): New.
826 (md_begin): Set them.
827 (tc_x86_cfi_init): Remove.
828 (tc_x86_regname_to_dw2regnum): Fix 32-bit register numbers;
829 return int, not unsigned long; don't as_bad here.
830 (tc_x86_frame_initial_instructions): Streamline; use
832 * config/tc-i386.h (tc_cfi_init): Remove.
833 (DWARF2_DEFAULT_RETURN_COLUMN): New.
834 (DWARF2_CIE_DATA_ALIGNMENT): New.
838 * symbols.c (temp_label_name): New.
839 (symbol_temp_new, symbol_temp_new_now, symbol_temp_make): New.
840 (symbol_set_value_now): New.
841 * symbols.h: Prototype them.
842 * dwarf2dbg.c: Use them.
843 (fake_label_name, symbol_new_now, set_symbol_value_now): Remove.
847 * config/tc-i860.c (target_xp): Declare variable.
848 (OPTION_XP): Declare macro.
849 (md_longopts): Add option -mxp.
850 (md_parse_option): Set target_xp.
851 (md_show_usage): Add -mxp usage.
852 (i860_process_insn): Recognize XP registers bear, ccr, p0-p3.
853 (md_assemble): Don't try expansions if XP_ONLY is set.
854 * doc/c-i860.texi: Document -mxp option and i860XP support.
858 * config/tc-mips.c (macro_build_jalr): Warning patrol.
862 * config/tc-mips.c (append_insn): Use actual relocation size for new
863 fixp's. Don't relax overflow checking for partial_inplace relocations.
864 Use the actual relocation type in combined relocs, not just the type
866 (macro_build_jalr): Use actual relocation size for new fix.
867 (s_cpsetup, s_gpdword): Likewise.
871 * config/tc-mips.c (macro): Don't use uninitialized tempreg.
875 * config/tc-xstormy16.c (md_pcrel_from_section): Do not produce
876 section relative offsets for relocs that will not be based on the
878 (xstormy16_md_apply_fix3): Remove previous patch to this
883 * config/tc-mips.c (mips_abicalls): New variable.
884 (md_parse_option): Use.
887 (mips_elf_final_processing): Set EF_MIPS_PIC and
888 EF_MIPS_CPIC dependent on above.
892 * tc-hppa.c (hppa_symbol_chars): Remove `,' and `!'.
896 * config/obj-elf.c: Include dwarf2dbg.h.
897 (elf_pseudo_tab): Add .file and .loc.
898 * config/tc-arc.c (md_pseudo_table): Remove .file and .loc.
899 * config/tc-arm.c: Likewise.
900 * config/tc-h8300.c: Likewise.
901 * config/tc-hppa.c: Likewise.
902 * config/tc-ia64.c: Likewise.
903 * config/tc-m68hc11.c: Likewise.
904 * config/tc-m68k.c: Likewise.
905 * config/tc-mmix.c: Likewise.
906 * config/tc-mn10300.c: Likewise.
907 * config/tc-ppc.c: Likewise.
908 * config/tc-sh.c: Likewise.
909 * config/tc-sparc.c: Likewise.
910 * config/tc-v850.c: Likewise.
911 * config/tc-frv.c: Likewise, and remove redundant inclusion of
913 * config/tc-ip2k.c: Likewise.
914 * config/tc-iq2000.c: Likewise.
915 * config/tc-xstormy16.c: Likewise.
916 * config/tc-xtensa.c: Likewise.
917 * Makefile.am: Regenerate dependencies.
918 * Makefile.in: Regenerate.
919 * aclocal.m4: Regenerate.
920 * config.in: Regenerate.
921 * configure: Regenerate.
922 * po/POTFILES.in: Regenerate.
923 * po/gas.pot: Regenerate.
927 * dw2gencfi.c (cfi_get_label): Use symbol_make for non
928 BFD_ASSEMBLER targets.
929 (cfi_startproc): Change type of saved_seg and cfi_seg to segT.
930 (dot_cfi_endproc): Only call bfd_set_section_flags for targets
931 defining BFD_ASSEMBLER.
935 * as.c (main): Remove tc_cfi_init().
936 * dw2gencfi.c (cfi_parse_arg): Allow regnames beginning
938 (cfi_pseudo_table): Add "cfi_register" entry.
939 (cfi_make_insn): Handle CFA_register.
940 (cfi_output_insn): Ditto.
942 (cfi_get_label): Add 'simple' modifier to .cfi_startproc.
943 (dot_cfi_endproc): Reuse already emitted CIEs.
944 * testsuite/gas/cfi/cfi-i386.d: New pattern.
945 * testsuite/gas/cfi/cfi-x86-64.d: Ditto.
949 * config/tc-xstormy16.c: Include dwarf2dbg.h.
950 (md_pseudo_table): Add entries for .loc and .line.
954 * dw2gencfi.c (cfi_parse_arg): Only use tc_regname_to_dw2regnum if
956 (dot_cfi_endproc): Avoid C99 construct.
960 * dw2gencfi.c, dw2gencfi.h: New files.
961 * config/tc-i386.c (tc_x86_cfi_init): New function.
962 * config/tc-i386.h (TARGET_USE_CFIPOP, tc_cfi_init): New defines.
963 * as.c (parse_args): Set verbose flag on --verbose.
964 (main): Call tc_cfi_init()/cfi_finish().
965 * as.h (verbose): New external variable.
966 * read.c (pobegin): Insert CFI pops to the list.
967 * symbols.c (local_symbol_make): Make symbol external.
968 * symbols.h (local_symbol_make): New prototype.
969 * Makefile.am: Add dw2gencfi.[ch] files. Run "make dep-am".
970 * Makefile.in: Regenerate.
971 * doc/as.texinfo: Added node "CFI directives" with description of
972 all implemented .cfi_* directives.
973 * doc/Makefile.in: Regenerate.
974 * po/POTFILES.in: Regenerate.
978 * config/tc-tic4x.c (md_assemble): Added support for one-line parallel
980 * config/tc-tic4x.h: Added DOUBLEBAR_PARALLEL definition
984 * config/tc-i860.c (i860_process_insn): Initialize fc after
985 each opcode mismatch.
989 * configure.in: Accept i[3-7]86 variants.
990 * configure: Regenerate.
994 * config/tc-mips.h: Fix comment formatting.
998 * read.c (old_buffer, old_input, old_limit): Remove variables.
999 (read_a_source_file): Delete label contin.
1000 <handling #APP/#NO_APP>: Use an "sb" to push #APP expansion into
1001 input as with macros, instead of in separate old_* variables.
1002 Zero-terminate string being scrubbed.
1006 * config/tc-xstormy16.c (skipping_fptr): New local variable.
1007 (md_assemble): Reset skipping_fptr.
1008 (md_operand): If @fptr() is followed by a minus sign, set
1009 skipping_fptr and ignore the fptr. If skipping_fptr is set and an
1010 @fptr is detected, ignore it and reset skipping_fptr.
1014 * config/tc-i860.c (MAX_FIXUPS): Define.
1015 (struct i860_fi fi[]): New struct.
1016 (struct i860_it the_insn): Add above as member and move fields
1017 exp, reloc, pcrel and fup into i860_fi.
1018 (md_assemble): Replace all instances of exp, reloc, pcrel
1019 and fup with fi[].exp, fi[].reloc, fi[].pcrel, fi[].fup.
1020 Add a loop to possibly emit multiple fix-ups for each insn.
1021 (i860_process_insn): Likewise.
1022 (i860_get_expression): Likewise.
1023 (md_apply_fix3): Use a bitwise check for OP_IMM_U5, not equality.
1027 * config/tc-s390.c (s390_target_format): Always call init_default_arch.
1031 * config/tc-ia64.c (alias_hash): New.
1032 (alias_name_hash): New.
1033 (secalias_hash): New.
1034 (secalias_name_hash): New.
1035 (md_pseudo_table): Add "secalias".
1036 (md_begin): Initialize alias_hash, alias_name_hash,
1037 secalias_hash and secalias_name_hash.
1038 (struct alias): New.
1039 (dot_alias): Implement .alias and .secalias directives.
1041 (ia64_adjust_symtab): New.
1043 (ia64_frob_file): New.
1045 * config/tc-ia64.h (ia64_adjust_symtab): New.
1046 (tc_adjust_symtab): Defined.
1047 (ia64_frob_file): New.
1048 (tc_frob_file): Defined.
1052 * tc-msp430.c: Add missing lines to known cpus list.
1056 * configure.in (MIPS_DEFAULT_ABI): AC_DEFINE.
1057 * config/tc-mips.c (mips_after_parse_args): Set mips_abi to it.
1058 * config.in, configure: Rebuilt.
1062 * config/tc-mips.c (tc_gen_reloc): Add addend just once if
1063 howto->partial_inplace is false.
1067 * config/tc-mips.c (mips_need_elf_addend_fixup): Remove
1068 symbol_used_in_reloc_p check.
1069 (md_apply_fix3): Remove check for howto->pcrel_offset.
1073 * config/obj-elf.c (obj_elf_parse_section_letters): Make it a
1074 fatal error for unknown section attribute.
1076 * config/tc-alpha.c (alpha_elf_section_letter): Return -1 for
1077 unknown section attribute.
1078 * config/tc-ia64.c (ia64_elf_section_letter): Likewise.
1079 * config/tc-ppc.c (ppc_section_letter): Likewise.
1081 * config/tc-ia64.c (ia64_elf_section_letter): Handle 'o'.
1082 (ia64_elf_section_type): Accept "unwind".
1086 * read.h (demand_copy_string): New.
1088 * config/tc-alpha.c (demand_copy_string): Removed.
1092 * write.h (FAKE_LABEL_NAME): Allow override
1093 (for targets that like eg. a leading dot in a local label).
1097 * config/tc-xstormy16.c (xstormy16_md_apply_fix3): Do not bias the
1098 addend with the symbol's value for pc-relative relocations against
1099 a defined symbol - this will be done automatically.
1103 * config/obj-elf.c (obj_elf_type): Accept "notype" and
1108 * config/tc-ia64.h (tc_canonicalize_section_name): New.
1110 * config/obj-elf.c (obj_elf_section_name): Call
1111 tc_canonicalize_section_name if it is defined.
1115 * config/tc-ia64.c (ia64_check_label): New.
1116 * config/tc-ia64.h (tc_check_label): New.
1118 * read.c (read_a_source_file): Call tc_check_label after
1119 creating a user-defined label if defined.
1123 * config/tc-ppc.c (md_show_usage): Mention -a32, -a64, -l, -le, -b,
1124 -be and split strings to below 509 bytes in length.
1128 * expr.h: Fix comments in operatorT typedef.
1129 * config/tc-z8k.c: Add 2003 to copyright message.
1130 Fold s_segm() and s_unseg() into one function s_segm(parm) which
1131 decides by the parameter.
1132 (md_begin): Don't set linkrelax. Only set Z8002 default if no
1133 command line argument was given to select the intended
1135 (get_interrupt_operand): Warn if NOP type code is emitted.
1136 (newfix): New parameter 'size', forward it to 'fix_new_exp'.
1137 (apply_fix): Call newfix with additional 'size' parameter.
1138 (build_bytes): Remove unused variable 'nib'. Detect overflow in
1139 4 bit immediate arguments.
1140 (md_longopts): Add 'linkrelax' option.
1141 (md_parse_option): Adapt to new s_segm function. Set 'linkrelax'
1142 variable when 'linkrelax' command line option is specified.
1143 (md_show_usage): Display 'linkrelax' option.
1144 (md_apply_fix3): Fix cases R_IMM4L, R_JR, and R_IMM8. Add cases
1145 R_CALLR and R_REL16.
1146 * config/tc-z8k.h: Undef WARN_SIGNED_OVERFLOW_WORD.
1150 * config/tc-ia64.c (ia64_number_to_chars): New function pointer.
1151 (ia64_float_to_chars): Likewise.
1152 (dot_byteorder): Set target_big_endian, ia64_number_to_chars
1153 and ia64_float_to_chars by tc_segment_info_data.endian from
1154 the current segment if byteorder == -1.
1155 (md_begin): Call dot_byteorder to set target_big_endian.
1156 (md_atof): Call ia64_float_to_chars to convert floating point.
1157 (ia64_float_to_chars_bigendian): New function.
1158 (ia64_float_to_chars_littleendian): Likewise.
1159 (ia64_elf_section_change_hook): Likewise.
1161 * config/tc-ia64.h (ia64_number_to_chars): New.
1162 (md_number_to_chars): Changed to (*ia64_number_to_chars)
1163 (ia64_elf_section_change_hook): New.
1164 (md_elf_section_change_hook): Defined.
1165 (ia64_segment_info_type): New struct.
1166 (TC_SEGMENT_INFO_TYPE): Defined.
1170 * config/tc-ia64.c (md_section_align): Deleted.
1172 * config/tc-ia64.h (SUB_SEGMENT_ALIGN): New.
1173 (md_section_align): New.
1177 * config/tc-ia64.c (stmt_float_cons): Fix alignment for real10
1179 (md_pseudo_table): Add "xreal16", "xreal16.ua", "real16" and
1181 (md_atof): Add 6 byte padding of zero for real16.
1185 * config/obj-elf.c (obj_elf_symver): Skip whitespace before the
1186 start of a version name.
1190 * configure.in (mips-*-*n*bsd*): Replace with...
1191 (mips-*-netbsd*, mips-*-openbsd*): These.
1192 * configure: Regenerate.
1196 * config/tc-hppa.c (hppa_symbol_chars): New.
1197 * config/tc-hppa.h (tc_symbol_chars): Likewise.
1201 * config/tc-mips.h (TC_FORCE_RELOCATION_SUB_SAME): Define again.
1205 * NEWS: Belatedly mention support for MIPS32 Release 2.
1209 * config/tc-mips.c (nopic_need_relax): Revert previous
1214 * config/tc-h8300.h (DWARF2_LINE_MIN_INSN_LENGTH): New
1215 * config/tc-h8300.c (dwarf2dbg.h): Include
1216 (md_pseudo_table): Handle .loc and .file
1217 (md_assemble): Call dwarf2_emit_insn if BFD_ASSEMBLER.
1218 * Makefile.am: Add dependency on dwarf2dbg.h for h8300 targets.
1219 * Makefile.in: Regenerate.
1223 * config/tc-h8300.c (Nmode, h8300hnmode, h8300snmode): New.
1224 (md_pseudo_table): Add h8300hn, h8300sn.
1225 * config/tc-h8300.h (COFF_MAGIC): Handle h8300hn, h8300sn.
1226 * doc/c-h8300.texi : Add documentation for new machine directives.
1230 * config/tc-ppc.c (ppc_symbol_chars): Define.
1231 * config/tc-ppc.h (tc_symbol_chars): Define.
1235 * config/tc-sh.c: Amend comment to refer to SuperH.
1236 * config/tc-sh.h: Likewise.
1237 (LISTING_HEADER): Amend to refer to SuperH.
1238 * config/tc-sh64.c: Change comment to refer to SuperH.
1239 * config/tc-sh64.h (LISTING_HEADER): Change to refer to SuperH.
1240 * doc/as.texinfo [SH, GENERIC]: Amend / Change to refer to SuperH.
1241 * doc/c-sh.texi: Amend to refer to SuperH.
1242 Add SuperH architecture documentation references.
1243 * doc/c-sh64.texi: Change to refer to SuperH.
1247 * app.c (do_scrub_chars): More checks for valid labels.
1251 * app.c (do_scrub_chars): Check for valid label.
1255 * doc/as.texinfo: Replace references to Mitsubishi M32R with
1256 references to Renesas M32R.
1257 * doc/c-m32r.texi: Likewise.
1261 * dwarf2dbg.c (get_filenum): Skip as-yet unassigned file numbers.
1262 (out_file_list): Assign non-null filename after generating error.
1266 * ehopt.c (check_eh_frame): For aug_size == 0
1267 in state_seeing_aug_size state skip the state_skipping_aug
1272 * doc/c-h8300.texi: Replace occurrances of 'Hitachi' with
1274 * doc/c-h8500.texi: Likewise.
1275 * doc/c-sh.texi: Likewise.
1276 * doc/c-sh64.texi: Likewise.
1277 * doc/h8.texi: Likewise.
1278 * config/tc-h8300.c: Likewise.
1279 * config/tc-h8300.h: Likewise.
1280 * config/tc-h8500.c: Likewise.
1281 * config/tc-h8500.h: Likewise.
1282 * config/tc-sh.c: Likewise.
1283 * config/tc-sh.h: Likewise.
1284 * config/tc-sh64.c: Likewise.
1285 * config/tc-sh64.h: Likewise.
1289 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): New.
1290 * config/tc-mips.c: Use signed add for n32 address arithmetic.
1291 (append_insn): When filling delay slots with instructions
1292 that have fixups that tc_gen_reloc might consider modifyable
1293 in variant frags, start a new frag.
1294 (load_address): Generate GOT_DISP with of without offset
1295 depending on whether symbol is local. For -xgot, use
1296 GOT_PAGE/GOT_OFST or GOT_HI16/GOT_LO16.
1297 (macro) <M_DLA_AB, M_LA_AB>: Likewise.
1298 <M_JAL_A>: In NewABI, use CALL16 or GOT_DISP for small got,
1299 CALL_HI16/CALL_LO16 or GOT_PAGE/GOT_OFST for big got.
1300 <ld_st>: In NewABI with small got, always use
1301 GOT_PAGE/GOT_OFST, with the latter in the load/store
1302 instruction. With big got, use GOT_HI16/GOT_LO16 or
1304 (tc_gen_reloc): Adjust variant frags with GOT_DISP in NewABI.
1305 Add tc_frag_data.tc_fr_offset to addends. Decay CALL16,
1306 GOT_OFST and GOT_DISP to GOT_DISP in NewABI.
1307 (md_convert_frag): Use memmove for safe copying of overlapping
1312 * doc/c-m68hc11.texi (M68HC11-Opts): Document -m68hcs12, -mshort,
1313 -mlong, -mshort-double and -mlong-double options; use table @code.
1314 (M68HC11-Syntax): Update to document 68HC12 operands.
1315 (M68HC11-Modifiers): New section for operand modifiers.
1316 (M68HC11-Directives): New section for specific assembler directives.
1317 (M68HC11-Branch): Fix Overfull hbox error.
1321 * config/tc-mips.c (macro): Add comments explaining the rationale
1326 * config/tc-mips.c (macro): Put back `+ 0x8000' in test for 64-bit
1327 constant address that Alexandre took out by accident. Reject
1328 64-bit addresses that are not sign extensions of 32 bits only if
1329 we don't support 64-bit address constants.
1333 * config/tc-ppc.c (md_apply_fix3): Generate ADDR16 relocs.
1337 * as.c (perform_an_assembly_pass): If using cgen, call
1339 * cgen.c (gas_cgen_begin): New function. If
1340 flag_signed_overflow_ok is set call cgen_set_signed_overflow_ok
1341 otherwise call cgen_clear_signed_overflow_ok.
1342 * cgen.h: Prototype gas_cgen_begin.
1346 * write.c (write_relocs): Remove unused variable.
1350 * config/tc-mips.c (HAVE_64BIT_ADDRESS_CONSTANTS): New.
1351 (macro): Use new macro to decide whether to emit constant address
1352 as 32 or 64 bits if addresses are 32-bit wide but registers are
1357 * config/tc-m68hc11.c (M6811_OP_CALL_ADDR): New internal define.
1358 (M6811_OP_PAGE_ADDR): New internal define.
1359 (get_operand): New modifier %page and %addr to obtain page and
1360 address part of a far-function.
1361 (fixup8): Use BFD_RELOC_M68HC11_PAGE for a %page modifier; don't
1362 complain on overflow for the BFD_RELOC_M68HC11_PAGE and truncation
1364 (fixup16): Use BFD_RELOC_M68HC11_LO16 for a %addr modifier.
1365 (find_opcode): Add comment.
1366 (md_estimate_size_before_relax): Force relocation of
1367 STATE_UNDEXED_OFFSET types when the symbol is not absolute.
1368 (tc_m68hc11_fix_adjustable): Check for BFD_RELOC_M68HC11_LO16
1369 instead of BFD_RELOC_LO16; temporarily make the BFD_RELOC_32
1370 on the symbol itself so that DWARF2 strings are merged correctly.
1374 * config/obj-coff.h (TARGET_FORMAT): Namespace cleanup, changed
1375 default tic4x target format to 'coff2-tic4x'.
1376 * config/tc-tic4x.c: Namespace cleanup. Replace s/c4x/tic4x/ and
1378 * config/tc-tic4x.h: Ditto
1382 * NEWS: Mention support for Xtensa architecture.
1386 * config/tc-arm.c (arm_force_relocation): Return 0 for OFFSET_IMM.
1390 * config/tc-mips.c (macro2): Adjust implementation of
1391 M_ULH, M_ULHU, M_ULW, and M_ULD so that they work properly
1392 in the case where the source and destination registers
1397 * Makefile.am (CPU_TYPES): Add xtensa.
1398 (TARGET_CPU_CFILES): Add config/tc-xtensa.c.
1399 (TARGET_CPU_HFILES): Add config/tc-xtensa.h.
1400 (xtensa-relax.o): New target.
1402 * Makefile.in: Regenerate.
1403 * configure.in: Handle xtensa-*-*. Add xtensa-relax.o to
1404 extra_objects for xtensa targets.
1405 * configure: Regenerate.
1406 * write.c (write_object_file): Add new md_post_relax_hook.
1407 * config/tc-xtensa.c: New file.
1408 * config/tc-xtensa.h: Likewise.
1409 * config/xtensa-istack.h: Likewise.
1410 * config/xtensa-relax.c: Likewise.
1411 * config/xtensa-relax.h: Likewise.
1412 * doc/Makefile.am (CPU_DOCS): Add c-xtensa.texi.
1413 * doc/Makefile.in: Regenerate.
1414 * doc/all.texi: Set new XTENSA variable.
1415 * doc/as.texinfo: Set new Xtensa variable. Describe
1416 Xtensa-specific options. Define line comment character for
1417 Xtensa. Add Xtensa processors to list of ELF targets where
1418 alignment is specified in bytes. Add new Xtensa-Dependent node.
1419 Add acknowledgements for those contributing to the Xtensa port.
1420 * doc/internals.texi: Describe new md_post_relax_hook.
1421 * doc/c-xtensa.texi: New file.
1426 * config/tc-arm.c: Remove presence of (r) and (tm) symbols.
1427 (ARM_ARCH_IWMMXT): Simplify.
1428 (insns): Place iwmmx instructions in correct place in table.
1429 (arm_add_note): New function: Add a note entry to a .note section.
1430 (md_begin): Make the default architecture be unknown.
1431 Suppress the creation of an arm note section.
1435 * config/tc-mips.c (nopic_need_relax): Check for S_IS_EXTERN.
1440 Contribute support for Intel's iWMMXt chip - an ARM variant:
1442 * config/tc-arm.c: (ARM_CEXT_IWMMXT, ARM_ARCH_IWMMXT, WR_PREFIX,
1443 WC_PREFIX, REG_TYPE_IWMMXT): New constants.
1444 (enum wreg_type, enum iwmmxt_insn_type): New types.
1445 (wr_register, wc_register, wcg_register): New macros.
1446 (iwmmxt_table): New variable.
1447 (wreg_required_here, do_iwmmxt_byte_addr, do_iwmmxt_tandc,
1448 do_iwmmxt_tbcst, do_iwmmxt_textrc, do_iwmmxt_textrm,
1449 do_iwmmxt_tinsr, do_iwmmxt_tmcr, do_iwmmxt_tmcrr, do_iwmmxt_tmia,
1450 do_iwmmxt_tmovmsk, do_iwmmxt_tmrc, do_iwmmxt_tmrrc,
1451 do_iwmmxt_torc, do_iwmmxt_waligni, do_iwmmxt_wmov,
1452 do_iwmmxt_word_addr, do_iwmmxt_wrwr, do_iwmmxt_wrwrwcg,
1453 do_iwmmxt_wrwrwr, do_iwmmxt_wshufh, do_iwmmxt_wzero,
1454 cp_byte_address_offset, cp_byte_address_required_here,
1455 check_iwmmxt_insn): New functions.
1456 (asm_opcode_insns): Add iWMMXt instructions.
1457 (md_begin): Set the mach value for iWMMXt targets. Create a note
1458 section to identify iwmmxt binaries.
1459 (md_apply_fix3): Handle BFD_RELOC_ARM_CP_OFF_IMM_S2.
1460 * doc/c-arm.texi: Document the support for the iWMMXt.
1461 * NEWS: Mention new support.
1465 * doc/as.texinfo: Rename the all occurances of C54X to TIC54X.
1466 * doc/all.texi: Likewise.
1467 * doc/c-tic54x.texi: Likewise.
1471 * config/tc-ia64.c (generate_unwind_image): Fix type of unw_rec to
1472 avoid aliasing issue.
1476 * config/tc-s390.c (s390_arch_size): Initialize to zero.
1477 (current_arch_mask): Rename to current_mode_mask.
1478 (current_arch_requested): Remove variable.
1479 (current_cpu): New variable.
1480 (init_default_arch): Set defaults values for s390_arch_size,
1481 current_mode_mask and current_cpu.
1482 (md_parse_option): New options -mesa, -mzarch and -march={g5,g6,z900}.
1483 (md_begin): Replace current_arch_mask by current_cpu.
1484 (md_assemble): Adapt check and error message to current_mode_mask and
1489 * macro.c (buffer_and_nest): Store more to sb instead of '\n'.
1490 * read.c (get_line_sb): Return end of line character or '\n' if
1491 it is zero or non-existent.
1495 * config/tc-mips.c (mips_validate_fix): New function.
1496 * config/tc-mips.h (TC_VALIDATE_FIX): Define.
1497 (mips_validate_fix): Declare.
1501 * Reverted 2003-03-02's patch.
1505 * dwarf2dbg.c (generic_dwarf2_emit_offset): New.
1506 (TC_DWARF2_EMIT_OFFSET): Provide default.
1507 (out_debug_aranges, out_debug_info): Use it.
1508 * config/tc-ia64.c (ia64_dwarf2_emit_offset): New.
1509 (ia64_cons_fix_new): Move FUNC_DTP_RELATIVE handling ...
1510 (ia64_gen_real_reloc_type): ... here.
1511 * config/tc-ia64.h (TC_DWARF2_EMIT_OFFSET): New.
1515 * config/tc-mips.c (s_mips_end): Remove !BFD_ASSEMBLER case.
1516 (s_mips_ent): Likewise.
1520 * config/tc-msp430.c (mcu_types): Add recently announced x1122
1521 and x1123 devices, add missed x437.
1522 (md_show_usage): Sort device list.
1526 * config/tc-sh.c (sh_dsp): Replace with preset_target_arch.
1527 (md_begin): Use preset_target_arch.
1528 (md_longopts): Make isa option unconditional.
1529 (md_parse_option): Make OPTION_DSP and OPTION_ISA sh4 / any
1530 set preset_target_arch.
1531 (md_apply_fix3): If BFD_ASSEMBLER, adjust SWITCH_TABLE fixups
1532 by -S_GET_VALUE (fixP->fx_subsy).
1533 (tc_gen_reloc): For SWITCH_TABLE fixups, the symbol is fixp->fx_subsy,
1534 and the addend is 0.
1535 Adjust addend of R_SH_IND12W relocations by fixp->fx_offset - 4.
1536 * config/tc-sh.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
1540 * config/tc-mips.c (append_insn): Add handling of
1541 BFD_RELOC_MIPSEMB_16_PCREL_S2. Avoid emitting unneeded
1542 BFD_RELOC_16_PCREL_S2 relocs and add earlier warnings about
1543 misaligned address and reange overflow.
1544 (macro_build): Add handling of BFD_RELOC_MIPSEMB_16_PCREL_S2. Add
1545 earlier warnings about misaligned address and reange overflow.
1546 (mips_ip): Add handling of BFD_RELOC_MIPSEMB_16_PCREL_S2.
1547 (md_apply_fix): Likewise. Fix warning output.
1548 (tc_gen_reloc): Add handling of BFD_RELOC_MIPSEMB_16_PCREL_S2.
1549 Allow BFD_RELOC_16_PCREL_S2 for all ABIs.
1550 (md_convert_frag): Add handling of BFD_RELOC_MIPSEMB_16_PCREL_S2.
1554 * NEWS: Mention availability of test generator program.
1558 * config/tc-v850.c (system_registers): Add v850e debug registers.
1559 (system_register_name): Accept up to 27 (the last v850e sys register).
1563 * doc/as.texinfo: Define new COFF-ELF variable to conditionalize text
1564 relevant to both COFF and ELF. Fix obvious typos and texinfo bugs.
1565 Capitalize section headings consistently. Format index entries more
1566 consistently. Unconditionalize text about whether text and data
1567 sections are alterable. Use @ifnottex for alternatives to @tex output
1568 so that HTML works. Clean up COFF vs. ELF descriptions of .section,
1569 .size and .type directives. Be more polite about bad bug reports.
1570 Move FDL into a separate file.
1571 * doc/fdl.texi: New file.
1575 * config/tc-mips.c (prev_reloc_op_frag): New variable.
1576 (macro): Check it to decide whether a new frag is needed.
1577 (my_getSmallExpression): Set it.
1581 * cgen.c: Fix typo: intial -> initial.
1585 * app.c (do_scrub_chars): Handle '||' in two states.
1589 * write.c (TC_FORCE_RELOCATION_SUB_SAME): Revert last change.
1590 * config/tc-s390.h (TC_FORCE_RELOCATION_SUB_SAME): Define.
1594 * configure.in: Add support for kaOS as cross build target system.
1595 * configure: Regenerated.
1599 * config/tc-arm.c (md_begin): If the Maverick co-processor is
1600 selected, set the EF_ARM_MAVERICK_FLOAT flag and
1601 bfd_mach_arm_ep9312 machine number.
1605 * config/tc-mips.c (reloc_needs_lo_p): New function.
1606 (fixup_has_matching_lo_p): New function.
1607 (append_insn): Use reloc_needs_lo_p to check whether a relocation
1608 might need a matching %lo(). Reuse the head of mips_hi_fixup_list
1609 if that fixup already has a matching %lo(). Don't call frag_wane here.
1610 (macro): Call frag_wane here if the last unmatched hi was in the
1612 (pic_need_relax): New function, split out from...
1613 (md_estimate_size_before_relax): ...here.
1614 (mips_frob_file): Use reloc_needs_lo_p. Use pic_need_relax to test
1615 whether BFD_RELOC_MIPS_GOT16 fixups refer to global symbols.
1619 * config/tc-mips.c (my_getSmallExpression): Rework bracket handling.
1623 * config/tc-ppc.c (ppc_elf_suffix): Undo part of last change so that
1628 * config/tc-ppc.c (mapping): Handle new TLS reloc specs.
1629 (ppc_elf_suffix): Don't warn for x+off@got when ppc64 and don't
1630 accept x@got+off etc.
1631 (md_assemble): Handle TLS relocs.
1632 (ppc_force_relocation): Force for all TLS relocs.
1633 (ppc_fix_adjustable): Likewise.
1634 (md_apply_fix3): Handle TLS relocs.
1638 * config/obj-elf.c (obj_elf_change_section): Set SEC_LINK_ONCE and
1639 SEC_LINK_DUPLICATES_DISCARD directly rather than using elf_linkonce_p.
1643 * config/tc-mips.c (enum small_ex_type): Remove.
1644 (imm_unmatched_hi): Remove.
1645 (md_assemble): Remove use of imm_unmatched_hi. Remove the last
1646 argument from calls to append_insn.
1647 (append_insn): Remove unmatched_hi parameter; check reloc_type[0]
1649 (macro_build): Update append_insn calls.
1650 (mips16_macro_build, macro_build_lui): Likewise.
1651 (mips_ip): Rework handling of small expressions. Move explicit
1652 relocation handling into my_getSmallExpression. Assume that the
1653 value of 'o' operands is zero if there is only one bracketed
1655 (percent_op): Make constant. Record the BFD relocation code
1656 associated with each operator.
1657 (my_getSmallParser, my_getPercentOp): Remove.
1658 (parse_relocation): New function.
1659 (my_getSamllExpression): Rework. Fill in relocations here
1660 rather than in mips_ip.
1664 * config/tc-i386.c (line_comment_chars): Add '#'. This makes the
1665 assembler's handling of # <linenum> "<filename>" directives work.
1669 * dwarf2dbg.c: Include filenames.h.
1670 (struct file_entry): Make filename const char *.
1671 (dirs, dirs_in_use, dirs_allocated): New variables.
1672 (get_filenum): Add NUM argument. Build directory table.
1673 (dwarf2_where): Adjust caller.
1674 (dwarf2_directive_file): Use get_filenum to allocate
1675 slot in file and directory tables.
1676 (dwarf2_directive_loc): Recreate full filename from
1677 directory and filename part if needed for listing.
1678 (out_file_list): Output directory table.
1679 Output main source file dirname before its filename.
1683 * config/tc-msp430.c: Replace occurences of 'tolower' with
1688 * config/tc-ia64.c (dot_vframesp): Correct error message.
1689 (dot_vframepsp): Ditto.
1693 * configure.in (em): Set to irix on all Irix systems.
1694 * configure: Rebuilt.
1695 * config/te-irix.h: New file.
1696 * config/tc-mips.c (mips_dwarf2_format): Use TE_IRIX to decide
1697 whether to use Irix-specific 64-bit format.
1701 * config/tc-s390.c (s390_elf_cons): Avoid designated initializers.
1705 * config/tc-sparc.c (sparc_ip): Handle TLS % operators.
1706 (tc_gen_reloc): Handle TLS relocs.
1707 (sparc_cons, cons_fix_new_sparc): Handle %r_tls_dtpoff.
1708 * config/tc-sparc.h (tc_fix_adjustable): Don't adjust TLS
1710 * config/obj-elf.c (obj_elf_section_word): Handle tls.
1711 (obj_elf_type): Handle tls_object.
1715 * config/tc-s390.c (s390_tls_suffix): New function.
1716 (elf_suffix_type): Add suffix enums for TLS relocations.
1717 (s390_elf_suffix): Add suffix strings for TLS relocations.
1718 (s390_elf_cons): Map new lenght/elf suffix combinations for TLS to
1720 (md_gather_operands): Map new instruction operand/elf suffix
1721 combinations for TLS to bfd relocations.
1722 (tc_s390_fix_adjustable): Add new TLS relocations.
1723 (tc_s390_force_relocation): Likewise.
1724 (md_apply_fix3): Likewise.
1728 * as.h: Update copyright date.
1729 * symbols.c: Likewise.
1730 * config/tc-d10v.h: Likewise.
1731 * config/tc-fr30.h: Likewise.
1732 * config/tc-i960.h: Likewise.
1733 * config/tc-mips.h: Likewise.
1735 * config/tc-hppa.h (DIFF_EXPR_OK): Define.
1736 (MD_APPLY_SYM_VALUE): Move.
1740 * NEWS: Announce sh2e support.
1744 * config/tc-sh.c (md_show_usage): Added sh2e next to sh3e.
1745 (sh_elf_final_processing): Handle arch_sh2e.
1749 * symbols.c (S_FORCE_RELOC): Add "strict" param.
1750 * symbols.h (S_FORCE_RELOC): Likewise.
1751 * config/obj-aout.h (S_FORCE_RELOC): Likewise.
1752 * config/obj-bout.h (S_FORCE_RELOC): Likewise.
1753 * config/obj-coff.h (S_FORCE_RELOC): Likewise.
1754 * config/obj-ieee.h (S_FORCE_RELOC): Likewise.
1755 * config/obj-vms.h (S_FORCE_RELOC): Likewise.
1756 * write.c (generic_force_reloc): New function.
1757 (TC_FORCE_RELOCATION): Use it here instead of S_FORCE_RELOC.
1758 (TC_FORCE_RELOCATION_SUB_SAME): Test TC_FORCE_RELOCATION too.
1759 (adjust_reloc_syms): Adjust S_FORCE_RELOC call.
1760 * as.h (generic_force_reloc): Declare.
1761 * doc/internals.texi (S_FORCE_RELOC): Update.
1762 (TC_FORCE_RELOCATION_SUB_SAME): Update.
1764 * config/tc-alpha.c (alpha_force_relocation): Adjust to use
1765 generic_force_reloc.
1766 (alpha_fix_adjustable): Likewise.
1767 * config/tc-arm.c (arm_force_relocation): Likewise.
1768 * config/tc-cris.c (md_cris_force_relocation): Likewise.
1769 * config/tc-frv.c (frv_force_relocation): Likewise.
1770 * config/tc-i386.c (md_apply_fix3): Likewise.
1771 * config/tc-ia64.c (ia64_force_relocation): Likewise.
1772 * config/tc-ip2k.c (ip2k_force_relocation): Likewise.
1773 * config/tc-m32r.c (m32r_force_relocation): Likewise.
1774 * config/tc-m68hc11.c (tc_m68hc11_force_relocation): Likewise.
1775 * config/tc-mcore.c (mcore_force_relocation): Likewise.
1776 * config/tc-mips.c (mips_force_relocation): Likewise.
1777 * config/tc-mmix.c (mmix_force_relocation): Likewise.
1778 * config/tc-ppc.c (ppc_force_relocation): Likewise.
1779 * config/tc-s390.c (tc_s390_force_relocation): Likewise.
1780 * config/tc-sh.c (sh_force_relocation): Likewise.
1781 (md_pcrel_from_section): Likewise.
1782 * config/tc-sparc.c (tc_gen_reloc): Likewise.
1783 * config/tc-v850.c (v850_force_relocation): Likewise.
1784 * config/tc-xstormy16.c (xstormy16_force_relocation): Likewise.
1785 * config/tc-i386.h (TC_FORCE_RELOCATION): Likewise.
1786 * config/tc-mcore.h (TC_FORCE_RELOCATION): Likewise.
1787 * config/tc-sparc.h (tc_fix_adjustable): Likewise.
1789 * config/tc-d10v.c (d10v_force_relocation): Delete.
1790 * config/tc-d10v.h (TC_FORCE_RELOCATION): Don't define.
1791 * config/tc-dlx.c (md_dlx_force_relocation): Delete.
1792 * config/tc-dlx.h (TC_FORCE_RELOCATION): Don't define.
1793 * config/tc-fr30.c (fr30_force_relocation): Delete.
1794 * config/tc-fr30.h (TC_FORCE_RELOCATION): Don't define.
1795 * config/tc-mn10300.c (mn10300_force_relocation): Delete.
1796 * config/tc-mn10300.h (TC_FORCE_RELOCATION): Don't define.
1797 (TC_FORCE_RELOCATION_SUB_SAME): Test TC_FORCE_RELOCATION too.
1798 * config/tc-i960.h (TC_FORCE_RELOCATION_SUB_SAME): Likewise.
1799 * config/tc-hppa.c (hppa_force_relocation): Adjust S_FORCE_RELOC call.
1800 * config/tc-mips.c (RELAX_BRANCH_TOOFAR): Warning fix.
1801 * config/tc-mips.h (TC_FORCE_RELOCATION_SUB_SAME): Don't define.
1802 * config/tc-openrisc.c (openrisc_force_relocation): Delete.
1803 * config/tc-openrisc.h (TC_FORCE_RELOCATION): Don't define.
1804 * config/tc-sparc.c (elf32_sparc_force_relocation): Delete.
1805 * config/tc-sparc.h (TC_FORCE_RELOCATION): Don't define for ELF.
1806 * config/tc-i386.c (i386_force_relocation): Delete.
1807 * config/tc-i386.h (TC_FORCE_RELOCATION): Don't define for
1809 (EXTERN_FORCE_RELOC): Fix TE_PE and STRICT_PE_FORMAT nesting.
1810 * config/tc-m68k.h (TC_FORCE_RELOCATION): Don't define.
1811 * config/tc-pj.h (TC_FORCE_RELOCATION): Don't define.
1812 * config/tc-sh.h (TC_FORCE_RELOCATION_SUB_ABS): Don't call
1814 (TC_FORCE_RELOCATION_SUB_SAME): Test TC_FORCE_RELOCATION too.
1815 * config/tc-sh64.h (TC_FORCE_RELOCATION_SUB_SAME): Likewise.
1819 * config/tc-sh64.c (shmedia_frob_section_type): Adjust for changed
1820 sh64_elf_section_data.
1821 * config/tc-sh64.h: Include elf32-sh64.h.
1822 * config/tc-m68hc11.c: Don't include stdio.h.
1823 (md_show_usage): Fix missing continuation.
1824 * Makefile.am: Run "make dep-am".
1825 * Makefile.in: Regenerate.
1829 * as.h: Include fopen-bin.h not fopen-same.h for mingw32 hosts.
1833 * configure.in: Handle *-*-aros*.
1834 * configure: Regenerated from configure.in.
1838 * config/tc-s390.c (elf_suffix_type): Add suffix enums for gotoff,
1839 gotplt and pltoff relocations.
1840 (s390_elf_suffix): Add suffix strings for gotoff, gotplt and pltoff.
1841 (s390_elf_cons): Map new lenght/elf suffix combinations for gotoff,
1842 gotplt and pltoff to bfd relocations.
1843 (md_gather_operands): Map new instruction operand/elf suffix
1844 combinations to bfd relocations.
1845 (tc_s390_fix_adjustable): Add new gotoff, gotplt and pltoff relocations
1846 to the list of unadjustable relocations.
1847 (tc_s390_force_relocation): Always emit relocations for gotoff, gotplt
1848 and pltoff relocations.
1849 (md_apply_fix3): Add the new relocations.
1853 * config/tc-s390.c (md_apply_fix3): Emit error message for relocations
1854 with a subsy symbol.
1858 * config/tc-m68hc11.c (tc_m68hc11_fix_adjustable): Prevent adjustment
1859 of relocs for memory bank addressing.
1863 * config/tc-m68hc11.c (md_show_usage): Update usage.
1864 (md_parse_option): Recognize -m68hcs12.
1865 (m68hc11_elf_final_processing): Set EF_M68HCS12_MACH flag to identify
1867 * doc/as.texinfo (Overview): Document new option -m68hcs12.
1871 * config/tc-ia64.c (ia64_cons_fix_new): Handle @dtprel() in data.
1875 * Makefile.am: Run "make dep-am".
1876 * Makefile.in: Regenerate.
1877 * po/POTFILES.in: Regenerate.
1881 * read.c (get_absolute_expr): New, split out from..
1882 (get_absolute_expression): ..here.
1883 * read.h (get_absolute_expr): Declare.
1884 * config/obj-elf.c (elf_common): Use offsetT for "temp" and "size".
1885 Trim size to arch bits_per_address, and test for negative input
1886 via get_absolute_expr.
1890 * config/tc-xstormy16.c (md_cgen_lookup_reloc): Adjust value based
1892 (xstormy16_md_apply_fix3): Use adjustment.
1896 * configure.in: Add iq2000-elf target.
1897 * configure: Regenerate.
1898 * config/tc-iq2000.c: New file.
1899 * config/tc-iq2000.h: Likewise.
1900 * po/gas.pot: Regenerate.
1904 * config/tc-mips.c: Update copyright years to include 2003.
1905 (mips_ip): Fix indentation of "+A", "+B", and "+C" handling.
1906 Additionally, clean up their code slightly and clean up their
1909 * doc/c-mips.texi: Add MIPS32r2 to ".set mipsN" documentation.
1913 * doc/Makefile.am (as.1): Depend on "asconfig.texi gasver.texi
1915 * doc/Makefile.in: Regenerate.
1919 * config/obj-elf.c (special_sections): Work around HP's incorrect usage
1920 of .init and .fini sections for array initializers and finalizers.
1924 * config/tc-mips.c (validate_mips_insn, mips_ip): Recognize
1925 the "+D" operand, which will be used only by the disassembler.
1929 * configure.in: Recognize mipsisa32r2, mipsisa32r2el, and
1931 * configure: Regenerate.
1932 * config/tc-mips.c (ISA_HAS_DROR, ISA_HAS_ROR): New defines.
1933 (macro_build): Handle "K" operand.
1934 (macro2): Use ISA_HAS_DROR and ISA_HAS_ROR in the places where
1935 CPU_HAS_DROR and CPU_HAS_ROR are currently used.
1936 (mips_ip): New variable "lastpos", and implement "+A", "+B",
1937 and "+C" operands for MIPS32 Release 2 ins/ext instructions.
1938 Implement "K" operand for MIPS32 Release 2 rdhwr instruction.
1939 (validate_mips_insn): Implement "+" as a way to extend the
1940 allowed operands, and implement "K", "+A", "+B", and "+C"
1942 (OPTION_MIPS32R2): New define.
1943 (md_longopts): Add entry for OPTION_MIPS32R2.
1944 (OPTION_ELF_BASE): Adjust to accomodate OPTIONS_MIPS32R2.
1945 (md_parse_option): Handle OPTION_MIPS32R2.
1946 (s_mipsset): Reimplement handling of ".set mipsN" options
1947 and add support for ".set mips32r2".
1948 (mips_cpu_info_table): Add entry for "mips32r2" (MIPS32 Release 2).
1949 (md_show_usage): Document "-mips32r2" option.
1950 * doc/as.texinfo: Document "-mips32r2" option.
1951 * doc/c-mips.texi: Likewise.
1955 * configure.in: Add msp430 target.
1956 * configure: Regenerate.
1957 * Makefile.am: Add msp430 target.
1958 * Makefile.in: Regenerate.
1959 * config/tc-msp430.c: New file: msp430 assembler.
1960 * config/tc-msp430.h: New file: target macros for msp430.
1961 * doc/Makefile.am: Add msp430 target.
1962 * doc/Makefile.in: Regenerate.
1963 * doc/as.texinfo: Include msp430 documenation.
1964 * doc/all.texi: Enable msp430 documentation.
1965 * doc/c-msp430.texi: New file: document msp430 specific features
1970 * dwarf2dbg.c (DWARF2_ADDR_SIZE): New macro.
1971 (dwarf2_finish): Use it.
1972 * doc/internals.texi (DWARF2_ADDR_SIZE): Document it.
1973 * config/tc-mips.h (DWARF2_ADDR_SIZE): Override.
1977 * config/tc-xstormy16.c (md_cgen_lookup_reloc): Support
1978 BFD_RELOC_XSTORMY16_12.
1982 * doc/as.texinfo (Invoking): Typo fix.
1983 * config/tc-tic54x.c (encode_operand): Comment typo fix.
1987 * doc/c-alpha.texi: Fix typos.
1988 * doc/c-arm.texi: Likewise.
1989 * doc/c-d10v.texi: Likewise.
1990 * doc/c-i370.texi: Likewise.
1991 * doc/c-i960.texi: Likewise.
1992 * doc/c-ia64.texi: Likewise.
1993 * doc/c-mmix.texi: Likewise.
1994 * doc/c-ns32k.texi: Likewise.
1995 * doc/c-pdp11.texi: Likewise.
1996 * doc/c-pj.texi: Likewise.
1997 * doc/c-sh64.texi: Likewise.
1998 * doc/c-sparc.texi: Likewise.
1999 * doc/c-tic54x.texi: Likewise.
2000 * doc/c-v850.texi: Likewise.
2001 * doc/c-vax.texi: Likewise.
2002 * doc/internals.texi: Likewise.
2006 * config/tc-mips.c (macro): In M_DROL, M_DROR, M_ROL, and M_ROR,
2007 use hardware rotate ops as appropriate. In M_DROL_I, M_DROR_I,
2008 M_ROL_I, and M_ROR_I, simplify code, clean up warnings, and
2009 arrange not to issue warnings about use of AT when AT is not
2014 * as.c (std_longopts): Duplicate --keep-locals entry in order to
2015 prevent it being confused with -k.
2019 * config/tc-xstormy16.c (md_cgen_lookup_reloc): If a relocation
2020 has already been set up, use it.
2024 * ChangeLog-9295: Fix a typo.
2026 * config/tc-d10v.c: Fix a comment typo.
2027 * config/tc-dlx.c: Likewise.
2028 * config/tc-h8300.h: Likewise.
2029 * config/tc-h8500.h: Likewise.
2030 * config/tc-mips.c: Likewise.
2031 * config/tc-s390.c: Likewise.
2032 * config/tc-sh.h: Likewise.
2033 * config/tc-tic80.h: Likewise.
2034 * config/tc-w65.h: Likewise.
2035 * config/tc-z8k.c: Likewise.
2036 * config/tc-z8k.h: Likewise.
2037 * testsuite/gas/h8300/cmpsi2.s: Likewise.
2041 * config/tc-d30v.c (check_range): Warning fixes, formatting.
2042 Simplify sign extension. Remove redundant unsigned < 0 test.
2043 * config/tc-i960.c (md_ri_to_chars): Prototype.
2044 * config/tc-mcore.c (md_pseudo_table): Fix typo.
2045 (dump_literals): Init brarsym, and test later instead of isforce.
2047 * config/tc-ns32k.c (encode_operand): Constify operandsP and suffixP.
2048 (parse): Constify line and lineptr.
2049 (md_begin): Calculate endop here.
2053 * config/obj-vms.c: Formatting. Include fnctl.h.
2054 (Create_VMS_Object_File): Fix creat call for sane unix systems.
2055 (Object_Record_Offset): Make it a size_t.
2056 (Flush_VMS_Object_Record_Buffer): Fix signed/unsigned warning.
2057 (VMS_TBT_Routine_End <Size>): Make var unsigned long.
2058 (VMS_Fix_Indirect_Reference <Offset>): Make arg addressT.
2059 (synthesize_data_segment <data_size>): Remove ATTRIBUTE_UNUSED.
2060 (vms_fixup_data_section <data_size>): Add here instead.
2061 * config/e-criself.c: Fix typo in last change.
2065 * write.c (write_object_file): Fix signed/unsigned warning.
2066 * config/e-crisaout.c (crisaout_bfd_name): Prototype.
2067 * config/e-criself.c (criself_bfd_name): Prototype.
2068 * config/obj-aout.c (s_sect): Remove unused function.
2069 * config/obj-bout.c (obj_bout_line <ignore>): Add ATTRIBUTE_UNUSED.
2070 * config/obj-coff.c (coff_last_bf): Don't declare for OBJ_XCOFF.
2071 (fixup_mdeps <h>): Add ATTRIBUTE_UNUSED.
2072 * config/obj-ecoff.c (ecoff_frob_file <addr>): Likewise.
2073 * config/obj-vms.c (setup_basic_type <spnt>): Likewise.
2074 (VMS_RSYM_Parse <Current_Routine>): Likewise.
2075 (vms_fixup_text_section <text_siz>): Likewise.
2076 (synthesize_data_segment <data_size>): Likewise.
2077 (vms_fixup_xtors_section <sect_no>): Likewise.
2078 (structure_count): Don't use implicit int type.
2079 * config/tc-a29k.c (insert_sreg): Prototype.
2080 (define_some_regs): Prototype, make static.
2081 (parse_operand): Likewise.
2082 (md_parse_option <c,arg>): Add ATTRIBUTE_UNUSED.
2083 (md_show_usage <stream>): Likewise.
2084 (md_section_align <segment>): Likewise.
2085 (md_convert_frag <all args>): Likewise.
2086 (md_estimate_size_before_relax <all args>): Likewise.
2087 (md_apply_fix3): Don't cast valP pointer type. Fix bogus >>='s.
2088 * config/tc-arm.c (arm_validate_fix): Only for OBJ_COFF or OBJ_ELF.
2089 * config/tc-d30v.c (md_parse_option <arg>): Add ATTRIBUTE_UNUSED.
2090 (md_undefined_symbol <name>): Likewise.
2091 (md_convert_frag <all args>): Likewise.
2092 (write_long <opcode>): Likewise.
2093 (tc_gen_reloc <seg>): Likewise.
2094 (md_estimate_size_before_relax <all args>): Likewise.
2095 (md_apply_fix3 <seg>): Likewise.
2096 (s_d30v_align <ignore>): Likewise.
2097 (build_insn): Correct format string.
2098 (md_apply_fix3): Likewise.
2099 * config/tc-fr30.c (md_parse_option <c,arg>): Add ATTRIBUTE_UNUSED.
2100 (md_undefined_symbol <name>): Likewise.
2101 (md_convert_frag <all args>): Likewise.
2102 (md_cgen_lookup_reloc <insn>): Likewise.
2103 (md_begin): Delete unused vars.
2104 (md_assemble): Likewise.
2105 (md_estimate_size_before_relax): Likewise.
2106 (fr30_relax_frag): #if 0 out, seems unused.
2107 (md_atof): Remove declaration of atof_ieee.
2108 (restore_colon): Prototype.
2109 * config/tc-frv.c (frv_insert_vliw_insn): Prototype.
2110 (frv_find_in_vliw): Likewise.
2111 (frv_debug_tomcat): Likewise.
2112 (frv_adjust_vliw_count): Likewise.
2113 (frv_tomcat_shuffle): Likewise.
2114 (frv_tomcat_analyze_vliw_chains): Likewise. Correct args to
2115 frv_find_in_vliw call.
2116 (md_atof): Remove declaration of atof_ieee.
2117 * config/tc-h8500.c (cons): Delete declaration.
2118 (md_begin <opcode>): Constify.
2119 (displacement_size, immediate_size, absolute_size): Remove.
2120 (build_relaxable_instruction <operand>): Add ATTRIBUTE_UNUSED.
2121 (tc_crawl_symbol_chain <headers>): Likewise.
2122 (md_undefined_symbol <name>): Likewise.
2123 (tc_headers_hook <headers>): Likewise.
2124 (md_parse_option <c,arg>): Likewise.
2125 (md_show_usage <stream>): Likewise.
2126 (md_convert_frag <headers, seg>): Likewise.
2127 (tc_coff_symbol_emit_hook <ignore>): Likewise.
2128 (md_atof): Remove declaration of atof_ieee.
2129 (tc_aout_fix_to_chars): Remove unused function.
2130 (parse_reg): Prototype.
2131 (parse_exp): Prototype.
2132 (skip_colonthing): Prototype. Use &&, not & in logical expressions.
2133 (parse_reglist): Prototype.
2134 (get_operand): Prototype.
2135 (get_operands): Prototype.
2136 (get_specific): Prototype. Make "this_index" signed.
2137 (check): Prototype, make static.
2139 (build_relaxable_instruction): Prototype, make static.
2140 (build_bytes): Prototype.
2141 (wordify_scb): Prototype.
2142 * config/tc-h8500.h (start_label): Declare.
2143 (tc_coff_sizemachdep): Declare.
2144 * config/tc-i370.c (i370_ebcdic <unused>): Add ATTRIBUTE_UNUSED.
2145 (i370_rmode <unused>): Likewise.
2146 (i370_csect <unused>): Likewise.
2147 (i370_dc <unused>): Likewise.
2148 (i370_ds <unused>): Likewise.
2149 (i370_elf_lcomm <unused>): Likewise.
2150 (i370_ltorg <ignore>): Likewise.
2151 (i370_using <ignore>): Likewise.
2152 (i370_drop <ignore>): Likewise.
2153 (i370_byte <ignore>): Likewise.
2154 (i370_tc <ignore>): Likewise.
2155 (md_estimate_size_before_relax <fragp, seg>): Likewise.
2156 (md_convert_frag <all args>): Likewise.
2157 (md_undefined_symbol <name>): Likewise.
2158 (md_pcrel_from_section <sec>): Likewise.
2159 (tc_gen_reloc <seg>): Likewise.
2160 (i370_section_letter): #if 0 unused functions.
2161 (i370_section_word, i370_section_type, i370_section_flags): Likewise.
2162 (symbol_locate): Prototype.
2163 * config/tc-i860.c (md_atof): Remove declaration of atof_ieee.
2164 (md_number_to_disp, md_number_to_field): Remove.
2165 (md_apply_fix3): Correct format string and cast "fup".
2166 * config/tc-i960.c (md_convert_frag): Add ATTRIBUTE_UNUSED to args.
2167 (s_endian <ignore>): Likewise.
2168 (md_undefined_symbol <name>): Likewise.
2169 (tc_crawl_symbol_chain <headers>): Likewise.
2170 (tc_set_bal_of_call): Likewise.
2171 (tc_coff_symbol_emit_hook <symbolP>): Likewise.
2172 (i960_handle_align <fragp>): Likewise.
2173 (i960_validate_fix <this_segment_type>): Likewise
2174 (tc_gen_reloc <section>): Likewise.
2175 (tc_coff_symbol_emit_hook): Only define for OBJ_COFF.
2176 (struct memS, struct regop): Forward declare.
2177 (brcnt_emit, brlab_next, cobr_fmt, ctrl_fmt, emit, get_args,
2178 get_cdisp, get_ispec, get_regnum, i_scan, mem_fmt, mema_to_memb,
2179 parse_expr, parse_ldconst, parse_memop, parse_po, parse_regop,
2180 reg_fmt, relax_cobr, s_leafproc, s_sysproc, shift_ok, syntax,
2181 targ_has_sfr, targ_has_iclass, tc_bfd_fix2rtype): Prototype.
2182 (md_chars_to_number, md_number_to_imm): Make static, prototype.
2183 (md_number_to_field): Likewise.
2184 (md_number_to_disp): Remove unused function.
2185 (md_atof): Remove declaration of atof_ieee.
2186 (md_apply_fix3): Correct md_number_to_imm call.
2187 * config/tc-ip2k.c (md_assemble): Warning fix.
2188 * config/tc-m32r.c (md_parse_option <arg>): Add ATTRIBUTE_UNUSED.
2189 (fill_insn <ignore>): Likewise.
2190 (debug_sym <ignore>): Likewise.
2191 (md_undefined_symbol <name>): Likewise.
2192 (m32r_scomm <ignore>): Likewise.
2193 (md_convert_frag <abfd>): Likewise.
2194 (md_cgen_lookup_reloc <insn>): Likewise.
2195 (m32r_record_hi16 <seg>): Likewise.
2196 (md_estimate_size_before_relax): #if 0 old_fr_fix.
2197 (allow_m32rx): Prototype.
2198 (first_writes_to_seconds_operands): Prototype.
2199 (writes_to_pc): Prototype.
2200 (can_make_parallel): Prototype.
2201 (make_parallel): Prototype.
2202 (target_make_parallel): Prototype.
2203 (assemble_two_insns): Prototype.
2204 (m32r_record_hi16): Prototype.
2205 (md_atof): Remove declaration of atof_ieee.
2206 * config/tc-m32r.h (m32r_fix_adjustable): Declare.
2207 (m32r_force_relocation): Prototype.
2208 (m32r_elf_section_change_hook): Prototype.
2209 * config/tc-m68k.c (tc_gen_reloc <section>): Add ATTRIBUTE_UNUSED.
2210 (md_show_usage): Fix signed/unsigned warning.
2211 * config/tc-m88k.c (get_reg): Make reg_prefix param unsigned.
2212 (calcop): Ditto for reg_prefix var.
2213 (hexval): Add cast to fix signed/unsigned warning.
2214 (md_number_to_disp): Delete unused function.
2215 (md_number_to_field): Likewise.
2216 (float_cons, cons, s_globl, s_space, s_set, s_lcomm): Remove decl.
2217 (match_name): Prototype.
2218 (get_bf2): Prototype.
2219 (get_bf_offset_expression): Prototype.
2220 * config/tc-mcore.c (mcore_s_literals <ignore>): Add ATTRIBUTE_UNUSED.
2221 (md_undefined_symbol <ignore>): Likewise.
2222 (md_create_short_jump <all args>): Likewise.
2223 (md_create_long_jump <all args>): Likewise.
2224 (md_convert_frag <abfd, sec>): Likewise.
2225 (md_apply_fix3 <segment>): Likewise.
2226 (md_section_align <segment>): Likewise.
2227 (md_pcrel_from_section <sec>): Likewise.
2228 (tc_gen_reloc <section>): Likewise.
2229 (reg_m, reg_n, immediate): Delete unused vars.
2230 (dump_literals): Fix signed/unsigned warning.
2231 (enter_literal): Likewise.
2232 (parse_imm): Likewise. Also fix format string.
2233 (parse_mem): Remove unused var.
2234 (md_assemble <LS>): Abort on unexpected inst.
2235 (md_atof): Remove declaration of atof_ieee.
2236 (md_parse_option): Remove unused vars.
2237 (md_apply_fix3): Fix format strings, cast args.
2238 (tc_gen_reloc): Delete unused var.
2239 * config/tc-mcore.h (tc_coff_sizemachdep): Declare.
2240 * config/tc-mn10200.c (md_parse_option <c, arg>): Add ATTRIBUTE_UNUSED.
2241 (md_undefined_symbol <name>): Likewise.
2242 (md_convert_frag <abfd>): Likewise.
2243 (tc_gen_reloc <seg>): Likewise.
2244 (check_operand <insn>): Likewise.
2245 (md_convert_frag): Fix format strings.
2246 (tc_gen_reloc): Delete fx_addsy - fx_subsy code.
2247 * config/tc-openrisc.c (ignore_pseudo): Prototype.
2248 (md_atof): Remove declaration of atof_ieee.
2249 * config/tc-or32.c (parse_operand): Prototype non-BFD too.
2250 (md_apply_fix3): Fix bogus >>='s.
2251 (md_undefined_symbol): Delete unused var.
2252 * config/tc-pj.c (little, big, parse_exp_save_ilp): Prototype.
2253 (c_to_r, ipush_code, fake_opcode, alias): Likewise.
2254 (fake_opcode): Adjust for pj_opc_int_t change.
2255 (md_begin): Likewise.
2256 (md_assemble): Likewise.
2257 (ipush_code): Correct parse_exp_save_ilp call. Test pending_reloc
2258 instead of non-existent third arg of parse_exp_save_ilp.
2259 (md_parse_option): Correct "little" and "big" calls.
2260 * config/tc-sparc.c (s_register): Only declare #ifdef OBJ_ELF.
2261 (md_apply_fix3 <segment>): Add ATTRIBUTE_UNUSED.
2262 (tc_gen_reloc <section>): Likewise.
2263 * config/tc-tic30.c: #include stdarg.h or varargs.h.
2264 (debug): Rewrite using VA_* macros.
2265 (md_estimate_size_before_relax): Add ATTRIBUTE_UNUSED to args.
2266 (md_convert_frag): Likewise.
2267 (md_parse_option): Likewise.
2268 (md_show_usage): Likewise.
2269 (md_undefined_symbol): Likewise.
2270 (tc_gen_reloc): Likewise.
2271 (md_operand): Likewise.
2272 (tc_aout_pre_write_hook): Delete.
2273 (struct tic30_insn): Make "operands" unsigned.
2274 (struct tic30_par_insn): Likewise.
2275 (md_assemble): Likewise for "count", "i" and "numops".
2276 (tic30_parallel_insn): Likewise for vars here.
2277 (tic30_operand): Likewise. Remove useless unsigned >= 0 comparison.
2278 * config/tc-tic30.h (tc_aout_pre_write_hook): Define as empty.
2279 * config/tc-tic80.c (obj_coff_section): Delete declaration.
2280 (md_estimate_size_before_relax): Add ATTRIBUTE_UNUSED on args.
2281 (md_undefined_symbol): Likewise.
2282 (md_parse_option): Likewise.
2283 (md_convert_frag): Likewise.
2284 (tc_coff_symbol_emit_hook): Likewise.
2285 (md_atof): Remove declaration of atof_ieee.
2286 (const_overflow): Warning fixes, tidy.
2287 (get_operands): Delete unused vars.
2288 (internal_error_a): Adjust format string to expect a long for arg.
2289 (find_opcode): Warning fixes, simplify.
2290 (build_insn): Cast internal_error_a arg.
2291 (md_begin): Likewise.
2292 (md_apply_fix3): Likewise.
2293 (md_assemble): Delete unused var.
2294 * config/tc-tic80.h (tc_coff_fix2rtype): Prototype.
2295 * config/tc-z8k.c (cons, obj_coff_section): Delete declarations.
2296 (whatreg, parse_reg, parse_exp): Make static, prototype.
2297 (checkfor, regword, regaddr, get_ctrl_operand): Prototype.
2298 (get_flags_operand, get_interrupt_operand, get_cc_operand): Likewise.
2299 (get_operand, get_operands, get_specific, newfix): Likewise.
2300 (apply_fix, build_bytes): Likewise.
2301 (md_atof): Remove declaration of atof_ieee.
2302 (tc_aout_fix_to_chars): Delete.
2303 (md_begin): Constify "opcode". Don't try to init opcode->idx.
2305 (md_parse_option): Fix s_segm and s_unseg calls.
2309 * tc-mips.c (RELAX_BRANCH_ENCODE): Remove reloc_s2 argument.
2311 (RELAX_BRANCH_RELOC_S2): Delete.
2312 (append_insn): Use only BFD_RELOC_16_PCREL_S2 for branches.
2313 Do not handle BFD_RELOC_16_PCREL.
2314 (macro_build, mips_ip): Likewise.
2315 (md_pcrel_from): Return 4 for undefined symbols regardless of
2317 (md_apply_fix3): Use only BFD_RELOC_16_PCREL_S2 for branches.
2318 Don't dereference howto if no such relocation is available.
2319 Do not apply hack for in-place zero addend in NEWABI.
2320 (md_convert_frag): Use only BFD_RELOC_16_PCREL_S2 for branches.
2325 * config/tc-mn10300.h (TC_VALIDATE_FIX_SUB): Define.
2326 (TC_LINKRELAX_FIXUP): Add comments.
2330 * config/tc-mn10300.c (set_arch_mach): Change argument type to
2332 (r_register_name, xr_register_name): Add prototype declarations.
2336 * config/tc-ia64.c (md_apply_fix3): Remove the PCREL hack
2337 copied from tc-i386.c.
2341 Fix Bug savannah/1825:
2342 * config/tc-m68hc11.c (STATE_INDEXED_PCREL): New relax code.
2343 (md_relax_table): Define specific relax for PC-rel offsets.
2344 (build_indexed_byte): Use a STATE_INDEXED_PCREL relax code.
2345 (m68hc11_relax_frag): Handle the new relax code.
2346 (md_convert_frag): Likewise.
2347 (md_estimate_size_before_relax): Likewise.
2351 * subsegs.c (section_symbol): Use the symbol, not the section, name.
2355 * config/ia64.c (enum reloc_func): Add FUNC_LT_RELATIVE_X.
2356 (ia64_gen_real_reloc_type): Handle it.
2357 (pseudo_func): Add @ltoffx.
2358 (md_begin): Build .<ltoffx>.
2359 (ia64_force_relocation): True for LTOFF22X and LDXMOV.
2363 * config/tc-arm.c (arm_force_relocation): Move out of #if OBJ_ELF.
2364 Move OBJ_COFF TC_FORCE_RELOCATION code here so that COFF handles
2365 ARM_IMMEDIATE and ARM_ADRL_IMMEDIATE relocs as for ELF.
2366 * config/tc-arm.h (TC_FORCE_RELOCATION): Define for both ELF and
2367 COFF to call arm_force_relocation.
2371 * config/tc-ia64.c (pseudo_func): Add "@pause" constant for "hint"
2373 (emit_one_bundle): Handle "hint" instruction.
2374 (operand_match): Match IA64_OPND_AR_CSD.
2378 * dwarf2dbg.c (scale_addr_delta): Correct parameter. Move prototype
2383 * config/tc-ppc.c (ppc_cleanup): Do not set SEC_MERGE flag on
2384 .PPC.EMB.apuinfo sections.
2388 * config/tc-ia64.c (operand_match): Add IA64_OPND_LDXMOV case.
2392 * config/tc-w65.c (s_longa): Prototype. Make static, specify int arg.
2393 (cons, s_align_bytes): Delete declaration.
2395 (md_begin): Constify "struct opinfo *" var. Don't try to make "name"
2397 (dot): Delete unused function.
2398 (w65_expression): Remove unused arg.
2399 (parse_exp): Prototype. Remove unused arg. Adjust w65_expression
2401 (get_operands): Prototype. Constify "struct opinfo *" arg. Fix
2403 (get_specific): Prototype. Constify "struct opinfo *" arg and return
2405 (check): Remove unused function.
2406 (build_Mytes): Prototype. Constify "struct opinfo *" arg. Abort
2407 on unhandled switch case.
2408 (md_assemble): Remove unused op_start, op_end, nlen, p vars. Constify
2410 (tc_crawl_symbol_chain): Delete unused function.
2411 (tc_headers_hook): Likewise.
2412 (tc_Nout_fix_to_chars): Likewise.
2413 (md_undefined_symbol): Add ATTRIBUTE_UNUSED.
2414 (md_parse_option): Likewise.
2415 (md_convert_frag): Likewise.
2416 (tc_coff_symbol_emit_hook): Likewise.
2417 (md_show_usage): Likewise.
2418 * config/tc-w65.h (tc_coff_sizemachdep): Declare.
2419 (TC_PARSE_CONS_EXPRESSION): w65_expression takes one arg.
2420 (w65_expression): Declare.
2422 * po/POTFILES.in: Regenerate.
2424 * config/tc-arm.c (arm_force_relocation): Return 0 for ARM_IMMEDIATE
2425 and ARM_ADRL_IMMEDIATE.
2429 * gasp.c: Delete. It has been deprecated.
2430 * NEWS: Mention that gasp has been removed.
2431 * Makefile.am: Remove references to gasp.
2432 * makefile.vms: Likewise.
2433 * mpw-make.sed: Likewise.
2434 * Makeile.in: Regenerate.
2435 * doc/gasp.texi: Delete.
2436 * doc/Makefile.am: Remove references to gasp.texi.
2437 * doc/Makefile.in: Regenerate.
2438 * macro.c: Delete references to gasp.
2439 Remove use of comment_char function parameter as it is no longer
2441 * macro.h: Update prototypes to remove comment_char parameter.
2443 * read.c (read_a_source_file, s_irp): Remove comment_char
2444 parameter from invocation of functions in macro.c
2448 * read.c (emit_expr) [!WORKING_DOT_WORD]: Initialize x->use_jump.
2452 * config/tc-m68hc11.c (md_begin): Fix qsort warning.
2453 (tc_gen_reloc): Mark section param as not used.
2457 Fix Bug savannah/1825:
2458 * config/tc-m68hc11.h (md_relax_frag): Define to support relaxations
2459 that are not pc-relative.
2460 (m68hc11_relax_frag): Declare.
2462 * config/tc-m68hc11.c (build_indexed_byte): Use a frag_var to handle
2463 the offsetable indexed addressing modes (n,r).
2464 (build_insn): Cleanup some locals.
2465 (m68hc11_relax_frag): New function imported from tc-cris.c to handle
2466 relaxation of difference between two symbols of same section.
2467 (md_convert_frag): For INDEXED_OFFSET relaxs, use the displacement
2468 only when this is a PC-relative operand and the offset is not absolute.
2469 (md_estimate_size_before_relax): Convert the INDEXED_OFFSET,UNDEF frag
2470 to INDEXED_OFFSET,STATE_BITS5 when the symbol is absolute; this will
2471 be handled by m68hc11_relax_frag.
2475 * config/tc-m68hc11.c (elf_flags): Set default ABI to gcc default
2476 (32-bit int, 64-bit double).
2477 (md_longopts): New options -mshort, -mlong, -mshort-double and
2478 -mlong-double to control the ABI.
2479 (md_show_usage): Update.
2480 (md_parse_option): Handle new options.
2481 * doc/as.texinfo (Overview): Document new options for HC11/HC12.
2485 * symbols.c (report_op_error): New function, broken out of ...
2486 (resolve_symbol_value): ... here. Use for both monadic and dyadic
2491 * config/tc-sh.c (md_apply_fix3): Take account of fx_offset
2492 for BFD_RELOC_32_PLT_PCREL.
2496 * dwarf2dbg.c, write.c, config/obj-aout.c, config/obj-coff.c,
2497 config/obj-ecoff.c, config/obj-elf.c, config/obj-som.c,
2498 config/tc-arm.c, config/tc-arm.h, config/tc-avr.c, config/tc-cris.c,
2499 config/tc-d10v.c, config/tc-d10v.h, config/tc-d30v.c, config/tc-d30v.h,
2500 config/tc-dlx.c, config/tc-dlx.h, config/tc-fr30.c, config/tc-fr30.h,
2501 config/tc-frv.c, config/tc-frv.h, config/tc-hppa.c, config/tc-i370.c,
2502 config/tc-i386.c, config/tc-i386.h, config/tc-m32r.c,
2503 config/tc-m68hc11.c, config/tc-mcore.c, config/tc-mcore.h,
2504 config/tc-mips.c, config/tc-mips.h, config/tc-mn10200.c,
2505 config/tc-mn10300.c, config/tc-mn10300.h, config/tc-openrisc.c,
2506 config/tc-openrisc.h, config/tc-ppc.c, config/tc-s390.c,
2507 config/tc-sh.c, config/tc-sh.h, config/tc-sh64.c, config/tc-tic54x.c,
2508 config/tc-v850.c, config/tc-v850.h, config/tc-xstormy16.c,
2509 config/tc-xstormy16.h: Replace boolean with bfd_boolean, true with
2510 TRUE, false with FALSE. Simplify comparisons of bfd_boolean vars with
2511 TRUE/FALSE. Formatting.
2515 * config/tc-arm.c (do_t_bkpt): If no operand is specified,
2516 assume a value of zero.
2517 (do_bkpt): Likewise.
2521 * config/tc-sh.c (md_apply_fix3): Call S_SET_THREAD_LOCAL
2522 for TLS relocations.
2526 * config/tc-ia64.c (md_apply_fix3): Add default case.
2530 * symbols.c (S_SET_THREAD_LOCAL): New function.
2531 * symbols.h (S_SET_THREAD_LOCAL): New prototype.
2532 * config/tc-i386.c (md_apply_fix3): Call S_SET_THREAD_LOCAL
2533 for TLS relocations.
2534 * config/tc-ia64.c (md_apply_fix3): Likewise.
2535 * config/tc-alpha.c (md_apply_fix3): Likewise.
2539 * write.c (subsegs_finish): For SEC_MERGE sections pad last fragment
2544 * dwarf2dbg.c (scale_addr_delta): New.
2545 (size_inc_line_addr): Use it.
2546 (emit_inc_line_addr): Use it.
2550 * config/tc-cris.c (cris_relax_frag): Fix typo in comment.
2551 (md_assemble): Don't pass on branches to constants as relaxable.
2556 * config/tc-tic4x.c (c4x_operands_match): Bugfix in direct mode
2560 * config/tc-hppa.h (tc_frob_symbol): Frob undefined unused symbols
2561 only if they have default visibility.
2565 * config/tc-alpha.c (alpha_fix_adjustable): Remove redundant
2570 * config/tc-mcore.c (mcore_pool_count): New function.
2571 (mcore_cons, mcore_float_cons, mcore_stringer, mcore_fill): Use it.
2575 * config/tc-mcore.c (md_begin): Use a const iterator. Don't
2576 coalesce the name fields to use the same pointer.
2578 * config/tc-sh.c (md_begin): Use a const iterator. Don't coalesce
2579 the name fields to use the same pointer.
2580 (get_specific): Check for opcodes with the same name using strcmp
2581 as well as comparing the pointer.
2585 * write.c (adjust_reloc_syms): Don't reduce SEC_MERGE fixups with
2590 * config/obj-elf.c (obj_elf_visibility): Overwrite only the
2591 visibility portion of st_other.
2595 * config/tc-h8300.c (struct h8_instruction): New type, used to
2596 wrap h8_opcodes with length, noperands, idx, and size fields
2597 (computed at run-time).
2598 (h8_instructions): New variable.
2599 (md_begin): Allocate the storage for h8_instructions. Fill
2600 h8_instructions with pointers to the appropriate opcode and the
2601 correct value for the additional fields.
2602 (clever_message): Update to use h8_instructions instead of
2604 (build_bytes): Ditto.
2605 (get_specific): Ditto.
2606 (md_assemble): Ditto.
2610 * config/tc-s390.c (tc_s390_fix_adjustable): Re-add patch to prevent
2611 adjustments to symbols in merge sections.
2615 * config/tc-alpha.c (s_alpha_prologue): as_bad when sym is NULL.
2619 * dwarf2dbg.h (dwarf2_format): New enum.
2620 * dwarf2dbg.c (DWARF2_FORMAT): Provide default definition.
2621 (out_debug_line, out_debug_info): Add code for handling 64-bit
2623 * config/tc-mips.h (mips_dwarf2_format): Declare.
2624 * config/tc-mips.c (mips_dwarf2_format): New function.
2625 * doc/internals.texi (DWARF2_FORMAT): Document.
2629 * config/tc-mips.c (s_change_section): Make sure input buffer
2630 is not accessed past the end. Don't hand
2631 obj_elf_change_section a pointer into the input buffer.
2635 * config/tc-mips.c (tc_gen_reloc): Fix typo in handling of
2640 * config/tc-tic4x.c: Fixed proper commandline
2641 parameters. Added support for new opcode-list format. General
2642 error message fixups.
2643 (c4x_inst_add): Reject insn not for our CPU
2644 (md_begin): Added matrix for setting the proper opcode-level &
2645 device-flags according to cpu type and revision. Rewrite the
2647 (c4x_operand_parse): Fix opcode bug
2648 (c4x_operands_match): New function argument. Added dry-run
2649 mechanism, that is optional error generation. Added constraint 'i'
2651 (c4x_insn_check): Added new function for post-verification of the
2653 (md_assemble): Check all opcodes before croaking because of an
2654 argument mismatch. Need this to be able to fully support
2655 ortogonally arguments.
2656 (md_parse_options): Revised commandprompt swicthes and added new
2658 (md_show_usage): Complete rewrite of printout.
2662 * config/tc-tic4x.c: Remove c4x_pseudo_ignore function.
2663 (c4x_operands_match): Added check for 8-bits LDF insn. Give
2664 warning when using constant direct bigger than 2^16. Add the new
2669 * configure.in: Use .gdbinit under Cygwin.
2670 * configure: Regenerate.
2674 * config/tc-i386.h (EXTERN_FORCE_RELOC): Define only if
2679 * config/tc-tic4x.c: Declare as many functions as possible as
2680 static. Maintenance on the general indenting. Removed unnecessary
2681 pseudo-ops and added new ones. Removed obsoleted c4x_pseudo_ignore
2682 function. Add support for new DSP, TMS320VC33. Fix bug for
2683 converting flonum constants.
2684 (c4x_do_align): Add proper align handling. Setup align to insert
2686 (c4x_gen_to_words): Support for extended TI type floats.
2687 (md_atof): Proper dumping of multiple-word littlenums.
2688 (c4x_atof): Added support for extended TI type floats.
2689 (c4x_stringer): Added new function to handle compact strings.
2690 (c4x_emit_char): Added new function argument to handle custom
2691 length inserts, like single-byte strings.
2692 * config/tc-tic4x.h: Add proper align handling with NOP's.
2693 * Makefile.am: Added tic4x dependecy
2694 * Makefile.in: Regenerate
2698 * macro.c (get_any_string): Correct logic for not going beyond end
2703 * config/tc-mmix.c (get_putget_operands): Mark both possible
2704 operands as invalid at beginning.
2706 * config/tc-mmix.c (md_convert_frag) <case STATE_GREG_DEF>:
2707 Initialize target of fixup to zero.
2711 * config/tc-mips.c (macro_build_lui): _gp_disp is not special on
2712 NEWABI, but we should still emit HI16_S for non-PIC n32.
2716 * config/tc-alpha.c (alpha_validate_fix): Move code ...
2717 (alpha_fix_adjustable): ... here.
2718 * config/tc-alpha.h (TC_VALIDATE_FIX): Remove.
2722 * config/te-pe.h (LEX_AT): Accept at-sign (@) as first character
2727 * config/tc-mips.c (support_64bit_objects): Check *l before it
2732 * config/obj-coff.c (obj-coff-section): Set SEC_DATA and
2733 SEC_LOAD flags for sections marked as 's'.
2737 * write.c (TC_FORCE_RELOCATION_SUB_ABS): Default to zero.
2738 (TC_FORCE_RELOCATION_SUB_LOCAL): Likewise when DIFF_EXPR_OK.
2739 * doc/internals.texi (TC_FORCE_RELOCATION_SUB_ABS): Document changed
2742 * dep-in.sed: Fix typo.
2746 * configure.in: Update ARM CPU patterns.
2747 * configure: Regenerated.
2751 * itbl-lex.l: Use #include <> for generated headers.
2752 * itbl-ops.c: Likewise.
2756 * doc/gasp.texi: Fix typo in deprecation note.
2760 * config/tc-arm.c (tc_gen_reloc): Allow an absolute reference to
2761 _GLOBAL_TABLE_OFFSET_ to be converted into a GOT reloc.
2765 * config/tc-mmix.h (mmix_frob_file_before_adjust): Don't declare.
2766 (tc_frob_file_before_adjust): Don't define.
2767 * config/tc-mmix.c (mmix_frob_local_reloc): Remove unused
2769 (mmix_frob_file_before_adjust): Remove ineffective function.
2773 * config/tc-cris.c (SIMPLE_EXPR): New macro.
2774 (cris_relax_frag): New function.
2775 (md_estimate_size_before_relax) <case ENCODE_RELAX
2776 (STATE_BASE_PLUS_DISP_PREFIX, STATE_UNDF)>: Pass on unresolved
2777 expressions that will become absolute expressions to relaxation.
2778 (md_convert_frag) <case ENCODE_RELAX (STATE_BASE_PLUS_DISP_PREFIX,
2779 STATE_WORD)>: Expect only absolute expressions. Use the symbol
2780 value, not distance to symbol.
2781 <case ENCODE_RELAX (STATE_BASE_PLUS_DISP_PREFIX, STATE_BYTE)>:
2782 Ditto. Correct placement of fixup.
2783 (md_assemble): Use SIMPLE_EXPR when dissecting expressions.
2785 * config/tc-cris.h (cris_relax_frag): Declare.
2786 (md_relax_frag): Define.
2790 * config/obj-elf.c (special_sections): Use correct types for init
2792 (obj_elf_change_section): Don't mess with init array section type.
2796 * config/tc-mips.c (mips_need_elf_addend_fixup): Return true
2797 for relocs against symbols in a merged section.
2801 * config/tc-mips.c (md_begin): Add $fcc registers to the symbol
2802 table as register names.
2806 * config/tc-s390.c (md_parse_option): Set s390_arch_size to 32
2811 * expr.c (operand): Add support for n.nn flonums.
2815 * itbl-parse.y (entry): Provide empty action.
2819 * configure.in (BFDVER_H): Set and subst.
2820 * dep-in.sed: Replace bfdver.h with $(BFDVER_H).
2821 * Makefile.am: Run "make dep-am".
2823 * configure: Regenerate.
2824 * Makefile.in: Regenerate.
2825 * doc/Makefile.in: Regenerate.
2829 * config/tc-i386.h (EXTERN_FORCE_RELOC): Define.
2830 (MD_APPLY_SYM_VALUE): Define for PE too.
2834 * Makefile.am: Run "make dep-am".
2835 (CPU_OBJ_VALID): sh64 coff is invalid.
2836 * as.c: #include "bfdver.h".
2837 * Makefile.in: Regenerate.
2838 * config.in: Regenerate.
2842 * config/tc-v850.c (CHECK_): Remove token pasting operator.
2846 * configure.in (mips64vr-elf, mips64vrel-elf): New config.
2847 * configure: Regenerate.
2852 * config/tc-mips.c (s_gpdword): New function.
2853 (mips_pseudo_table): Add .gpdword.
2854 (mips_need_elf_addend_fixup): never for NEWABI.
2855 (md_apply_fix3): Don't mark BFD_RELOC64 after GPREL16 or
2857 (s_cpadd): Generate .cpadd on NEWABI.
2861 * config/tc-ppc.c (ppc_cleanup): Make 'i' unsigned int.
2865 * config/tc-mips.h (mips_relax_frag): Take segment as argument.
2866 (md_relax_frag): Adjust macro.
2867 * config/tc-mips.c (mips_relax_branch): New variable.
2868 (RELAX_BRANCH_ENCODE, RELAX_BRANCH_P, RELAX_BRANCH_LIKELY,
2869 RELAX_BRANCH_LINK, RELAX_BRANCH_TOOBAR): New.
2870 (RELAX_MIPS16_P): Adjust.
2871 (append_insn): Emit branch to non-constant in a frag_var if
2872 branch-relaxation is desirable and possible.
2873 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH): New options.
2874 (OPTION_ELF_BASE): Adjust.
2875 (md_parse_option): Handle new options.
2876 (md_apply_fix3): Update comment on EMBEDDED_PIC conditional
2878 (relaxed_branch_length): New function.
2879 (md_estimate_size_before_relax): Handle branch frags.
2880 (mips_relax_frag): Likewise.
2881 (md_convert_frag): Handle branch frags. Warn if branch is
2886 * config/tc-sh.c (sh_force_relocation): Make sure TLS relocs get
2888 (md_apply_fix3): Add TLS relocs.
2889 (sh_parse_name): Support @TLSGD, @TLSLDM, @GOTTPOFF, @TPOFF and
2895 * config/tc-i386.c (output_jump): Set fx_signed for loop/jcxz.
2896 (md_estimate_size_before_relax): Likewise for 8 bit branches.
2900 * config/tc-sh.c (assemble_ppi): Initialize reg_x / reg_y / reg_n
2906 * config/tc-sh64.c (sh64_target_format): Add support for sh64
2911 * contig/tc-sh.c (sh_local_pcrel): New.
2912 (sh_force_relocation): Use sh_local_pcrel.
2913 (md_pcrel_from_section): Check the relocation type whether it
2914 should be resolved locally. Use S_FORCE_RELOC.
2918 * config/tc-mips.h (TC_FORCE_RELOCATION_SUB_SAME): Define.
2919 (TC_FORCE_RELOCATION): Tidy arg.
2929 * doc/c-mips.texi: Add entries for -march=vr4120,vr4130,vr4181,
2930 vr5400 and vr5500. Add entry for -mfix-vr4122-bugs.
2931 * config/tc-mips.c (CPU_HAS_DROR, CPU_HAS_ROR): New macros.
2932 (hilo_interlocks): True for CPU_VR5500.
2933 (gpr_interlocks, cop_interlocks): True for CPU_VR5400 and CPU_VR5500.
2934 (mips_fix_vr4122_bugs): New.
2935 (append_insn): Work around 4122 errors if mips_fix_vr4122_bugs.
2936 (mips_emit_delays): Likewise.
2937 (macro2) [M_DROLI]: Use dror or dror32 if CPU_HAS_DROR.
2938 [M_ROLI]: Likewise ror if CPU_HAS_ROR.
2939 (validate_mips_insn, mips_ip): Handle '[', ']', 'e' and '%'.
2940 (OPTION_FIX_VR4122, OPTION_NO_FIX_VR4122): New options.
2941 (md_longopts): Add -mfix-vr4122-bugs and -no-mfix-vr4122-bugs.
2942 (OPTION_ELF_BASE): Bump.
2943 (md_parse_option): Handle the new options.
2944 (mips_cpu_info_table): Add entries for vr4120, vr4130, vr4181,
2949 * config/tc-mips.c (md_apply_fix3): Subtract the symbol value
2950 twice if howto->pcrel_offset is true.
2955 * config/tc-vax.c (md_estimate_size_before_relax): Only try to
2956 convert undefined references to GOT32/PLT32 if PIC code is
2957 requested. Fix comment.
2961 * config/tc-sh.c (sh_force_relocation): Return 0 for
2962 some PC relative relocations when not relaxing.
2966 * config/tc-i386.c (tc_i386_fix_adjustable): Add x86-64 TLS relocs.
2967 Define them if not BFD_ASSEMBLER.
2968 (lex_got): Handle @tlsgd, @dtpoff and @tpoff in 64-bit mode, add
2970 (md_apply_fix3): No addend for BFD_RELOC_X86_64_TLSGD,
2971 BFD_RELOC_X86_64_TLSLD and BFD_RELOC_X86_64_GOTTPOFF.
2972 (tc_gen_reloc): Handle x86-64 TLS relocs.
2976 * config/tc-avr.c (md_apply_fix3): Reinstate code handling pcrel
2977 fixups to current or absolute section.
2981 * config/tc-v850.c (v850_offset): Use frag_var instead of frag_now_fix
2986 * config/tc-mips.c (CPU_HAS_MIPS16): Add mips-lsi-elf as MIPS16
2987 capable configuration.
2988 (macro_build): Check for MIPS16 capability, not for actual MIPS16 code
2990 (mips_ip): Likewise.
2994 * config/tc-mips.c (append_insn): Fix jump overflow check.
2998 * config/tc-i386.c (process_operands): Warn about "lea" segment
3003 * write.c: Delete set_segment_vma and prototype. Update all callers.
3007 * config/tc-i386.c (md_apply_fix3): Replace S_IS_EXTERNAL,
3008 S_IS_WEAK etc. with S_FORCE_RELOC call. Correct comment.
3009 Rename "fseg" to "sym_seg".
3010 * Makefile.am: Run "make dep-am".
3011 * Makefile.in: Regenerate.
3012 * doc/Makefile.in: Regenerate.
3016 * symbols.c (colon): Do not allow symbols to be created in the
3017 absolute section if WORKING_DOT_WORD is not defined and
3018 new_broken_words would require a new frag to be created.
3022 * expr.c (expr): Simplify foo-foo here.
3023 (clean_up_expression): Remove O_subtract code.
3025 * write.h (struct fix): Add fx_dot_value.
3026 (dot_value): Declare.
3027 * write.c (dot_value): New var.
3028 (fix_new_internal): Save dot_value as fx_dot_value.
3029 (fixup_segment): Adjust fx_offset using fx_dot_value.
3030 * expr.c (expr): Update dot_value.
3034 * config/tc-i386.c (tc_i386_fix_adjustable): Handle
3035 BFD_RELOC_386_TLS_IE and BFD_RELOC_386_TLS_GOTIE.
3036 (BFD_RELOC_386_TLS_IE, BFD_RELOC_386_TLS_GOTIE): Define to 0
3038 (lex_got): Handle @GOTNTPOFF and @INDNTPOFF.
3039 (md_apply_fix3, tc_gen_reloc): Handle BFD_RELOC_386_TLS_IE and
3040 BFD_RELOC_386_TLS_GOTIE.
3044 * config/tc-alpha.c (md_pcrel_from): Only adjust special for
3046 (alpha_force_relocation): Don't special-case branch type relocs.
3050 * config/tc-m68k.c (select_control_regs): Handle situation where
3051 architecture has not yet been selected.
3055 * config/tc-mips.c (IS_SEXT_32BIT_NUM): Move closer to top of file.
3056 (IS_SEXT_16BIT_NUM): New macro.
3057 (macro_build_ldst_constoffset): New function, to build a set of
3058 instructions to do a load or store from a constant offset relative
3059 to a given register.
3060 (macro, s_cprestore): Use macro_build_ldst_constoffset to implement
3061 .cprestore pseudo-op.
3065 * config/tc-mips.c (md_apply_fix3): Just return for BFD_RELOC_8.
3069 * config/tc-mips.c (s_change_section): Fix parsing. Code cleanup.
3073 * tc-mips.c (load_address): Use BFD_RELOC_MIPS_GOT_DISP for newabi.
3074 (macro): Likewise for la. Likewise for ld.
3075 (mips_after_parse_args): Make -xgot optional, not the default.
3076 (md_apply_fix3): Allow composite relocation to set up gp.
3077 (tc_gen_reloc): Allow relaxing for newabi.
3078 Relax R_MIPS_CALL16 to R_MIPS_GOT_PAGE/R_MIPS_GOT_OFST if local.
3079 Relax R_MIPS_GOT16/R_MIPS_LO16 to R_MIPS_GOT_DISP if local.
3083 * config/tc-arm.c (md_apply_fix3): Note that an implemented
3084 BFD_RELOC_ARM_IMMEDIATE has been done.
3085 (tc_gen_reloc): Do not issue reloc number of unimplemented
3086 BFD_RELOC_ARM_IMMEDIATE and BFD_RELOC_ARM_OFFSET_IMM relocs -
3087 their name is already in the error message - plus remove them
3088 from the default case.
3090 * config/tc-arm.c (do_ldmstm): Warn about unpredictable
3091 behavior of instructions.
3095 * config/tc-tic4x.c: Add function declarations and ATTRIBUTE_UNUSED.
3096 Convert functions to K&R format.
3100 * config/tc-mips.c (pdr_seg): Define only for ELF.
3101 (s_change_section): Remove unused variable. Don't use for nonELF.
3105 * config/obj-elf.c (obj_elf_change_section): Move prototype to
3107 * config/obj-elf.h (obj_elf_change_section): Likewise.
3111 * dwarf2dbg.c (out_debug_abbrev): Add support for the DW_AT_name field.
3112 (out_debug_info): Likewise.
3116 * config/tc-i386.h (ELF_TARGET_FORMAT): New macro.
3117 (TARGET_FORMAT): Use ELF_TARGET_FORMAT instead of "elf32-i386".
3118 * config/tc-i386.c (i386_target_format): Likewise.
3119 * config/tc-alpha.h (ELF_TARGET_FORMAT): New macro.
3120 (TARGET_FORMAT): Use ELF_TARGET_FORMAT instead of "elf64-alpha".
3124 * config/tc-ppc.c (md_assemble): Do not count FAKE operands
3125 when deciding if any operands have been skipped.
3129 * NEWS: New TI port supports both C4x and C3x series of DSPs.
3131 * po/tr.po: Updated Turkish translation.
3135 * config/tc-i386.c (md_apply_fix3): Allow addend for
3136 BFD_RELOC_386_TLS_LDO_32, BFD_RELOC_386_TLS_LE and
3137 BFD_RELOC_386_TLS_LE_32.
3141 * config/tc-hppa.c (md_apply_fix3): Don't set fx_done for
3142 marker relocations such as ENTRY/EXIT.
3143 * config/tc-hppa.h (MD_APPLY_SYM_VALUE): Definition applies
3144 to both OBJ_ELF and OBJ_SOM.
3148 * doc/internals.texi (md_apply_fix3): Expand.
3149 (TC_VALIDATE_FIX, TC_FORCE_RELOCATION, TC_FORCE_RELOCATION_ABS,
3150 TC_FORCE_RELOCATION_LOCAL, TC_FORCE_RELOCATION_SUB_SAME,
3151 TC_FORCE_RELOCATION_SUB_ABS, TC_FORCE_RELOCATION_SUB_LOCAL,
3152 TC_VALIDATE_FIX_SUB, MD_APPLY_SYM_VALUE, S_FORCE_RELOC,
3153 EXTERN_FORCE_RELOC): Document.
3154 (TC_HANDLES_FX_DONE, obj_fix_adjustable): Remove.
3155 * as.h: Don't include struc-symbol.h for arc.
3157 * cgen.c (gas_cgen_md_apply_fix3): Remove *valP fudges and code to
3158 subtract absolute symbol.
3159 * obj.h (struct format_ops): Add frob_file_before_fix.
3160 * subsegs.c (section_symbol): Set BSF_SECTION_SYM flag.
3161 * symbols.c (S_FORCE_RELOC): New function.
3162 * symbols.h (S_FORCE_RELOC): Declare.
3163 * write.c (TC_FORCE_RELOCATION): Change default.
3164 (TC_FORCE_RELOCATION_ABS): Define.
3165 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3166 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3167 (TC_FORCE_RELOCATION_SECTION): Don't define.
3168 (TC_FORCE_RELOCATION_SUB_SAME): Define this instead.
3169 (TC_FORCE_RELOCATION_SUB_ABS): Define.
3170 (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
3171 (TC_VALIDATE_FIX_SUB): Define.
3172 (TC_FIX_ADJUSTABLE): Don't define.
3173 (MD_APPLY_SYM_VALUE): Define this instead.
3174 (abs_section_sym): New variable.
3175 (adjust_reloc_syms): Use S_FORCE_RELOC. Remove obj_fix_adjustable
3176 call. Don't symbol_mark_used_in_reloc here. Simplify link_once tests.
3177 Don't put the absolute section sym on fixups here.
3178 (fix_segment): New function.
3179 (write_relocs): Don't call fixup_segment from here.
3180 (write_object_file): Instead call tc_frob_file_before_fix,
3181 obj_frob_file_before_fix, and fix_segment prior to symbol table code.
3182 Don't output the absolute section symbol.
3183 (fixup_segment): Rewrite.
3184 * write.h (abs_section_sym): Declare.
3185 * config/obj-aout.c (obj_aout_frob_file_before_fix): Rename from
3187 (aout_format_ops): Adjust to suit.
3188 * config/obj-aout.h (obj_frob_file): Don't define.
3189 (obj_frob_file_before_fix): Define.
3190 (obj_aout_frob_file_before_fix): Rename from obj_aout_frob_file.
3191 (S_FORCE_RELOC): Define.
3192 * config/obj-bout.h (S_FORCE_RELOC): Define.
3193 * config/obj-coff.c (coff_format_ops): Init new field.
3194 * config/obj-coff.h: Formatting fixes.
3195 (obj_sec_sym_ok_for_reloc): Define.
3196 (S_FORCE_RELOC): Define.
3197 * config/obj-ecoff.c (ecoff_frob_file_before_fix): Split out ..
3198 (ecoff_frob_file): .. from here.
3199 (ecoff_format_ops): Add new function.
3200 * config/obj-ecoff.h (ecoff_frob_file_before_fix): Declare.
3201 (obj_frob_file_before_fix): Define.
3202 * config/obj-elf.c (elf_format_ops): Init new field.
3203 * config/obj-elf.h (obj_sec_sym_ok_for_reloc): Expand comment.
3204 * config/obj-ieee.h: Formatting fixes.
3205 (S_FORCE_RELOC): Define.
3206 * config/obj-multi.h (obj_frob_file_before_fix): Define.
3207 * config/obj-vms.h (S_FORCE_RELOC): Define.
3208 * config/tc-alpha.c (md_apply_fix3): Correct GPDISP comment.
3209 (alpha_force_relocation): Use S_FORCE_RELOC, and don't return 0
3210 for BFD_RELOC_32 and BFD_RELOC_64.
3211 (alpha_fix_adjustable): Remove extern and weak tests.
3212 (alpha_before_fix): Rename from alpha_adjust_symtab.
3213 (alpha_adjust_relocs): Rename from alpha_adjust_symtab_relocs.
3214 * config/tc-alpha.h (struct fix, struct alpha_reloc_tag): Declare.
3215 (TC_VALIDATE_FIX): Tweak param name.
3216 (TC_FORCE_RELOCATION, tc_fix_adjustable): Likewise.
3217 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3218 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3219 (MD_APPLY_SYM_VALUE): Define.
3220 (tc_adjust_symtab): Don't define.
3221 (alpha_adjust_symtab): Don't declare.
3222 (tc_frob_file_before_fix): Define.
3223 (alpha_before_fix): Declare.
3224 (TC_INIT_FIX_DATA): Tweak param names.
3225 * config/tc-arc.c: Include "struc-symbol.h".
3226 (md_pcrel_from): Remove undefined sym fudge.
3227 (md_apply_fix3): Remove *valP fudges and code to subtract abs sym.
3228 Don't set fx_addnumber.
3229 (tc_gen_reloc): Remove spurious fx_addnumber comment.
3230 * config/tc-arc.h (MD_APPLY_SYM_VALUE): Define.
3231 (EXTERN_FORCE_RELOC): Define.
3232 * config/tc-arm.c (md_apply_fix3 <case BFD_RELOC_ARM_GOTPC>): Remove.
3233 (tc_gen_reloc): Fudge ARM_GOTPC addend.
3234 (arm_validate_fix): Return void.
3235 (arm_fix_adjustable <elf version>): Remove extern and weak tests.
3236 Add plt and got reloc tests.
3237 (arm_force_relocation): Call S_FORCE_RELOC.
3238 * config/tc-arm.h (struct fix): Forward declare.
3239 (TC_VALIDATE_FIX): No longer set add_symbolP.
3240 (arm_validate_fix): Adjust declaration.
3241 (TC_FORCE_RELOCATION <pe version): Call S_FORCE_RELOC.
3242 (TC_FORCE_RELOCATION): Tweak param name.
3243 (TC_INIT_FIX_DATA): Likewise.
3244 (obj_fix_adjustable): Don't define.
3245 (tc_fix_adjustable): Call arm_fix_adjustable.
3246 (TC_FIX_ADJUSTABLE): Don't define.
3247 (MD_APPLY_SYM_VALUE): Define.
3248 (EXTERN_FORCE_RELOC): Define.
3249 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3250 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3251 * config/tc-avr.c (md_apply_fix3): Don't cast valP pointer type.
3252 Remove *valP fudges and code to subtract abs sym. Don't set
3254 * config/tc-avr.h: Formatting.
3255 (EXTERN_FORCE_RELOC): Define.
3256 (MD_APPLY_SYM_VALUE): Define.
3257 * config/tc-cris.c (tc_gen_reloc): Don't use fx_addnumber.
3258 (md_apply_fix3): Remove code to subtract abs sym.
3259 (md_cris_force_relocation): Update comment. Call S_FORCE_RELOC.
3260 * config/tc-cris.h (TC_FORCE_RELOCATION): Tweak param name.
3261 (IS_CRIS_PIC_RELOC, tc_fix_adjustable): Likewise.
3262 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3263 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3264 (tc_fix_adjustable): Remove extern, weak tests.
3265 * config/tc-d10v.c (tc_gen_reloc): Don't fiddle address of
3266 BFD_RELOC_VTABLE_INHERIT relocs. Don't use fx_addnumber.
3267 (md_apply_fix3): Don't cast valP pointer type. Remove *valP fudges
3268 and code to subtract abs sym.
3269 (d10v_fix_adjustable): Remove extern, weak, SEC_MERGE tests.
3270 (d10v_force_relocation): Call S_FORCE_RELOC.
3271 * config/tc-d10v.h: Don't include write.h.
3272 (struct fix): Instead, forward declare.
3273 (MD_PCREL_FROM_SECTION): Tweak param names.
3274 (md_pcrel_from_section): Use "struct fix" instead of "fixS".
3275 (d10v_fix_adjustable): Likewise.
3276 (obj_fix_adjustable): Don't define.
3277 (tc_fix_adjustable): Define.
3278 (TC_FORCE_RELOCATION): Tweak param name.
3279 (MD_APPLY_SYM_VALUE): Define.
3280 (EXTERN_FORCE_RELOC): Define.
3281 * config/tc-d30v.c (tc_gen_reloc): Don't use fx_addnumber.
3282 (md_apply_fix3): Don't cast valP pointer type. Remove *valP fudges
3283 and code to subtract abs sym.
3284 * config/tc-d30v.h: Don't include write.h.
3285 (struct fix): Instead, forward declare.
3286 (MD_PCREL_FROM_SECTION): Tweak param names.
3287 (md_pcrel_from_section): Use "struct fix" instead of "fixS".
3288 (MD_APPLY_SYM_VALUE): Define.
3289 (EXTERN_FORCE_RELOC): Define.
3290 * config/tc-dlx.c (md_dlx_force_relocation): Call S_FORCE_RELOC.
3291 (md_dlx_fix_adjustable): Don't test for NULL fx_addsy.
3292 (md_apply_fix3): No need to test fx_pcrel before setting fx_done.
3293 (tc_gen_reloc): Don't fiddle with BFD_RELOC_VTABLE_INHERIT relocs.
3294 Adjust the address for BFD_RELOC_VTABLE_ENTRY, not the addend.
3295 Don't use fx_addnumber.
3296 * config/tc-dlx.h: Don't include write.h.
3297 (md_convert_frag): We're not alpha twice over.
3298 (EXTERN_FORCE_RELOC): Define.
3299 (obj_fix_adjustable): Don't define.
3300 (tc_fix_adjustable): Define.
3301 (TC_FIX_ADJUSTABLE): Don't define.
3302 (MD_APPLY_SYM_VALUE): Define.
3303 (LOCAL_LABELS_DOLLAR): Undef always.
3304 * config/tc-fr30.c (fr30_force_relocation): Call S_FORCE_RELOC.
3305 (fr30_fix_adjustable): Remove extern, weak tests.
3306 * config/tc-fr30.h (MD_APPLY_SYM_VALUE): Define.
3307 (obj_fix_adjustable): Don't define.
3308 (tc_fix_adjustable): Define.
3309 (struct fix): Forward declare.
3310 (MD_PCREL_FROM_SECTION): Tweak param name. Remove duplicate.
3311 * config/tc-frv.c (frv_force_relocation): Call S_FORCE_RELOC.
3312 (frv_fix_adjustable): Don't do extern, weak tests.
3313 * config/tc-frv.h (MD_APPLY_SYM_VALUE): Define.
3314 (TC_FIX_ADJUSTABLE): Don't define.
3315 (obj_fix_adjustable): Don't define.
3316 (tc_fix_adjustable): Define.
3317 (struct fix): Forward declare.
3318 (MD_PCREL_FROM_SECTION): Tweak param name. Remove duplicate.\
3319 * config/tc-h8300.c (md_apply_fix3): Don't cast valP pointer type.
3320 * config/tc-h8300.h (struct fix): Forward declare.
3321 (EXTERN_FORCE_RELOC): Define.
3322 * config/tc-hppa.c (md_apply_fix3): Remove subtraction of sym value.
3323 (hppa_fix_adjustable): Don't test extern or weak. Don't
3324 symbol_mark_used_in_reloc.
3325 (hppa_force_relocation): Use S_FORCE_RELOC instead of S_IS_EXTERNAL
3327 * config/tc-hppa.h (EXTERN_FORCE_RELOC): Define
3328 (TC_FORCE_RELOCATION, TC_INIT_FIX_DATA): Tweak param name.
3329 (TC_FORCE_RELOCATION_SECTION): Don't define.
3330 (MD_APPLY_SYM_VALUE): Define.
3331 * config/tc-i370.c (md_apply_fix3): Remove *valP fudges and code
3332 to subtract abs sym.
3333 * config/tc-i370.h: Remove ifdef OBJ_ELF tests.
3334 (MD_APPLY_SYM_VALUE): Define.
3335 (MD_PCREL_FROM_SECTION): Tweak param name.
3336 * config/tc-i386.c (tc_i386_fix_adjustable): Remove weak, extern tests.
3337 (md_apply_fix3): Test fx_addsy, not fx_pcrel.
3338 (i386_force_relocation): New function.
3339 * config/tc-i386.h (TC_COFF_FIX2RTYPE): Tweak param name.
3340 (TC_VALIDATE_FIX): Likewise.
3341 (TC_FIX_ADJUSTABLE): Don't define.
3342 (MD_APPLY_SYM_VALUE): Define.
3343 (TC_FORCE_RELOCATION <elf version>): Call i386_force_relocation.
3344 (i386_force_relocation): Declare.
3345 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3346 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3347 (TC_FORCE_RELOCATION <coff version>): Call S_FORCE_RELOC.
3348 * config/tc-i860.c (md_apply_fix3): Don't cast valP pointer type.
3349 * config/tc-i860.h (MD_APPLY_SYM_VALUE): Define.
3350 (EXTERN_FORCE_RELOC): Define.
3351 * config/tc-i960.c (reloc_callj): Remove declaration. Return false.
3352 (md_apply_fix3): Don't cast valP pointer type. Move code here from
3353 old fixup_segment. No need to test fx_pcrel before setting fx_done.
3354 (i960_validate_fix): Remove add_symbolPP arg and add_symbolP macro.
3355 Use fx_addsy instead of add_symbolP, as_bad_where instead of as_bad.
3356 Remove #if 0 code. Invert return boolean.
3357 * config/tc-i960.h (TC_COUNT_RELOC): Tweak param name.
3358 (TC_COFF_FIX2RTYPE, TC_ADJUST_RELOC_COUNT, TC_VALIDATE_FIX): Likewise.
3359 (tc_headers_hook, tc_coff_fix2rtype): Remove declaration.
3360 (tc_coff_sizemachdep): Prototype.
3361 (i960_handle_align): Likewise.
3362 (i960_validate_fix): Adjust declaration.
3363 (reloc_callj): Likewise.
3364 (EXTERN_FORCE_RELOC): Define.
3365 (TC_FORCE_RELOCATION_SUB_SAME): Define.
3366 (TC_FORCE_RELOCATION_ABS): Define.
3367 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3368 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3369 (TC_FIX_ADJUSTABLE): Don't define.
3370 (MD_APPLY_SYM_VALUE): Define.
3371 * config/tc-ia64.c (ia64_force_relocation): Call S_FORCE_RELOC.
3372 * config/tc-ia64.h (MD_APPLY_SYM_VALUE): Define.
3373 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3374 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3375 * config/tc-ip2k.c (ip2k_force_relocation): Call S_FORCE_RELOC.
3376 * config/tc-ip2k.h (MD_APPLY_FIX3): Don't define.
3377 (MD_APPLY_SYM_VALUE): Define.
3378 (EXTERN_FORCE_RELOC): Define.
3379 (TC_FORCE_RELOCATION): Tweak param name.
3380 * config/tc-m32r.c (m32r_force_relocation): Call S_FORCE_RELOC.
3381 (m32r_fix_adjustable): Don't test extern, weak.
3382 * config/tc-m32r.h (MD_PCREL_FROM_SECTION): Tweak param name.
3383 (MD_APPLY_SYM_VALUE): Define.
3384 (obj_fix_adjustable): Don't define.
3385 (tc_fix_adjustable): Define.
3386 (tc_frob_file): Don't define.
3387 (tc_frob_file_before_fix): Define.
3388 (EXTERN_FORCE_RELOC): Define.
3389 * config/tc-m68hc11.c (tc_gen_reloc): Set addend to zero. Adjust
3390 BFD_RELOC_VTABLE_ENTRY address.
3391 (tc_m68hc11_force_relocation): Call S_FORCE_RELOC.
3392 (tc_m68hc11_fix_adjustable): Don't test relaxable_symbol.
3393 (md_apply_fix3): Remove *valP fudges and code to subtract abs sym.
3394 Remove duplicated fx_done code.
3395 * config/tc-m68hc11.h (MD_APPLY_SYM_VALUE): Define.
3396 (EXTERN_FORCE_RELOC): Define.
3397 * config/tc-m68k.c (tc_m68k_fix_adjustable): Don't test
3399 * config/tc-m68k.h (TC_COFF_FIX2RTYPE): Tweak param name.
3400 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3401 (TC_FIX_ADJUSTABLE): Don't define.
3402 (EXTERN_FORCE_RELOC): Define.
3403 (MD_APPLY_SYM_VALUE): Define.
3404 (TC_FORCE_RELOCATION): Call S_FORCE_RELOC.
3405 * config/tc-mcore.c (md_apply_fix3): Don't cast valP pointer type.
3406 Remove fx_addsy tests.
3407 (mcore_force_relocation): Call S_FORCE_RELOC.
3408 (mcore_fix_adjustable): Don't test fx_addsy.
3409 * config/tc-mcore.h (MD_PCREL_FROM_SECTION): Tweak param name.
3410 (EXTERN_FORCE_RELOC): Define.
3411 (obj_fix_adjustable): Don't define.
3412 (tc_fix_adjustable): Define.
3413 (MD_APPLY_SYM_VALUE): Define.
3414 * config/tc-mips.c (enum mips_pic_level): Move to tc-mips.h.
3415 (mips_pic): No longer static.
3416 (mips_force_relocation): Call S_FORCE_RELOC.
3417 (mips_fix_adjustable): Remove extern, weak tests.
3418 * config/tc-mips.h (enum mips_pic_level): Declare.
3419 (mips_pic): Declare.
3420 (tc_frob_file): Don't define.
3421 (tc_frob_file_before_fix): Define this instead.
3422 (EXTERN_FORCE_RELOC): Define.
3423 * config/tc-mmix.c (md_apply_fix3): Replace real_reg_section tests
3424 with reg_section tests. Set fx_done instead of calling
3425 symbol_clear_used_in_reloc on bad relocs.
3426 (tc_gen_reloc): Zero fx_addsy on bad relocs.
3427 (mmix_force_relocation): Remove weak sym test. Call S_FORCE_RELOC.
3428 (mmix_adjust_symtab): Simplify list handling. Abort on any
3430 * config/tc-mmix.h (tc_fix_adjustable): Remove weak tests. Check
3431 BFD_RELOC_MMIX_LOCAL.
3432 (tc_frob_symbol): Keep user defined syms in reg_section. Don't punt.
3433 (EXTERN_FORCE_RELOC): Define.
3434 (MD_PCREL_FROM_SECTION): Tweak param name.
3435 (tc_frob_file): Don't define.
3436 (tc_frob_file_before_fix): Define this instead.
3437 * config/tc-mn10300.c (mn10300_force_relocation): Call S_FORCE_RELOC.
3438 Remove SEC_CODE checks.
3439 (mn10300_fix_adjustable): Remove extern and weak tests.
3440 * config/tc-mn10300.h (EXTERN_FORCE_RELOC): Define.
3441 (TC_FORCE_RELOCATION): Tweak param name.
3442 (obj_fix_adjustable): Don't define.
3443 (TC_FORCE_RELOCATION_SUB_SAME): Define to handle SEC_CODE.
3444 * config/tc-ns32k.h (TC_FIX_DATA_PRINT): Tweak param name.
3445 * config/tc-openrisc.c (openrisc_force_relocation): Call S_FORCE_RELOC.
3446 (openrisc_fix_adjustable): Don't test fx_addsy.
3447 * config/tc-openrisc.h (MD_APPLY_SYM_VALUE): Define.
3448 (obj_fix_adjustable): Don't define.
3449 (tc_fix_adjustable): Define this instead.
3450 (MD_PCREL_FROM_SECTION): Remove duplicate. Tweak param name.
3451 * config/tc-or32.c (md_apply_fix3): Don't cast valP pointer type.
3452 (tc_gen_reloc): Don't fiddle with BFD_RELOC_VTABLE_INHERIT relocs.
3453 Adjust the address for BFD_RELOC_VTABLE_ENTRY, not the addend.
3454 * config/tc-or32.h (EXTERN_FORCE_RELOC): Define.
3455 (MD_APPLY_SYM_VALUE): Define.
3456 * config/tc-pj.c (md_apply_fix3): Don't cast valP pointer type.
3457 Don't subtract symbol value.
3458 * config/tc-pj.h (md_pcrel_from): Tweak param name.
3459 (EXTERN_FORCE_RELOC): Define.
3460 (TC_FORCE_RELOCATION): Call S_FORCE_RELOC.
3461 (MD_APPLY_SYM_VALUE): Define.
3462 (obj_fix_adjustable): Don't define.
3463 (tc_fix_adjustable): Define this instead.
3464 * config/tc-ppc.c (ppc_frob_symbol <coff version>): Ignore absolute
3466 (ppc_force_relocation <coff version>): Call S_FORCE_RELOC.
3467 (ppc_force_relocation <elf version>): New.
3468 (ppc_fix_adjustable <elf version>): Remove extern and weak tests.
3469 (md_apply_fix3): Don't subtract symbol values for ELF. Update
3470 comments. Don't subtract fx_subsy as that is already done.
3471 * config/tc-ppc.h (tc_fix_adjustable): Tweak param name.
3472 (MD_PCREL_FROM_SECTION): Likewise.
3473 (TC_FORCE_RELOCATION): Define for both ELF and XCOFF as calling
3474 ppc_force_relocation.
3475 (TC_FORCE_RELOCATION_SECTION): Delete.
3476 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3477 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3478 (MD_APPLY_SYM_VALUE): Define.
3479 * config/tc-s390.c: #include "dwarf2dbg.h".
3480 (s390_insn): Remove excess parens.
3481 (tc_s390_fix_adjustable): Remove extern, weak, SEC_MERGE tests.
3482 (tc_s390_force_relocation): Call S_FORCE_RELOC.
3483 (md_apply_fix3): Add ATTRIBUTE_UNUSED on "seg". Abort when fx_subsy
3484 non-NULL. Don't subtract off fx_addsy value.
3485 * config/tc-s390.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3486 (TC_FORCE_RELOCATION, MD_PCREL_FROM_SECTION): Tweak param name.
3487 (TC_FIX_ADJUSTABLE): Don't define.
3488 (MD_APPLY_SYM_VALUE): Define.
3489 * config/tc-sh.c (SWITCH_TABLE_CONS): Move to tc-sh.h.
3490 (SWITCH_TABLE): Likewise.
3491 (sh_force_relocation): Call S_FORCE_RELOC.
3492 (sh_fix_adjustable): Remove "return 1" cases handled by the default.
3493 Replace TC_RELOC_RTSYM_LOC_FIXUP with reloc type tests.
3494 (md_apply_fix3 <case BFD_RELOC_32_PLT_PCREL>): Simplify,
3495 fx_addnumber is zero on entry. Save val in fx_addnumber.
3496 (tc_gen_reloc): Don't subtract fx_subsy.
3497 * config/tc-sh.h (struct fix): Move.
3498 (obj_fix_adjustable): Don't define.
3499 (tc_fix_adjustable): Define this instead.
3500 (TC_FIX_ADJUSTABLE): Don't define.
3501 (MD_APPLY_SYM_VALUE): Define.
3502 (SWITCH_TABLE_CONS): Define.
3503 (SWITCH_TABLE): Define.
3504 (TC_FORCE_RELOCATION_SUB_SAME): Define.
3505 (TC_VALIDATE_FIX_SUB): Define.
3506 (MD_PCREL_FROM_SECTION): Tweak param name.
3507 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3508 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3509 (TC_FORCE_RELOCATION_SUB_ABS): Define.
3510 * config/tc-sh64.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3511 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3512 (TC_FORCE_RELOCATION_SUB_SAME): Define.
3513 (TC_VALIDATE_FIX_SUB): Define.
3514 (MD_PCREL_FROM_SECTION): Tweak param name.
3515 * config/tc-sparc.c (md_apply_fix3): Don't subtract off symbol value.
3516 (tc_gen_reloc): Use S_FORCE_RELOC.
3517 (elf32_sparc_force_relocation): Call S_FORCE_RELOC.
3518 * config/tc-sparc.h (TC_FORCE_RELOCATION <coff version>): Remove.
3519 (TC_FORCE_RELOCATION_ABS): Define this instead.
3520 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3521 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
3522 (tc_fix_adjustable): Remove extern and weak tests. Use S_FORCE_RELOC.
3523 (MD_APPLY_SYM_VALUE): Define.
3524 (TC_FIX_DATA_PRINT): Tweak param name.
3525 * config/tc-tic30.c (USE_STDOUT): Don't define.
3526 (md_parse_option): Remove stupid debug code.
3527 (tc_gen_reloc): Don't use fx_addnumber.
3528 * config/tc-v850.c (v850_fix_adjustable): Remove extern and weak tests.
3529 (v850_force_relocation): Remove weak test. Call S_FORCE_RELOC.
3530 * config/tc-v850.h (obj_fix_adjustable): Don't define.
3531 (tc_fix_adjustable): Define this instead.
3532 (TC_FIX_ADJUSTABLE): Don't define.
3533 (MD_APPLY_SYM_VALUE): Define this instead.
3534 (TC_FORCE_RELOCATION, MD_PCREL_FROM_SECTION): Tweak param name.
3535 * config/tc-vax.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
3536 (TC_FIX_ADJUSTABLE): Don't define.
3537 (MD_APPLY_SYM_VALUE): Define this instead.
3538 (tc_fix_adjustable): Remove extern and weak tests.
3539 * config/tc-w65.h (struct fix): Forward declare.
3540 * config/tc-xstormy16.c (xstormy16_force_relocation): Call
3542 (xstormy16_fix_adjustable): Remove extern and weak tests. Don't
3543 call xstormy16_force_relocation; Instead test for FPTR16 reloc.
3544 (xstormy16_md_apply_fix3): Remove *valP fudges and code to subtract
3546 * config/tc-xstormy16.h (MD_APPLY_FIX3): Don't define.
3547 (MD_APPLY_SYM_VALUE): Define.
3548 (obj_fix_adjustable): Don't define.
3549 (tc_fix_adjustable): Define this instead.
3550 (MD_PCREL_FROM_SECTION): Remove duplicate. Tweak param name.
3554 * config/tc-ppc.c (ppc_frob_symbol): Formatting, warning fix.
3555 (ppc_fix_adjustable <coff version>): Cleanup.
3557 * config/tc-ppc.c (PPC_HIGHER, PPC_HIGHEST): Fix warning.
3558 (md_parse_option): No -a64 without BFD64.
3559 (ppc_set_cpu): Select appropriate cpu when ppc_obj64.
3560 (ppc_arch): Use bfd_mach_rs6k for bfd_arch_rs6000.
3564 * config/tc-ppc.c (md_begin): Do not insert non-BookE32
3565 instructions into the hash table if the target cpu is the BookE32.
3569 * read.c (do_align): Use ATTRIBUTE_UNUSED_LABEL for label, not
3574 * tc-v850.c (v850_relax): Declare.
3575 (v850_longcode): New routine.
3576 (v850_handle_align): New routine.
3577 (md_pseudo_table): Add longcall and longjump.
3578 (md_parse_option): Check for relax option.
3579 (tc_gen_reloc): Handle BFD_RELOC_V850_LONGCALL,
3580 BFD_RELOC_V850_LONGJUMP, and BFD_RELOC_V850_ALIGN.
3581 (md_apply_fix3): Likewise.
3582 (v850_force_relocation): Likewise.
3583 (v850_comm): Change the current section.
3584 (md_assemble): Ensure that the correct value is put in the
3586 (v850_sdata, v850_tdata, v850_zdata, v850_sbss, v850_tbss,
3587 v850_zbss, v850_rosdata, v850_rozdata): Fix section book keeping.
3588 Remove redundant v850ea support.
3589 * tc-v850.h (HANDLE_ALIGN): Define.
3590 (v850_handle_align): Declare.
3591 * doc/c-v850.c: Document -mrelax, .longcall and .longjump.
3595 * configure.in: Add tic4x-coff* and c4x-coff*-coff-coff targets.
3596 * configure: Regenerate.
3597 * NEWS: Mention new port.
3601 * config/obj-coff.c: Add sdef definition.
3602 * config/obj-coff.h: Add tic4x include file and set
3604 * config/tc-tic4x.c: New file.
3605 * config/tc-tic4x.h: New file.
3609 * write.c (BFD_FAST_SECTION_FILL): Remove unused macro.
3610 (TC_ADJUST_RELOC_COUNT): Tweak param name.
3611 (TC_FORCE_RELOCATION, TC_FORCE_RELOCATION_SECTION): Likewise.
3612 (TC_FIX_ADJUSTABLE, MD_PCREL_FROM_SECTION): Likewise.
3613 (RELOC_ENUM): Define.
3614 (fix_new_internal): Use RELOC_ENUM.
3615 (fix_new, fix_new_exp): Likewise.
3616 (adjust_reloc_syms): Comment. Remove unnecessary tests on sym != NULL.
3617 Replace gotos with continue.
3618 (write_relocs): Formatting. Avoid symbol loops in
3619 RELOC_EXPANSION_POSSIBLE case too. Report bfd_reloc_outofrange
3620 errors, and error number in other cases.
3621 (fixup_segment): Remove param names from prototype. Rename
3622 "this_segment_type" to "this_segment". Update linkrelax comment.
3623 Remove "size, "place" and "where" local vars. Formatting. Update
3624 "no symbol" comment. Remove #if 0 and #if 1.
3626 * app.c (do_scrub_chars): Don't test IGNORE_NONSTANDARD_ESCAPES. Tidy.
3630 * dwarf2dbg.c: Always include dwarf2dbg.h.
3631 (dwarf2_directive_file): Adjust dummy version args.
3632 * ecoff.c (ecoff_directive_weakext): Add ATTRIBUTE_UNUSED.
3633 * expr.c (clean_up_expression <O_subtract>): Allow subtraction
3634 when symbol values differ.
3635 * read.c (do_align): Add ATTRIBUTE_UNUSED to label.
3636 (pseudo_set <O_subtract>): Remove unnecessary segment test.
3637 * config/obj-bout.c (obj_pseudo_table): Warning fix.
3641 * config/tc-w65.c (md_section_align): Fix typo.
3642 (md_parse_option): Return 0, not 1.
3646 * doc/as.texinfo (Section): Note that if '@' is a comment
3647 character then another symbol is used to prefix the section's
3652 * config/tc-z8k.c (get_operands): Adjust ptr variable also in
3657 * configure.in: Add support for sh-**-nto* target.
3658 * configure: Regenerate.
3662 * config/tc-h8300.h (TC_LINKRELAX_FIXUP): Define.
3666 * config/tc-ppc.c (ppc_cleanup): Do something only if format
3668 (ppc_apuinfo_section_add): Define only if format is ELF.
3669 (md_assemble): Emit APUinfo section only if format is ELF.
3674 * config/tc-arc.c (md_pseudo_table <dwarf2_directive_file>): Cast.
3675 * config/tc-frv.c: Likewise.
3676 * config/tc-hppa.c: Likewise.
3677 * config/tc-ia64.c: Likewise.
3678 * config/tc-ip2k.c: Likewise.
3679 * config/tc-m68hc11.c: Likewise.
3680 * config/tc-m68k.c: Likewise.
3681 * config/tc-mmix.c: Likewise.
3682 * config/tc-mn10300.c: Likewise.
3683 * config/tc-sh.c: Likewise.
3684 * config/tc-sparc.c: Likewise.
3685 * config/tc-v850.c: Likewise.
3689 * config/tc-mips.c (macro2): Implement rotates by zero using shifts
3696 * config/tc-ppc.c (PPC_OPCODE_CLASSIC): Enable this everywhere
3697 PPC_OPCODE_PPC is, except for BookE architectures.
3698 (md_parse_option): Add support for -mspe.
3699 (md_show_usage): Add -mspe.
3700 (md_parse_option): Add support for -me500 and
3701 -me500x2 to generate code for Motorola e500 core complex.
3702 (md_show_usage): Add -me500 and -me500x2.
3704 (PPC_APUINFO_ISEL, PPC_APUINFO_PMR, PPC_APUINFO_RFMCI,
3705 PPC_APUINFO_CACHELCK, PPC_APUINFO_SPE, PPC_APUINFO_EFS,
3706 PPC_APUINFO_BRLOCK): New macros.
3708 (ppc_cleanup): New function.
3709 (ppc_apuinfo_section_add): New function.
3711 (md_assemble): Collect info and write the APUinfo section.
3713 * config/tc-ppc.h (md_cleanup): Define.
3714 (ppc_cleanup): Export.
3715 (ELF_TC_SPECIAL_SECTIONS): Add .PPC.EMB.apuinfo section.
3719 * config/obj-elf.c (obj_elf_change_section): Make non-static.
3720 config/tc-mips.c (s_change_section): New function to support
3721 IRIX .section pseudo-op.
3725 * config/tc-v850.c (md_assemble): Fix assembling of "callt 0x3f".
3729 * config/tc-mips.c (macro_build_jalr): Make sure we generate
3730 the fix-up against on the right frag.
3731 (s_cpsetup): Likewise. Parse third argument as expression, to
3732 handle global symbols and forward/backward labels correctly.
3736 * read.c (stringer): Catch attempts to create strings in the abs
3739 * config/tc-alpha.c: Fix compiling for COFF targets.
3740 Some minor formatting tidyups.
3744 * config/tc-m68hc11.h (MD_PCREL_FROM_SECTION): Remove.
3745 (TC_HANDLES_FX_DONE): Define to let md_apply_fix3 set fx_done flag
3746 according to the reloc.
3747 (tc_fix_adjustable, tc_m68hc11_fix_adjustable): Define.
3748 (TC_FORCE_RELOCATION): Define.
3749 (tc_m68hc11_force_relocation): Declare.
3751 * config/tc-m68hc11.c (md_pseudo_table): Add relax command.
3752 (s_m68hc11_relax): New function for relax group.
3753 (build_insn, build_jump_insn): Emit a M68HC11_RL_JUMP reloc at
3754 beginning of jump instruction.
3755 (md_pcrel_from): Rename from md_pcrel_from_section and fix
3756 address computation.
3757 (tc-gen_reloc): Update.
3758 (md_estimate_size_before_relax): Create the BFD_RELOC_16_PCREL as
3760 (tc_m68hc11_force_relocation): New function, handle new relocs.
3761 (tc_m68hc11_fix_adjustable): New to make sure there are enough
3762 reloc for the linker relax pass.
3763 (md_apply_fix3): Handle M68HC11_RL_JUMP, M68HC11_RL_GROUP
3768 * config/tc-m68hc11.c (m68hc11_elf_final_processing): New function.
3769 (md_pseudo_table): Add .mode, .far and .interrupt pseudo op.
3770 (s_m68hc11_mode): New function for .mode pseudo op.
3771 (s_m68hc11_mark_symbol): New function for .far and .interrupt
3773 * config/tc-m68hc11.h (elf_tc_final_processing): Define.
3774 (m68hc11_elf_final_processing): Declare.
3778 * config/tc-m68hc11.c (md_begin): Take into account additional
3779 page operand for call instruction.
3780 (print_opcode_format): Likewise.
3781 (check_range): Likewise for page range checking.
3782 (get_operand): Don't skip a possible comma in operands.
3783 (fixup8): Generate BFD_RELOC_M68HC11_PAGE reloc.
3784 (fixup16): Likwise with BFD_RELOC_M68HC11_LO16.
3785 (fixup24): New to handle call reloc.
3786 (build_insn): Handle missing page operand for call instruction.
3788 (md_apply_fix3): Take into account new relocs.
3789 (get_operand): Fix the mode for indexed indirect addressing.
3790 (build_indexed_byte): Fix post index byte for indexed indirect mode.
3794 * config/tc-mips.c (mips_ip): Don't work out the value of
3795 constant %hi()s here.
3799 * config/tc-i386.c (tc_i386_fix_adjustable): Test OUTPUT_FLAVOR
3800 for ELF, and don't bother checking ELF relocs when non-ELF.
3801 (i386_immediate): Allow absolute_section expressions for aout.
3802 (i386_displacement): Likewise. Also test bfd_is_com_section.
3803 (md_estimate_size_before_relax): Test OUTPUT_FLAVOR for ELF.
3804 (md_apply_fix3): Hack for bfd_install_relocation when fx_pcrel,
3805 not when fx_addsy. Remove dead code.
3809 * configure.in: Add support for ppc-*-nto* target.
3810 * configure: Regenerate.
3814 * config/tc-i386.h: Reorganize.
3818 * config/tc-mips.c (macro): Handle a register plus a 16-bit
3819 immediate offset in "dla" and "la" expansions.
3823 * configure.in: bfd_gas=yes for all i386 targets. Formatting.
3824 Remove "bfd_gas=yes" from target table when covered later.
3825 Consolidate some entries.
3826 * configure: Regenerate
3830 * config/tc-i386.c (output_insn): Save frag_now and frag_now_fix ()
3831 at start of insn, pass it to output_disp and output_imm.
3832 (output_disp): Added arguments. If _GLOBAL_OFFSET_TABLE_ is seen
3833 in displacement for R_386_32 reloc, use R_386_GOTPC and compute
3835 (output_imm): Added arguments. Compute properly addend for
3837 (md_apply_fix3): Remove R_386_GOTPC handling.
3841 * config/tc-alpha.c (cpu_types): Enabled ev67, ev68, -m21264a
3842 and m21264b processor names and cpu types.
3843 * doc/c-alpha.texi: Documented new types.
3847 * config/tc-ppc.c (md_apply_fix3): Adjust 16 bit XCOFF reloc offset.
3851 * config/tc-mips.c (tc_gen_reloc): Extend GP-relative addend
3852 handling to BFD_RELOC_MIPS16_GPREL.
3856 * config/tc-arm.c (add_to_lit_pool): Ensure that offset to literal
3857 pool is computed using signed arithmetic so that proper sign
3858 extension is performed if X_add_number is a 64-bit integer.
3863 * dwarf2dbg.c (dwarf2_finish): Don't emit unreferenced
3864 .debug_line section unless it has line information.
3868 * config/tc-ns32k.h (md_pcrel_adjust): Supply prototype.
3869 * config/tc-ns32k.c (convert_iif, md_parse_option, md_show_usage):
3870 Allow default displacement size to be an option "--disp-size-default".
3871 (md_number_to_disp): Make error messages include value. Use %d to
3872 print integers, not %s.
3873 (fix_new_ns32k): Conditionally set fx_no_overflow so we don't
3874 get duplicate messages sometimes.
3875 (convert_iif): Grow frag to max possible instruction size. Avoid
3876 creating unnecessary fixes.
3877 (md_number_to_field) Add prototype.
3878 (encode_operand, parse, convert_iif, md_fix_pcrel_adjust): Add
3879 prototypes and make static.
3880 (struct addr_mode): Make mode and scaled_mode explicitly signed.
3881 (evaluate_expr): Use resultP argument instead of exprP global.
3882 (get_addr_mode): Quiten compiler warnings.
3883 (encode_operand): eliminate unused variables. Quiten compiler
3884 warnings. Eliminate nul character in format strings.
3885 (parse): argc is unsigned.
3886 (reloc): Type cast index to quieten compiler.
3887 (md_pcrel_adjust, md_apply_fix3): Remove unused variable.
3888 (md_convert_frag): Note unused parameters. Remove unused
3890 (md_create_long_jump, md_create_short_jump,
3891 md_undefined_symbol_name, md_section_align, tc_gen_reloc): Note
3896 * NEWS: Retroactively add entry for Lars Brinkhoff's contribution
3897 of the PDP-11 and 2.11BSD a.out support.
3901 * config/tc-v850.c (md_assemble): Fix range check for immediate
3906 * config/tc-mips.c (mips_cpu_info_table): Clean up entries related
3907 to Broadcom SB-1 core support.
3911 * config/tc-mips.c (mips_target_format): Fix formatting.
3912 Add recognition of n32 ABI formats.
3916 * tc-mips.c (load_address): Don't clobber $at when loading a
3917 64-bit address in non-PIC code if noat is in effect.
3922 * config/tc-mips.c (macro): Use codes 6 and 7 in trap instructions
3923 used in division/multiply macro expansions similarly to how they
3924 are used in the variants with break instructions.
3929 * configure.in: Add support for arm-*-nto target.
3930 * configure: Regenerate.
3934 * config/tc-arm.c (struct literal_pool): Add fields to allow
3935 multiple literal pools to be maintained.
3936 (find_literal_pool): New function.
3937 (find_or_make_literal_pool): New function.
3938 (add_to_literal_pool): Use find_or_make_literal_pool.
3939 (arm_s_text, arm_s_data, arm_s_section): Remove - no longer
3941 (s_ltorg): Use find_literal_pool.
3942 (arm_cleanup): Dump all literal pools.
3943 * doc/c-arm.texi: Document new behaviour of only dumping literal
3948 * config/tc-ppc.c (ppc_set_cpu): Use PPC_OPCODE_64 as the default
3949 rather than PPC_OPCODE_32 for powerpc64*.
3953 * po/es.po: Updated Spanish translation.
3954 * po/fr.po: Updated French translation.
3958 * doc/c-mips.texi: Remove -mcpu. Document -mabi.
3959 * configure.in (MIPS_CPU_STRING_DEFAULT): New configuration macro.
3960 (USE_E_MIPS_ABI_O32, MIPS_DEFAULT_64BIT): New configuration macros.
3961 * configure, config.in: Regenerate.
3962 * config/tc-mips.c (file_mips_abi): Rename to mips_abi.
3963 (mips_set_options): Remove "abi" field.
3964 (mips_opts): Update accordingly. Replace all uses of mips_opts.abi
3967 (mips_arch_string, mips_arch_info): New vars.
3968 (mips_tune_string, mips_tune_info): New vars.
3969 (ABI_NEEDS_32BIT_REGS, ABI_NEEDS_64BIT_REGS): New macros.
3970 (HAVE_32BIT_GPRS, HAVE_32BIT_FPRS): Don't check the ABI.
3971 (mips_isa_to_str, mips_cpu_to_str): Remove.
3972 (mips_ip): If the selected architecture is a generic ISA rather
3973 than a processor, only mention the ISA level in error messages.
3974 (OPTION_MCPU): Remove.
3976 (md_longopts): Add -mfp64, remove -mcpu.
3977 (mips_set_option_string): New fn.
3978 (md_parse_option): Make -mipsN update file_mips_isa rather than
3979 mips_opts.isa. Use mips_set_option_string to set -march or -mtune.
3980 Don't let -mgp32 and -mfp32 change the ABI.
3981 (show): Move to end of file. Constify string argument.
3982 (md_show_usage): Move to the end of the file. Read available
3983 architectures from mips_cpu_info_table.
3984 (mips_set_architecture): New fn.
3985 (mips_after_parse_args): Rework. Remove -mcpu handling. -mipsN
3986 is an alias for -march=mipsN. Don't change the ABI based on other
3987 flags. Infer the register size from the ABI as well as the
3988 architecture. Complain about more conflicting arguments.
3989 Unify logic with GCC.
3990 (s_mipsset): Don't change the ABI.
3991 (mips_elf_final_processing): Check USE_E_MIPS_ABI_O32.
3992 (mips_cpu_info_table): Remove Generic-MIPS* entries, keeping just
3993 "mipsN"-type entries. Remove entries that vary only in the
3994 manufacturer's prefix, or that have "000" replaced by "k".
3995 Remove TARGET_CPU entries. Make r2000 entry use CPU_R3000.
3996 (mips_strict_matching_cpu_name_p, mips_matching_cpu_name_p): New fns.
3997 (mips_parse_cpu): New fn.
3998 (mips_cpu_info_from_name, mips_cpu_info_from_cpu): Remove.
3999 (mips_cpu_info_from_isa): Minor formatting tweak.
4003 * po/sv.po: Updated Swedish translation.
4004 * po/es.po: Updated Spanish translation.
4008 * Makefile.am: Run "make dep-am".
4009 * Makefile.in: Regenerate.
4010 * doc/Makefile.in: Regenerate.
4011 * config.in: Regenerate.
4012 * po/POTFILES.in: Regenerate.
4016 * po/fr.po: Updated French translation.
4020 * CONTRIBUTORS: Ken Raeburn is no longer the maintainer for
4025 * config/tc-mips.c (mips_need_elf_addend_fixup): Use S_IS_EXTERNAL
4026 instead of S_IS_EXTERN.
4027 (md_estimate_size_before_relax): Likewise.
4028 (mips_fix_adjustable): Likewise.
4032 * config/tc-mips.c (mips_pic_level): Remove IRIX4_PIC.
4037 * cond.c (s_ifdef): Treat a referenced but not yet defined
4038 symbol as if it were undefined, in exactly the same way as
4041 * doc/as.texinfo: Document that .ifdef, .ifndef and .equiv
4042 consider referenced bug not yet defined symbols to be
4051 * configure.in: Add ip2k configuraton.
4052 * configure: Regenerate.
4053 * Makefile.am: Add ip2k configuraton.
4054 * Makefile.in: Regenerate.
4055 * configure: Regenerate.
4056 * Makefile.in: Regenerate.
4057 * config/tc-ip2k.c: New file.
4058 * config/tc-ip2k.h: New files.
4059 * NEWS: Mention new support.
4060 * doc/Makefile.am (CPU_DOCS): Add c-ip2k.texi.
4061 * doc/Makefile.in: Regenerate.
4062 * doc/all.texi: Set IP2K
4063 * doc/as.texinfo: Add IP2K description.
4064 * doc/c-ip2k.texi: New file.
4068 * NEWS: Reformat to match style of other NEWS files.
4072 * tc-i386.c (i386_align_code): Implement x86_64 neutral code fillers.
4076 * config/tc-z8k.c (build_bytes): Correct order of memset args.
4080 * NEWS: Add 'Changes in 2.13'.
4084 * config/tc-hppa.h (LABELS_WITHOUT_COLONS): Define if TE_NetBSD.
4088 * config/tc-i386.c (process_suffix): Merge CODE_64BIT JumpByte
4089 case with non CODE_64BIT case. Don't warn on "qword ptr" if
4094 * config/tc-ppc.c (ppc_elf_frob_symbol): Delete.
4095 (ppc_frob_file_before_adjust): New function.
4096 * config/tc-ppc.h (tc_frob_symbol): Don't define.
4097 (ppc_elf_frob_symbol): Don't declare.
4098 (tc_frob_file_before_adjust): Define.
4099 (ppc_frob_file_before_adjust): Declare.
4101 * config/tc-ppc.c (md_pseudo_table): Warning fix.
4102 (ppc_cpu): Make it unsigned long to agree with struct powerpc_opcode
4105 (ppc_xcoff64): Rename to ppc_obj64.
4106 (md_parse_option <m>): Encode old ppc_size value in ppc_cpu.
4107 (ppc_set_cpu): Set PPC_OPCODE_32 too.
4108 (ppc_arch): Use ppc_obj64 instead of ppc_size to select bfd_mach_ppc64
4110 (ppc_target_format): Use ppc_obj64 to select format.
4111 (md_begin): Adjust for PPC_OPCODE_32/64 in ppc_cpu.
4112 (ppc_insert_operand): Use ppc_obj64 instead of ppc_size.
4113 (ppc_elf_suffix): Likewise. Don't depend on BFD_DEFAULT_TARGET_SIZE.
4114 (tc_frob_symbol): Likewise.
4115 (md_assemble): Use ppc_obj64 instead of ppc_size. Don't depend on
4116 BFD_DEFAULT_TARGET_SIZE.
4118 (ppc_is_toc_sym): Likewise.
4119 (md_apply_fix3): Likewise.
4120 * config/tc-ppc.h (TC_FORCE_RELOCATION): Don't depend on
4121 BFD_DEFAULT_TARGET_SIZE.
4122 (ELF_TC_SPECIAL_SECTIONS): Likewise.
4123 (tc_frob_symbol): Likewise.
4127 * config/tc-mips.c (macro_build): Handle MIPS16 insns.
4128 (mips_ip): Likewise.
4132 * config/tc-i386.c (md_pseudo_table <file>): Warning fix.
4133 (BFD_RELOC_8, BFD_RELOC_8_PCREL): Define for non-BFD.
4134 (md_apply_fix3): Formatting. Remove redundant test.
4135 (tc_gen_reloc): Remove redundant code.
4136 (tc_i386_force_relocation): Delete. Movy body of function to..
4137 * config/tc-i386.h (TC_FORCE_RELOCATION): .. here.
4141 * configure.in: Add hppa-*-openbsd* target, change
4142 alpha*-*-openbsd* format to elf, and use elf for sparc-*-openbsd*
4144 * configure: Regenerate.
4148 * config/tc-mips.c (macro): Shift the 32-bit address range
4149 accessible with a lone "lui" down by 32768.
4153 * config/tc-mips.c (load_address): Use non-trapping "daddu"
4154 instead of "dadd" in address calculations.
4159 * config/tc-i386.c (process_suffix): Remove intel mode movsx and
4161 (md_assemble): Instead, zap the suffix here.
4165 * NEWS: Remove next release number until the release is actually
4170 * Makefile.am (check-DEJAGNU): Revert 2002-06-25 change.
4172 * Makefile.in: Regenerate.
4176 * config/tc-s390.c (tc_s390_fix_adjustable): Prevent any adjustment
4177 to symbols in merge sections, even non pc-relative ones.
4181 * config/tc-m68hc11.h (m68hc11_listing_header): Fix warning.
4185 * config/tc-m68hc11.h (ELF_TC_SPECIAL_SECTIONS): New sections
4186 .softregs and .eeprom.
4190 * NEWS: Note the next release is 2.13.
4194 * po/tr.po: New translation imported.
4198 * ecoff.c: (get_tag): Replace strcpy with xstrdup.
4199 (ecoff_directive_def): Likewise.
4200 (ecoff_directive_tag): Likewise.
4201 * listing.c (file_info): Likewise.
4202 * hash.c (what): Likewise.
4206 * Makefile.am (check-DEJAGNU): Set LC_ALL=C and export it.
4207 * Makefile.in: Regenerated.
4211 * config/tc-sh.c (get_specific): Revert 2002-05-01 change.
4212 (assemble_ppi): Generate warning if the same register is used
4213 twice as destination in the same padd / pmuls instruction.
4217 From Catherine Moore, Michael Meissner, Richard Sandiford, Dave Brolley
4218 * po/POTFILES.in: Add tc-frv.c, tc-frv.h.
4219 * configure.in: Support frv-*-*.
4220 * Makefile.am (CPU_TYPES): Add frv.
4221 (TARGET_CPU_CFILES): Add tc-frv.c.
4222 (TARGET_CPU_HFILES): Add tc-frv.h.
4223 (DEPTC_frv_coff): New variable.
4224 (DEPTC_frv_elf): New variable.
4225 (DEPOBJ_frv_coff): New variable.
4226 (DEPOBJ_frv_elf): New variable.
4227 (DEP_frv_coff): New variable.
4228 (DEP_frv_elf): New variable.
4229 * tc-frv.c: New file.
4230 * tc-frv.h: New file.
4234 * config/obj-elf.h (TARGET_SYMBOL_FIELDS): Conditionally define.
4238 * config/tc-sh.c (assemble_ppi): Initialize reg_n.
4242 * config/tc-i370.h (tc_comment_chars): Define for i370-elf.
4247 * dwarf2dbg.h (dwarf2_directive_file): Return char *.
4248 * dwarf2dbg.c (dwarf2_directive_file): Return filename.
4249 * config/tc-mips.c (s_mips_file): Call s_app_file_string
4250 and new_logical_line for the first .file directive.
4251 * read.c (s_app_file_string): New function.
4252 (s_app_file): Call it.
4253 * read.h (s_app_file_string): Add declaration.
4257 * configure.in: Remove MIPS_STABS_ELF.
4258 * configure: Regenerated.
4259 * config.in: Regenerated.
4260 * config/obj-elf.h (ECOFF_DEBUGGING): Define to mips_flag_mdebug
4262 * config/tc-mips.c (mips_pseudo_table): Remove #ifdef around
4264 (pdr_seg): Declare unconditionally.
4265 (md_begin): Always generate .pdr unless ECOFF_DEBUGGING or not ELF.
4266 (s_mips_end): Likewise. Generate stabs function markers.
4267 (s_mips_ent): Generate stabs function markers.
4268 (s_mips_frame): Always generate .pdr unless ECOFF_DEBUGGING or not
4270 (s_mips_mask): Likewise.
4271 (mips_flag_mdebug): New.
4272 (md_longopts): Add "mdebug" and "no-mdebug".
4273 (md_parse_options): Add OPTION_MDEBUG and OPTION_NO_MDEBUG.
4274 (mips_after_parse_args): Set mips_flag_mdebug.
4275 * doc/as.texinfo: Add "-mdebug" and "-no-mdebug" for MIPS.
4279 * config/tc-mips.c (md_apply_fix3): Don't subtract the symbol's
4280 value twice for RELA relocations.
4284 * symbols.c (resolve_symbol_value): Initialise final_val.
4286 * subsegs.c (subsegs_print_statistics): Cast frchp to void *.
4290 * config/tc-ppc.c (ppc_subseg_align): Delete.
4291 (ppc_change_csect): Default csect align is 2.
4292 * config/tc-ppc.h (SUB_SEGMENT_ALIGN): Delete
4296 * config/tc-avr.c (mcu_types): Update.
4300 * configure.in (vax-*-netbsdelf*, vax-*-netbsdaout*)
4301 (vax-*-netbsd*): New targets.
4302 * configure: Regenerate.
4303 * config/aout_gnu.h (enum machine_type): Add M_VAX4K_NETBSD.
4304 * config/tc-vax.c: Add support for ELF and PIC.
4305 (flag_want_pic): New flag.
4306 (float_cons): Fix prototype.
4307 (md_apply_fix3): Adjust for BFD_ASSEMBLER.
4308 (md_assemble): Introduce a new is_absolute local, and use it
4309 rather than repeating the test. Make fatal errors actually
4310 fatal by using as_fatal as appropriate. Adjust for BFD_ASSEMBLER.
4311 Add support for ELF. Add support for PIC.
4312 (md_convert_frag): Adjust for BFD_ASSEMBLER.
4313 (tc_aout_fix_to_chars): Only include if OBJ_AOUT and not
4315 (vax_reg_parse): Make the % register prefix mandatory for ELF,
4316 optional for a.out, and not allowed for VMS. Adjust all callers.
4317 (md_create_short_jump): Add ATTRIBUTE_UNUSED to unused arguments.
4318 (md_create_long_jump): Likewise.
4319 (md_undefined_symbol): Likewise.
4320 (md_section_align): Likewise.
4321 (md_shortopts): Allow -k and -K for ELF.
4322 (md_parse_option): Set flag_want_pic if -k or -K.
4323 (tc_headers_hook): New function if OBJ_AOUT and not BFD_ASSEMBLER.
4324 (tc_gen_reloc): New function if BFD_ASSEMBLER.
4325 * config/tc-vax.h (tc_headers_hook): Remove.
4326 (TARGET_FORMAT): Set according to object format and target
4328 (BFD_ARCH, TARGET_ARCH): Define.
4329 (NO_RELOC): Adjust for BFD_ASSEMBLER.
4330 (TC_RELOC_RTSYM_LOC_FIXUP, TC_FIX_ADJUSTABLE)
4331 (tc_fix_adjustable): Define if BFD_ASSEMBLER.
4332 * config/vax-inst.h (VAX_JSB, VAX_CALLS, VAX_CALLG): Define.
4336 * Makefile.am: Run "make dep-am".
4337 * Makefile.in: Regenerate.
4339 * as.c: Replace CONST with const.
4340 * write.c: Likewise.
4341 * config/obj-coff.c: Likewise.
4342 * config/tc-a29k.c: Likewise.
4343 * config/tc-arm.c: Likewise.
4344 * config/tc-dlx.c: Likewise.
4345 * config/tc-h8300.c: Likewise.
4346 * config/tc-h8500.c: Likewise.
4347 * config/tc-i370.c: Likewise.
4348 * config/tc-i860.c: Likewise.
4349 * config/tc-i960.c: Likewise.
4350 * config/tc-m68hc11.c: Likewise.
4351 * config/tc-m68k.c: Likewise.
4352 * config/tc-m88k.c: Likewise.
4353 * config/tc-mcore.c: Likewise.
4354 * config/tc-mips.c: Likewise.
4355 * config/tc-ns32k.c: Likewise.
4356 * config/tc-pdp11.c: Likewise.
4357 * config/tc-pj.c: Likewise.
4358 * config/tc-s390.c: Likewise.
4359 * config/tc-sh.c: Likewise.
4360 * config/tc-sparc.c: Likewise.
4361 * config/tc-tahoe.c: Likewise.
4362 * config/tc-tic80.c: Likewise.
4363 * config/tc-v850.c: Likewise.
4364 * config/tc-vax.c: Likewise.
4365 * config/tc-w65.c: Likewise.
4366 * config/tc-z8k.c: Likewise.
4370 Based on patch from Matt Green:
4371 * config/obj-elf.h (ECOFF_DEBUGGING): Make configurable.
4372 * config/tc-mips.c (s_mips_file): Renamed from s_file.
4373 (s_mips_loc): New function.
4374 (mips_nonecoff_pseudo_table): Call them.
4375 (append_insn): Call dwarf2_emit_insn.
4379 * config/tc-mips.c (mips_opts): Fix comment, all ASE fields are set
4381 (file_ase_mips16): New veriable.
4382 (mips_eabi64): Remove.
4383 (CPU_HAS_MIPS16): New define.
4384 (CPU_HAS_MDMX): Fix data type.
4385 (md_begin): Code cleanup. Use file_ase_mips16.
4386 (mips_elf_final_processing): Handle mips16 header flag.
4387 Handle EABI flag without intermediate variable.
4391 * doc/as.texinfo: Update MIPS documentation.
4395 * config/tc-mips.c: Add define for $zero register.
4396 (md_begin): Add $zero as alias name for $0.
4397 (insn_uses_reg): Use ZERO define.
4398 (mips_ip): Add $zero as alias name for $0.
4399 (mips16_ip): Likewise.
4400 (s_cplocal): Demand empty rest of input line.
4401 (tc_get_register): Likewise. Add support for $kt0, kt1 register
4402 names. Use ZERO define. Fix input_line_pointer progress.
4406 * symbols.c: Replace CONST by const throughout.
4407 (symbol_find_exact): Split out from..
4408 (symbol_find_base): ..here.
4409 * symbols.h: Replace CONST by const throughout.
4410 (symbol_find_exact): Declare.
4411 * config/obj-elf.c: #include "struc-symbol.h".
4412 (elf_frob_file): If group name matches an exported symbol, use that
4413 symbol for the signature and ".group" as the section name.
4417 * config/tc-sh.c (parse_at): @(symbol,pc) is A_DISP_PC again,
4422 * tc-mips.c (mips_after_parse_args): Always set mips_opts.ase_mips3d
4423 and mips_opts.ase_mdmx if they are uninitialized.
4427 * gas/config/tc-hppa.c (pa_ip): Replace "L$0\001" with FAKE_LABEL_NAME.
4428 (hppa_force_relocation): Check if a stub just before the start symbol
4429 of the last call_info is reachable before forcing relocation. Fix
4434 * config/tc-mips.c (mips_after_parse_args): New function.
4435 (md_begin): Move processing of defaults to mips_after_parse_args.
4436 config/tc-mips.h (md_after_parse_args): Define.
4440 * configure.in (sh5*): Set cpu_type to sh64 and endian to big.
4441 (sh5le*, sh64le*): Set cpu_type to sh64 and endian to little.
4442 (sh5*-*-netbsd*, sh64*-*-netbsd*): New targets.
4443 * configure: Regenerate.
4444 * config/tc-sh64.c (sh64_target_format): Add support for NetBSD
4449 * config/tc-sh64.h (MD_PCREL_FROM_SECTION): Undef before redefining.
4453 * config/obj-elf.c (obj_elf_change_section): Set and check elf
4454 linkonce flag. Print all warnings.
4455 (obj_elf_section): Parse ",comdat" for groups.
4456 (elf_frob_file): Set SEC_LINK_ONCE on COMDAT groups. Check
4457 consistency of comdat flag.
4461 * config/tc-alpha.c (alpha_adjust_symtab_relocs): Fix thinko
4462 with LITERALs without sequence numbers.
4466 * config/tc-alpha.c: Move LITUSE constants to "elf/alpha.h".
4467 Rename them LITUSE_ALPHA_*.
4471 * config/tc-h8300.c (get_operand): Allow stm.l and ldm.l insns to
4472 accept parentheses enclosed register lists.
4476 * Makefile.am: Run "make dep-am".
4477 * Makefile.in: Regenerate.
4478 * po/POTFILES.in: Regenerate.
4482 * configure.in: Add i386-*-nto-qnx*.
4483 * configure: Regenerate.
4487 * config/tc-mips.c (mips_ip): Use unsigned long values for
4492 * config/tc-mips.c (s_cpsetup): Fix initialization of
4493 mips_cpreturn_register and mips_cpreturn_offset.
4497 * config/tc-mips.c (s_cpsetup): Fix comment.
4501 * config/tc-mips.c (md_begin): Add $ra as alias name for $31.
4502 (mips_ip): Likewise.
4503 (mips16_ip): Likewise.
4504 (tc_get_register): Likewise.
4509 * config/tc-mips.c (mips_set_options): New "ase_mdmx" member.
4510 (mips_opts): Initialize "ase_mdmx" member.
4511 (file_ase_mdmx): New variable.
4512 (CPU_HAS_MDMX): New macro.
4513 (md_begin): Initialize mips_opts.ase_mdmx and file_ase_mdmx
4514 based on command line options and configuration defaults.
4515 (macro_build): Note in comment that use of MDMX in macros is
4516 not currently allowed.
4517 (validate_mips_insn): Add support for the "O", "Q", "X", "Y", and
4518 "Z" MDMX operand types.
4519 (mips_ip): Accept MDMX instructions if mips_opts.ase_mdmx is set,
4520 and add support for the "O", "Q", "X", "Y", and "Z" MDMX operand
4522 (OPTION_MDMX, OPTION_NO_MDMX, md_longopts, md_parse_option):
4523 Add support for "-mdmx" and "-no-mdmx" options.
4524 (OPTION_ELF_BASE): Move to accomodate new options.
4525 (s_mipsset): Support ".set mdmx" and ".set nomdmx".
4526 (mips_elf_final_processing): Set MDMX ASE ELF header flag if
4527 file_ase_mdmx was set.
4528 * doc/as.texinfo: Document -mdmx and -no-mdmx options.
4529 * doc/c-mips.texi: Likewise, and document ".set mdmx" and ".set
4534 * config/tc-mips.c (OPTION_NO_M7000_HILO_FIX): Rename to
4535 OPTION_MNO_7000_HILO_FIX. Add alternate "mno-fix7000"
4536 command line switch conforming to gcc conventions.
4537 * doc/c-mips.texi: Document -mno-fix7000 instead of no-mfix-7000.
4541 * config/tc-mips.c (macro_build_jalr): New Function.
4542 (md_begin): NewABI uses big GOTs.
4543 (macro_build): Recognize BFD_RELOC_MIPS_GOT_DISP,
4544 BFD_RELOC_MIPS_GOT_PAGE, BFD_RELOC_MIPS_GOT_OFST as valid.
4545 (load_address): Add some NewABI PIC support.
4547 (md_apply_fix): Special handling for BFD_RELOC_MIPS_JALR.
4548 (tc_gen_reloc): Don't encode NewABI vtables in REL relocations.
4552 * config/tc-mips.c (load_address): Use mips_gp_register instead
4553 of hardcoded value. Remove dbl parameter, use HAVE_32BIT_ADDRESSES
4555 (macro): Use mips_gp_register instead of hardcoded value.
4559 * expr.h (operatorT): Add O_md17..O_md32.
4560 * config/tc-alpha.c (O_lituse_tlsgd, O_lituse_tlsldm, O_tlsgd,
4561 O_tlsldm, O_gotdtprel, O_dtprelhi, O_dtprello, O_dtprel, O_gottprel,
4562 O_tprelhi, O_tprello, O_tprel): New.
4563 (USER_RELOC_P, alpha_reloc_op_tag, debug_exp): Include them.
4564 (DUMMY_RELOC_LITUSE_TLSGD, DUMMY_RELOC_LITUSE_TLSLDM): New.
4565 (LITUSE_TLSGD, LITUSE_TLSLDM): New.
4566 (struct alpha_reloc_tag): Add master, saw_tlsgd, saw_tlsld,
4567 saw_lu_tlsgd, saw_lu_tlsldm. Make multi_section_p a bit field.
4568 (md_apply_fix3): Handle TLS relocations.
4569 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
4570 (alpha_adjust_symtab_relocs): Sort LITERAL relocs after the
4571 associated TLS reloc. Check lituse_tls relocs match up.
4572 (emit_insn): Handle TLS relocations.
4575 * doc/c-alpha.texi: Add docs for tls relocations.
4579 * config/tc-mips.c (mips_gprel_offset): New variable.
4580 (s_gpvalue): Use it.
4584 * gas/config/tc-d10v.c (check_resource_conflict): Only check
4585 write-write conflicts.
4586 (md_assemble): Reformat introductory comment.
4587 (parallel_ok): Prevent packing only if the first
4588 instruction cannot be packed.
4593 * config/tc-d10v.c (build_insn): Check for immediates.
4597 * config/tc-mips.c: Replace GP in comments by $gp.
4598 (mips_big_got): Initialize.
4599 (mips_trap): Initialize.
4600 (load_address): Use mips_gp_register instead of hardcoded value.
4601 Remove dbl parameter, use HAVE_32BIT_ADDRESSES instead.
4602 (macro): Use mips_gp_register instead of hardcoded value.
4603 (macro2): Change load_address calls.
4604 (md_pcrel_from): Comment formatting.
4605 (s_cpload): Use mips_gp_register instead of hardcoded value.
4606 (s_cprestore): Likewise. Comment formatting.
4607 (s_gpword): Fix data type.
4608 (s_cpadd): Use mips_gp_register instead of hardcoded value.
4609 (nopic_need_relax): Replace GP in comments by $gp.
4610 (mips_elf_final_processing): Better comment.
4614 * configure.in: Add DLX configuraton
4615 * Makefile.am: Add DLX configuraton
4616 * configure: Regenerate.
4617 * Makefile.in: Regenerate.
4618 * config/tc-dlx.c: New file.
4619 * config/tc-dlx.h: New files.
4620 * NEWS: Mention new support.
4624 * config/obj-coff.c (write_object_file): Add missing semicolon.
4628 * config/tc-mips.c (mips_emit_delays): Replace magic constant for RA
4629 by the define. Remove superfluous check of mips_opts.mips16.
4630 (append_insn): Likewise. Canonicalize variable increments.
4631 (macro_build): Canonicalize variable increments.
4632 (macro_build_lui): Likewise.
4633 (load_register): Likewise.
4634 (load_address): Move pointer initialization.
4635 (macro): Move pointer to a more local scope. Canonicalize variable
4636 increments. Better comments. Replace magic constant for RA by the
4638 (macro2): Replace magic constant for RA by the define. Canonicalize
4639 variable increments.
4640 (mips_ip): Canonicalize variable increments.
4641 (mips16_ip): Replace magic constant for RA by the define.
4642 (my_getSmallParser): Canonicalize variable increments/decrements.
4643 (my_getPercentOp): Likewise.
4644 (my_getSmallExpression): Likewise.
4645 (s_align): Likewise.
4646 (s_mipsset): Likewise.
4647 (s_cpsetup): Likewise.
4648 (s_insn): Remove superfluous check of mips_opts.mips16.
4649 (s_mips_stab): Likewise.
4650 (mips_handle_align): Canonicalize variable increments.
4651 (s_mips_ent): Likewise.
4652 (s_mips_end): Add comment.
4656 * configure.in: Use ns32k-*-netbsd* instead of ns32k-pc532-netbsd*.
4657 * configure: Regenerate.
4661 * Makefile.am (OBJS): Depend on ansidecl.h and fopen-same.h.
4662 * Makefile.in: Regenerate.
4663 * dep-in.sed: Reorder to match OBJS in Makefile.am.
4664 * configure.in (ALL_OBJ_DEPS): Add symcat.h when need_bfd.
4665 * configure: Regenerate.
4666 * as.h: Use #include "" instead of <> for local header files.
4667 * flonum-konst.c: Likewise.
4668 * flonum-mult.c: Likewise.
4670 * listing.c: Likewise.
4671 * config/tc-ia64.h: Likewise.
4672 * config/tc-v850.h: Likewise.
4676 * configure.in: Add a target for i386-netbsdpe.
4677 * configure: Regenerate.
4681 * config/obj-elf.c (elf_common): Renamed from obj_elf_common.
4682 (obj_elf_common): Call elf_common.
4683 (obj_elf_tls_common): New function.
4684 (elf_pseudo_tab): Support .tls_common.
4685 (special_sections): Add .tdata and .tbss.
4686 (obj_elf_change_section): Set SEC_THREAD_LOCAL for SHF_TLS
4688 (obj_elf_parse_section_letters): Support T in section flags (SHF_TLS).
4689 (obj_elf_parse_section_letters): Include T in error message.
4690 * config/tc-ppc.c (ppc_section_letter): Likewise.
4691 * config/tc-alpha.c (alpha_elf_section_letter): Likewise.
4692 (tc_gen_reloc): Handle SEC_THREAD_LOCAL the same way as
4694 * config/tc-sparc.c (md_apply_fix3): Likewise.
4695 * config/tc-i386.c (tc_i386_fix_adjustable): Add TLS relocs.
4696 Define them if not BFD_ASSEMBLER.
4697 (lex_got): Support @TLSGD, @TLSLDM, @GOTTPOFF, @TPOFF, @DTPOFF
4699 (md_apply_fix3): Add TLS relocs.
4700 * config/tc-ia64.c (enum reloc_func): Add FUNC_DTP_MODULE,
4701 FUNC_DTP_RELATIVE, FUNC_TP_RELATIVE, FUNC_LT_DTP_MODULE,
4702 FUNC_LT_DTP_RELATIVE, FUNC_LT_TP_RELATIVE.
4703 (pseudo_func): Support @dtpmod(), @dtprel() and @tprel().
4704 (ia64_elf_section_letter): Include T in error message.
4705 (md_begin): Support TLS operators.
4706 (md_operand): Likewise.
4707 (ia64_gen_real_reloc_type): Support TLS relocs.
4708 * write.c (adjust_reloc_syms): Don't change symbols in
4709 SEC_THREAD_LOCAL sections to STT_SECTION + addend.
4713 * config/tc-arm.c (md_apply_fix3): For the Thumb BLX reloc
4714 round the relocation up rather than down.
4718 * config/obj-coff.c (obj_coff_section): Silently ignore an 'a'
4720 * doc/as.texinfo: Document that the COFF version of .section
4721 ignores the 'a' flag.
4725 * config/tc-alpha.c (assemble_tokens): Protect use of
4726 ALPHA_RELOC_TABLE with #ifdef RELOC_OP_P.
4728 * write.c (size_seg): Check adjustment to last frag.
4729 (SUB_SEGMENT_ALIGN): If HANDLE_ALIGN defined, pad out last frag to
4731 * config/obj-coff.c (SUB_SEGMENT_ALIGN): Likewise.
4732 * config/obj-ieee.c (SUB_SEGMENT_ALIGN): Likewise.
4733 (write_object_file): Invoke md_do_align if available, and use
4734 frag_align_code on text sections.
4735 * config/obj-vms.h (SUB_SEGMENT_ALIGN): Now two args.
4736 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Likewise.
4737 * config/tc-ppc.h (SUB_SEGMENT_ALIGN): Likewise.
4738 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Likewise.
4739 * config/tc-i386.h (SUB_SEGMENT_ALIGN): Likewise. Define for
4744 * dwarf2dbg.c (dwarf2_directive_loc): Call listing_source_file
4749 * config/tc-arm.c (arm_s_section): Enable for COFF builds as well
4754 * dwarf2dbg.c (dwarf2_emit_insn): Emit only one line symbol
4755 for one .loc for compiler.
4759 * config/tc-mips.c (macro): Relax warning, it's toot strict for
4764 * config/tc-mips.c (macro2): Add 64 bit drol, dror macros.
4765 Optimize the rotate by zero case.
4769 * configure.in: Remove accidental enabling of bfd_gas=yes for
4771 * configure: Regenerate.
4775 * app.c: Fix formatting.
4777 * ehopt.c: Likewise.
4779 * input-file.c: Likewise.
4780 * listing.c: Likewise.
4781 * macro.h: Likewise.
4782 * stabs.c: Likewise.
4783 * symbols.c: Likewise.
4787 * config/obj-generic.c: Delete file.
4788 * config/obj-generic.h: Likewise.
4792 * config/tc-avr.c (mcu_types): Update for new devices.
4796 * config/tc-mips.c (macro): Warn about wrong la/dla use.
4800 * config/tc_mips.c (s_cpsetup): Fix completely bogus code which had
4801 worked sometimes by accident. Fix copy&paste comment.
4805 * config/tc-mips.c (md_begin): Fix .reginfo and .MIPS.option section
4806 alignment for NewABI. Let n32 use .reginfo. Remove useless casts.
4807 (mips_elf_final_processing): Let n32 use .reginfo.
4811 * config/tc-mips.c (append_insn): Fix too small range of variable.
4815 * config/tc-arm.c (arm_cleanup): Remove redundant call to
4816 listing_prev_line().
4820 * config/tc-arm.c (md_assemble): Remove redundant call to
4823 * dwarf2dbg.c (dwarf2_emit_insn): Do not reset
4826 * stabs.c (s_stab_generic): Fix grammatical error in warning
4831 * write.c (subsegs_finish): Don't specially align last subseg.
4835 * stabs.c (s_stab_generic): Warn about a description field that is
4838 * config/obj-coff.c: Fix compile time warnings when compiling
4839 without BFD_ASSEMBLER defined.
4842 * config/tc-sh.c (md_pcrel_from): Define for use with sh-hms
4844 (md_pcrel_from_section): Use md_pcrel_from().
4848 * dwarf2dbg.c (dwarf2_emit_insn): Use the 'current' struct filled
4849 by dwarf2_directive_loc, instead of calling dwarf2_where.
4853 * config/obj-coff.h: Fix formatting.
4854 * config/tc-mcore.c: Likewise.
4855 * config/tc-mn10300.c: Likewise.
4856 * config/tc-openrisc.c: Likewise.
4857 * config/tc-or32.c: Likewise.
4858 * config/tc-pdp11.c: Likewise.
4859 * config/tc-ppc.c: Likewise.
4860 * config/tc-ppc.h: Likewise.
4861 * config/tc-sh64.c: Likewise.
4862 * config/tc-sh.c: Likewise.
4863 * config/tc-tic54x.c: Likewise.
4864 * config/tc-xstormy16.c: Likewise.
4865 * config/tc-xstormy16.h: Likewise.
4869 * config/obj-coff.c: Fix formatting.
4870 * config/obj-elf.c: Likewise.
4871 * config/tc-alpha.c: Likewise.
4872 * config/tc-arm.c: Likewise.
4873 * config/tc-d10v.c: Likewise.
4874 * config/tc-d30v.c: Likewise.
4875 * config/tc-h8300.c: Likewise.
4876 * config/tc-hppa.c: Likewise.
4880 * config/tc-i386.c (md_estimate_size_before_relax) Don't lose
4881 reloc when no_cond_jump_promotion.
4885 * config/tc-i960.c (md_estimate_size_before_relax): Return size of
4886 current variable part of frag.
4890 * config/tc-mmix.c: Fix formatting.
4891 * config/tc-mmix.h: Likewise.
4895 * configure: Regenerate.
4899 * config/tc-m68k.c: Fix formatting.
4903 * Makefile.am: Honour DESTDIR.
4904 * Makefile.in: Regenerate.
4908 * config/tc-ia64.c: Fix formatting.
4909 * config/tc-ia64.h: Likewise.
4913 * config/tc-mips.c: Fix formatting.
4914 * config/tc-s390.c: Likewise.
4915 * config/tc-s390.h: Likewise.
4919 * config/tc-s390.c (md_gather_operands): Emit dwarf2 line-number
4920 information for instructions.
4924 * as.h: Fix formatting.
4927 * dwarf2dbg.c: Likewise.
4928 * frags.h: Likewise.
4930 * macro.c: Likewise.
4932 * stabs.c: Likewise.
4933 * symbols.c: Likewise.
4937 * app.c (mri_pseudo): Only declare for TC_M68K.
4939 * config/tc-ppc.c (mapping): Map sectoff to BFD_RELOC_16_BASEREL.
4940 (ppc_elf_validate_fix): Replace BFD_RELOC_32_BASEREL with
4941 BFD_RELOC_16_BASEREL.
4942 (md_assemble): Likewise.
4943 (md_apply_fix3): Likewise.
4947 * config/tc-arm.c (thumb_add_sub): Do not convert a subtract of
4948 zero into an add of zero - it is not the same.
4952 * tc-sh.c (get_specific): Generate warning if the same
4953 destination register is used in parallel instructions.
4957 * config/tc-i386.c (extra_symbol_chars): Add '[' to the list.
4961 * write.c (cvt_frag_to_fill): Set fr_offset to zero on .org
4962 backwards to prevent cascading errors.
4966 * configure.in: Add support for powerpc-*-windiss.
4967 * configure: Regenerated.
4971 * config/tc-s390.c (md_parse_option): Formatting.
4973 * config/tc-i386.c: Formatting fixes, add missing space in error
4978 * config/tc-z8k.c (build_bytes): Add support for new cases:
4979 CLASS_IGNORE and ARG_NIM4.
4980 (md_assemble): Prevent destruction of input_line_pointer if
4981 get_operands returns failure.
4985 * config/tc-mips.c (macro_build): Do _not_ allow MIPS-3D
4986 instructions to be generated by macros.
4990 * config/tc-i386.c (output_jump, output_disp)
4991 (md_estimate_size_before_relax): Don't set fx_pcrel_adjust any
4993 (md_apply_fix3): Remember addend value for rela relocations.
4994 (tc_gen_reloc): Correctly compute pc-relative relocation addend.
4998 * config/tc-mips.c (macro_build): Add close-parenthesis missing
4999 from previous change.
5003 * config/tc-mips.c (macro_build): Add warning if macro instructions
5004 are expanded into a branch delay slot.
5008 * dwarf2dbg.c (dwarf2_gen_line_info): Do emit duplicate line
5009 numbers, gdb relies on them to detect the start of the prologue.
5013 * config/tc-s390.c (tc_s390_fix_adjustable): Prevent adjustments to
5014 symbols in merge sections.
5018 * as.c (main): Don't reference _bfd_chunksize.
5022 * config/tc-d10v.c (d10v_fix_adjustable): Prevent adjustments to
5023 symbols in merge sections.
5027 * doc/invoke.texi (TC_LARGEST_EXPONENT_IS_NORMAL): Document.
5028 * config/atof-ieee.c (TC_LARGEST_EXPONENT_IS_NORMAL): Add an
5029 argument for the precision.
5030 (gen_to_words): Update accordingly.
5034 * as.c (parse_args <OPTION_VERSION>): Use VERSION is
5035 BFD_VERSION_STRING unavailable.
5036 * config/tc-i386.c (INLINE): Define (for non-BFD assembler).
5040 * config/tc-sh.h (TC_FIX_ADJUSTABLE): Disable adjusting if
5041 symbol_used_in_reloc_p is true.
5043 * config/tc-sh.c (md_apply_fix3): Don't zero relocations on big
5048 * dep-in.sed: Cope with absolute paths.
5049 * Makefile.am (dep.sed): Subst TOPDIR.
5051 * Makefile.in: Regenerate.
5055 * config/tc-mips.c (mips16_macro_build): Cast type mismatch.
5056 (mips_ip): Remove unused variable.
5057 (md_apply_fix3): Cast signed/unsignes mismatches. Replace
5058 unsigned char with bfd_byte.
5059 (s_file): Remove unused variable.
5060 (s_mips_ent): Likewise.
5064 * config/tc-d10v.c (d10v_insert_operand): Fix warning in as_bad_where.
5066 (find_opcode): Fix warning in as_warn.
5067 * config/tc-d10v.h: Update Copyright.
5073 * config/tc-d10v.c (check_resource_conflict): New function to
5074 check for resource conflicts.
5077 * config/tc-d10v.c (build_insn): Check for unresolved imm4 or
5079 * config/tc-d10v.c (find_opcode): Emit a warning if one of the
5080 reserved control registers is used.
5081 * config/tc-d10v.c (build_insn): Check for unresolved imm4 or
5085 * tc-d10v.c (parallel_ok): Prevent packing only if the first
5086 instruction cannot be packed.
5089 * gas/config/tc-d10v.c (check_resource_conflict): Only check
5090 write-write conflicts.
5091 (md_assemble): Reformat introductory comment.
5092 * opcodes/d10v-opc.c (d10v_opcodes): `btsti' does not modify its
5097 * symbols.c (resolve_symbol_value <O_uminus, O_bit_not,
5098 O_logical_not>): Derive final_seg from add_symbol.
5099 <O_multiply..O_logical_or>: More final_seg twiddles.
5103 * config/tc-ia64.c (ia64_cons_fix_new): Handle 8 byte iplt reloc
5108 * config/tc-i386.c (output_jump): Set fx_pcrel_adjust to size of
5109 field for pc-relative fixups.
5110 (output_disp): Likewise.
5111 (md_estimate_size_before_relax): Likewise.
5112 (tc_gen_reloc): Subtract fx_pcrel_adjust instead of fx_size for
5113 pc-relative fixups in 64bit mode.
5117 * config/te-aix5.h: Typo fix.
5121 * Makefile.am: Run "make dep-am".
5122 * Makefile.in: Regenerate.
5123 * doc/Makefile.in: Regenerate.
5127 * config/tc-arm.c (vfp_dp_reg_required_here): Fix typo
5128 (vfp_sp_reg_pos -> vfp_dp_reg_pos).
5132 * config/tc-mips.c (md_estimate_size_before_relax): Do not modify
5133 the EXTENDED bit here; report the estimate according to the
5138 * configure.in: Add AIX 5 support.
5139 * config/tc-ppc.c (ppc_target_format): Add AIX 5 64 bit target.
5140 * config/te-aix5.h: New file. AIX 5 support.
5141 * configure: Regenerate.
5145 * po/fr.po: Updated version.
5149 * doc/c-mips.texi (Machine Dependencies): Add new node.
5153 * config/tc-mips.c (mips_set_options): New "ase_mips3d" member.
5154 (mips_opts): Initialize "ase_mips3d" member.
5155 (file_ase_mips3d): New variable.
5156 (CPU_HAS_MIPS3D): New macro.
5157 (md_begin): Initialize mips_opts.ase_mips3d and file_ase_mips3d
5158 based on command line options and configuration defaults.
5159 (macro_build, mips_ip): Accept MIPS-3D instructions if
5160 mips_opts.ase_mips3d is set.
5161 (OPTION_MIPS3D, OPTION_NO_MIPS3D, md_longopts, md_parse_option):
5162 Add support for "-mips3d" and "-no-mips3d" options.
5163 (OPTION_ELF_BASE): Move to accomodate new options.
5164 (s_mipsset): Support ".set mips3d" and ".set nomips3d".
5165 (mips_elf_final_processing): Add a comment indicating that a
5166 MIPS-3D ASE ELF header flag should be set, when one exists.
5167 * doc/as.texinfo: Document -mips3d and -no-mips3d options.
5168 * doc/c-mips.texi: Likewise, and document ".set mips3d" and ".set
5169 nomips3d" directives.
5173 * config/tc-mmix.c (md_estimate_size_before_relax): Don't consider
5174 a weak symbol in same section to be within reach.
5178 * po/fr.po: Updated version.
5182 * config/tc-ia64.c (fixup_unw_records): Clear region when seeing a
5183 body record so that an error is given for misplaced .save
5188 * config/tc-i386.h (REX_OPCODE): Define.
5189 (REX_MODE64, REX_EXTX, REX_EXTY, REX_EXTZ): Define.
5190 (rex_byte): typedef to int.
5191 * config/tc-i386.c: Group prototypes and vars together.
5192 Formatting fixes. Remove occurrences of "register" keyword.
5195 (mode_from_disp_size): Add INLINE keyword to prototype.
5196 (fits_in_signed_byte): Likewise.
5197 (fits_in_unsigned_byte): Likewise.
5198 (fits_in_unsigned_word): Likewise.
5199 (fits_in_signed_word): Likewise.
5200 (fits_in_unsigned_long): Likewise.
5201 (fits_in_signed_long): Likewise.
5202 (type_names): Constify.
5203 (intel_float_operand): Constify param.
5204 (add_prefix): Use REX_OPCODE.
5205 (md_assemble): Likewise. Modify for changed rex_byte.
5206 (parse_insn): Split out of md_assemble.
5207 (parse_operands): Likewise.
5208 (swap_operands): Likewise.
5209 (optimize_imm): Likewise.
5210 (optimize_disp): Likewise.
5211 (match_template): Likewise.
5212 (check_string): Likewise.
5213 (process_suffix): Likewise.
5214 (check_byte_reg): Likewise.
5215 (check_long_reg): Likewise.
5216 (check_qword_reg): Likewise.
5217 (check_word_reg): Likewise.
5218 (finalize_imm): Likewise.
5219 (process_operands): Likewise.
5220 (build_modrm_byte): Likewise.
5221 (output_insn): Likewise.
5222 (output_branch): Likewise.
5223 (output_jump): Likewise.
5224 (output_interseg_jump): Likewise.
5225 (output_disp): Likewise.
5226 (output_imm): Likewise.
5230 * doc/as.texinfo: Wrap @menu in @ifnottex, not @ifinfo.
5234 * config/tc-i386.c (tc_gen_reloc): Don't attempt to handle 8 byte
5235 relocs except when BFD64.
5237 * write.c (number_to_chars_bigendian): Don't abort when N is
5238 larger than sizeof (VAL).
5239 (number_to_chars_littleendian): Likewise.
5243 * config/tc-hppa.c (md_apply_fix3): Add cast.
5244 (hppa_fix_adjustable): Adjust list of selectors using e_lrsel and
5249 * tc-pdp11.c: Use VAX float format support for PDP-11 target.
5250 (parse_ac5): New function for parsing float regs in float operand.
5251 (parse_expression): Remove attempt to make literals be octal.
5252 (parse_op_no_deferred): Support float literals.
5253 (parse_op): Reject attempts to refer to float regs.
5254 (parse_fop): New function, like parse_op but for float operand.
5255 (md_assemble): Add cases to parse float operands. Also fix
5256 IMM3, IMM6, IMM8 cases to pick up the operand from the right spot.
5260 * config/obj-elf.c (special_section): Add .init_array,
5261 .fini_array and .preinit_array.
5263 * config/tc-ia64.h (ELF_TC_SPECIAL_SECTIONS): Remove
5264 .init_array and .fini_array.
5268 * config/obj-elf.c (elf_copy_symbol_attributes): Don't copy
5270 (obj_frob_symbol): Copy visibility.
5274 * config/tc-alpha.c (s_alpha_text): Use obj_elf_text for OBJ_ELF, not
5276 (s_alpha_data): Use obj_elf_data for OBJ_ELF, not s_data.
5280 * po/es.po: Updated.
5284 * config/tc-mips.c (mips_need_elf_addend_fixup): For embedded-PIC
5285 only, undo the changes made on 2001-06-08, with the
5286 effect being that common or extern symbols are
5287 adjusted for embedded-PIC, but weak symbols are not.
5288 (md_estimate_size_before_relax: Likewise, with the effect
5289 that extern symbols are treated the same as weak symbols
5290 only if not embedded-PIC.
5291 (mips_fix_adjustable) Likewise, with the effect that
5292 weak or extern symbols are not adjusted for embedded-PIC.
5293 (md_apply_fix3): Tweak so that the case where value is zero
5294 is handled more correctly for embedded-PIC code.
5298 * doc/as.texinfo (Overview): Add missing @ifset IA64
5300 * configure.in (LINGUAS): Add es.po.
5301 * configure: Regenerate.
5302 * po/es.po: New file.
5306 * config/tc-mips.c (set_at): Fix handling of 64bit register loads.
5307 (macro): Likewise. Fix la/dla address expansions for EMBEDDED_PIC
5308 and NO_PIC cases. Code cleanup.
5309 (macro2): Fix handling of 64bit register loads.
5313 * doc/as.texinfo: Add entry for IA64.
5314 * doc/c-ia64.texi: New file.
5318 * config/tc-hppa.c: Update copyright date.
5320 * doc/c-ppc.texi (PowerPC-Opts): Add -mpower4, -maltivec and -m7400
5321 Remove references to chip manufacturers.
5322 * config/tc-ppc.c (md_parse_option): Handle -mpower4 option.
5324 (md_show_usage): Remove references to chip manufacturers. Mention
5326 (md_begin): Test power4 opcode flag bits.
5330 * config/tc-ia64.c (dot_restore): Issue error message of epilogue
5331 count exceeds prologue count.
5332 (md_show_usage): Describe -mconstant-gp and -mauto-pic.
5333 (unwind.label_prologue_count): New member.
5337 (get_saved_prologue_count): New function.
5338 (save_prologue_count): New function.
5339 (free_saved_prologue_count): New function.
5340 (dot_label_state): Record state label by calling save_prologue_count().
5341 (dot_copy_state): Restore prologue count by calling
5342 get_saved_prologue_count().
5343 (generate_unwind_image): Free up list of saved prologue
5344 counts by calling free_saved_prologue_counts().
5348 * config/tc-tic54x.c: Add missing prototypes and remove ANSI style
5349 function declarations.
5353 * NEWS: Note that GASP is now deprecated.
5354 * Makefile.am: Do not build gasp-new by default.
5355 * Makefile.in: Regenerate.
5356 * doc/Makefile.am: Do not install gasp.info.
5357 * doc/Makefile.in: Regenerate.
5358 * gas/gasp.texi: Note that gasp is now deprecated.
5362 * NEWS: Mark 2.12 branch.
5366 * config/tc-xstormy16.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
5370 * config/tc-mips.c (md_parse_option): Complain about invalid -mabi
5375 * config/tc-s390.c (md_parse_option): Add switches -m31 and -m64.
5376 Make bit size independent of architecture switch.
5377 (md_begin): Add warning for -m64 with -Aesa.
5378 (s390_md_end): Use renamed architecture defines.
5382 * config/obj-coff.h: Check !target_big_endian, not shl, for coff-sh.
5386 * doc/as.texinfo (Machine Dependencies): Fix typo: MMIX used
5391 * config/tc-mips.c (md_estimate_size_before_relax): Really
5392 make sure we treat weak like extern only for ELF. (Fixes
5393 patch from 2001-07-25.)
5397 * doc/as.texinfo: Add duplicate directory entry so that "info gas"
5402 * NEWS: Mention support for MMIX.
5406 * config/tc-mips.c (mips_need_elf_addend_fixup): Restructure into
5407 a sequence of indpendent 'if' statements for easier debugging
5408 and future modification.
5412 * config/tc-m68k.c (md_show_usage): No longer display a
5413 hard-coded "68020" for the default CPU, instead display the
5414 canonical name of the true, configured default CPU.
5415 (m68k_elf_final_processing): Mark objects for sub-68020
5416 CPUs with the new EF_M68000 flag.
5420 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust
5421 pc-relative relocations to merge sections in 64-bit mode.
5425 * NEWS: Document floating point number handling in gasp.
5426 * gasp.c: Include <assert.h> and "xregex.h".
5427 (is_flonum): New function.
5428 (chew_flownum): Likewise.
5429 (change_base): Consume flonums from the input, where possible.
5430 * doc/gasp.texi (Constants): Document floating point numbers.
5434 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust final types
5435 that implicitly use LR and RR selectors.
5439 * config/tc-mn10300.c (other_registers): Added `epsw'. Mark `pc'
5440 and `epsw' as available on AM33 and above only.
5441 (other_register_name): Add logic to handle machine type encoded in
5446 * config/tc-ppc.c (md_apply_fix3): Check for insert fop invalid for
5451 * config/tc-sparc.c (U0x80000000, U0xffffffff): New constants.
5456 * i386.c (md_assemble): Support 32bit address prefix.
5457 (i386_displacement): Likewise.
5458 (i386_index_check): Accept 32bit addressing in 64bit mode.
5462 * config/tc-sh.c (dot): Removed unused function.
5466 * Makefile.am: "make dep-am".
5467 * Makefile.in: Regenerate.
5468 * aclocal.m4: Regenerate.
5469 * config.in: Regenerate.
5470 * configure: Regenerate.
5471 * doc/Makefile.in: Regenerate.
5475 * doc/c-alpha.texi: New file.
5476 * doc/Makefile.am (CPU_DOCS): Add it.
5477 * doc/all.texi, doc/as.texinfo: Add hooks for Alpha.
5481 * config/tc-alpha.c (O_samegp): New.
5482 (USER_RELOC_P): Include it.
5483 (alpha_reloc_op_tag, debug_exp, find_macro_match): Add it.
5484 (md_apply_fix3): Handle BFD_RELOC_ALPHA_BRSGP.
5485 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
5486 (alpha_validate_fix): New.
5487 * config/tc-alpha.h (TC_VALIDATE_FIX): New.
5491 * doc/c-cris.texi: New.
5492 * doc/all.texi: @set CRIS.
5493 * doc/as.texinfo: Ditto. Add CRIS gas manpage option overview.
5494 Include c-cris.texi.
5495 * doc/Makefile.am (CPU_DOCS): Add c-cris.texi
5496 * doc/Makefile.in: Regenerate.
5500 * config/tc-mips.c (IS_SEXT_32BIT_NUM): New macro to
5501 determine if a number is a sign-extended 32-bit number.
5502 (load_register): Use IS_SEXT_32BIT_NUM.
5503 (macro): Check if load/store macro handling is using a
5504 constant 32-bit address on 64-bit address systems, and if
5505 so optimize the generation of that address.
5509 * config/tc-alpha.c (alpha_force_relocation): Don't assert that
5510 we've eliminated all foreign relocation types yet.
5511 (alpha_fix_adjustable): Likewise.
5515 Contribute sh64-elf.
5518 * doc/c-sh64.texi: Fix citation of SH64 architecture manual.
5520 * config/tc-sh.c (md_relax_table): Added default sizes for
5521 non-PC-relative UNDEF_MOVI, and relaxation sequences for
5522 MOVI_16, MOVI_32 and MOVI_48.
5523 * config/tc-sh64.c (shmedia_md_apply_fix3): Fix warning.
5524 (shmedia_md_convert_frag): Handle non-PC-relative UNDEF_MOVI
5526 (shmedia_md_estimate_size_before_relax): Remove redundant
5527 blocks. Set fragP->fr_var even if relaxation type unchanged.
5528 Retain UNDEF_MOVI until expression decays to number.
5530 * config/tc-sh64.c (shmedia_init_reloc): Handle new SHmedia PIC
5531 relocation types. Take fixP->fx_addnumber into account too.
5532 (shmedia_md_apply_fix): Likewise.
5533 (shmedia_md_convert_frag): Likewise.
5534 (shmedia_build_Mytes): Likewise.
5535 (sh64_consume_datalabel): Complain about nested datalabel.
5536 Support PIC relocs. Call sh_parse_name.
5537 * config/tc-sh64.h (TC_RELOC_RTSYM_LOC_FIXUP): Extend definition
5538 in tc-sh.h to SHmedia reloc types.
5539 * config/tc-sh.c (SH64PCRELPLT, MOVI_PLT, MOVI_GOTOFF,
5540 MOVI_GOTPC): New relaxation constants.
5541 (md_relax_table): Introduce relaxation directives for PIC-related
5543 (sh_PIC_related_p): Handle datalabel.
5544 (sh_check_fixup): Choose SH5 PIC relocations.
5545 (sh_cons_fix_new): Added BDF_RELOC_64.
5546 (md_apply_fix3, sh_parse_name): Handle GOTPLT.
5548 * config/tc-sh64.c (sh64_max_mem_for_rs_align_code): If the
5549 current ISA is SHmedia, get 7 bytes.
5551 * config/tc-sh.c (md_apply_fix3): Treat shmedia_md_apply_fix3 as a
5553 * config/tc-sh64.c (shmedia_apply_fix): Rename to
5554 shmedia_apply_fix3 and make void.
5556 * config/tc-sh64.c (s_sh64_abi): Remove unused arguments passed to
5559 * config/tc-sh64.h (md_parse_name): Take &c as argument.
5561 * doc/Makefile.am (CPU_DOCS): Added c-sh64.texi
5562 * doc/Makefile.in(CPU_DOCS): Ditto.
5563 * doc/c-sh64.texi: New file.
5564 * doc/as.texinfo: Add SH64 support.
5566 * config/tc-sh64.c (shmedia_get_operands): Rename A_RESV_Fx to
5567 A_REUSE_PREV so that its purpose is more obvious.
5568 (shmedia_build_Mytes): Ditto.
5570 * config/tc-sh64.c (sh64_vtable_entry): New, strip datalabels
5572 (sh64_vtable_inherit): Ditto.
5573 (strip_datalabels): New, strip "datalabel" from given line.
5574 * config/tc-sh.c (md_pseudo_table): Add sh64-specific vtable
5577 * config/tc-sh64.c (shmedia_md_assemble): Move dwarf2_emit_insn
5579 (shmedia_build_Mytes): ... to here.
5581 * config/tc-sh.c: Remove sh64-specific uaquad now that there
5584 * config/tc-sh64.h (DWARF2_LINE_MIN_INSN_LENGTH): Override.
5585 * config/tc-sh64.c (shmedia_md_assemble): Offset recorded insn
5586 address by one in call to dwarf2_emit_insn.
5588 Implement ".abi" pseudo and correct .cranges descriptors. Correct
5589 alignment handling broken by imported changes.
5590 * config/tc-sh64.h (HANDLE_ALIGN): Override definition in tc-sh.h.
5591 (sh64_handle_align): Declare.
5592 (MAX_MEM_FOR_RS_ALIGN_CODE): Override definition in tc-sh.h.
5593 (sh64_max_mem_for_rs_align_code): Declare.
5594 (enum sh64_isa_values): Moved here from tc-sh64.c.
5595 (md_do_align): Define.
5596 (sh64_do_align): Declare.
5597 (struct sh64_tc_frag_data): New.
5598 (TC_FRAG_TYPE): Change to struct sh64_tc_frag_data. Users
5600 (TC_FRAG_INIT): Change to set new datatype.
5601 (struct sh64_segment_info_type): Rename member
5602 last_flushed_location to last_contents_mark. All users changed.
5603 (md_elf_section_change_hook, TC_CONS_FIX_NEW): Do not define.
5604 (shmedia_elf_new_section, sh64_tc_cons_fix_new): Do not prototype.
5605 * config/tc-sh.c (md_pseudo_table): Add ".abi".
5606 (sh_elf_cons) [HAVE_SH64]: Call sh64_update_contents_mark instead
5607 of unsetting seen_insn.
5608 (md_assemble) [HAVE_SH64] <before new SHcompact sequence>: Also
5609 call sh64_update_contents_mark.
5610 (sh_handle_align): Remove HAVE_SH64-conditioned code.
5611 * config/tc-sh64.c (sh64_isa_mode): Correct type from boolean to
5612 enum sh64_isa_values.
5613 (sh64_set_contents_type): Drop segT parameter. All callers changed.
5614 (emitting_crange): Boolean guard moved to file scope from function
5615 scope in sh64_set_contents_type.
5617 (sh64_update_contents_mark): New; most split out from
5618 sh64_flush_pending_output.
5619 (shmedia_md_end): Call sh64_update_contents_mark. Set
5620 sh64_isa_mode to sh64_isa_sh5_guard unless sh64_isa_unspecified.
5621 (sh64_do_align): New function.
5622 (sh64_max_mem_for_rs_align_code): New function.
5623 (sh64_handle_align): Rename from shmedia_do_align. Make
5624 non-static. Add head comment. Emit zero bytes for n bytes modulo
5625 four. Change return-type to void.
5626 (shmedia_elf_new_section): Remove.
5627 (shmedia_md_assemble): Call sh64_update_contents_mark.
5628 (s_sh64_mode): Ditto. Do not call md_flush_pending_output. Make
5629 new frag. Call sh64_update_contents_mark after making the new
5631 (sh64_flush_pending_output): Just call sh64_update_contents_mark
5632 and sh_flush_pending_output.
5633 (sh64_flag_output): Also call md_flush_pending_output, but add
5634 condition on not emitting_crange.
5635 (sh64_tc_cons_fix_new): Remove.
5637 * config/tc-sh64.c (shmedia_do_align): Fix to work with new
5638 alignment handling scheme imported from sourceware.
5640 * config/tc-sh64.h (TARGET_FORMAT): Define.
5641 (sh64_target_format): Prototype.
5642 * config/tc-sh64.c (sh64_target_mach): New function.
5644 * config/tc-sh64.c (shmedia_md_end): When equating a symbol, use
5645 zero_address_frag instead of copying the frag of the symbol.
5646 (shmedia_frob_file_before_adjust): Ditto.
5647 (shmedia_md_apply_fix) <case BFD_RELOC_SH_IMM_MEDLOW16>: Cast mask
5648 to valueT to remove signedness.
5649 (shmedia_md_convert_frag): Add parameter final. Rename parameter
5650 headers to output_bfd. Do not evaluate symbols if final is false;
5652 (shmedia_md_estimate_size_before_relax) <case C (MOVI_IMM_32,
5653 UNDEF_MOVI) et al>: If symbol cannot be modified to be PC-relative
5654 to the current frag, call shmedia_md_convert_frag to emit fixups
5655 and make frag_wane neutralize the frag. Update comments.
5656 * config/tc-sh.c (md_convert_frag): Change caller of
5657 shmedia_md_convert_frag.
5659 * config/tc-sh64.h: Tweak comments and correct formatting.
5660 * config/tc-sh64.c: Ditto.
5661 (shmedia_md_convert_frag) <PT/PTA/PTB 32, 48 and 64 bit
5662 expansion, MOVI pcrel expansion>: Fix thinko calculating offset
5663 for the no-relocation case.
5664 (shmedia_check_limits): Fix range check being off-by-one for PTA.
5665 * config/tc-sh.c: Ditto. Add proper comments to #ifdef/#ifndef
5667 (SH64PCREL16_F): Increment for proper max-PTA handling. Update
5669 (SH64PCREL16_M, MOVI_16_M): Correct range thinko.
5670 (SH64PCREL48_M, MOVI_48_M): Similar; don't count in length of
5672 (SH64PCREL32_M, MOVI_32_M): Ditto; handle overflowing expression.
5675 * config/tc-sh64.c (shmedia_md_apply_fix) <second switch, case
5676 BFD_RELOC_SH_PT_16>: Set lowest bit in field to be relocated to 1.
5677 (shmedia_md_convert_frag) <case C (SH64PCREL16_32, SH64PCREL16) et
5678 al>: Set lowest bit of field to relocate to 1 and rest to empty,
5679 if reloc is emitted.
5681 New options plus bugfixes.
5682 * config/tc-sh.c (md_longopts): New options "-no-expand" and
5684 (md_parse_option): Handle new options.
5685 (md_show_usage): Add blurb for new options.
5686 * config/tc-sh64.c (SHMEDIA_BFD_RELOC_PT): New macro.
5687 (sh64_expand, sh64_pt32): New variables.
5688 (shmedia_init_reloc): Handle BFD_RELOC_SH_PT_16.
5689 (shmedia_md_apply_fix): Hold original fixP->fx_r_type in
5690 orig_fx_r_type. Change SHMEDIA_BFD_RELOC_PT into
5691 BFD_RELOC_SH_PT_16. Handle BFD_RELOC_SH_PT_16 as pc-relative.
5692 <resolved previously-pc-relative relocs>: Handle
5693 SHMEDIA_BFD_RELOC_PT and BFD_RELOC_SH_PT_16.
5694 (shmedia_md_convert_frag) <case C (SH64PCREL16PT_64, SH64PCREL16),
5695 case C (SH64PCREL16PT_32, SH64PCREL16)>: Modify to PTB if operand
5696 points to SHcompact code.
5697 <case C (SH64PCREL16_32, SH64PCREL16), case C (SH64PCREL16_64,
5698 SH64PCREL16)>: Check that ISA of what operand points at and
5699 PTA/PTB matches, or emit error.
5700 (shmedia_check_limits): Handle BFD_RELOC_SH_PT_16 and
5701 SHMEDIA_BFD_RELOC_PT.
5702 (shmedia_immediate_op): If pcrel, emit fixup also for constant
5704 (shmedia_build_Mytes) <case A_IMMS16>: Also check sh64_expand in
5705 condition for MOVI expansion.
5706 <case A_PCIMMS16BY4>: Handle expansion to 32 bits only, if
5707 sh64_pt32. Emit only a BFD_RELOC_SH_PT_16 fixup if not
5709 <case A_PCIMMS16BY4_PT>: Likewise, but emit a SHMEDIA_BFD_RELOC_PT
5711 (sh64_target_format): Error-check setting of sh64_pt32 and
5712 sh64_expand. Fix typo in check for sh64_shcompact_const_crange.
5713 (shmedia_md_pcrel_from_section): Handle BFD_RELOC_SH_PT_16 and
5714 SHMEDIA_BFD_RELOC_PT as coming from SHmedia code.
5716 * config/tc-sh64.c: Improve comments.
5717 (shmedia_md_convert_frag): Remove inactive is_pt_variant code. Do
5718 not say the linker will check correctness of PTA/PTB expansion.
5719 (shmedia_md_end): Make non-static.
5720 * config/tc-sh64.h (md_end): Define to shmedia_md_end. Add
5722 * config/tc-sh.c (sh_finalize): Remove.
5723 * config/tc-sh.h (md_end): Do not define.
5724 Remove prototype for sh_finalize.
5726 * config/tc-sh64.c (shmedia_frob_section_type): Use a struct
5727 sh64_section_data container when storing section type in tdata
5728 field in elf_section_data.
5729 * config/tc-sh.c (sh_elf_final_processing): Change from EF_SH64 to
5731 * Makefile.am: Update dependencies.
5732 * Makefile.in: Regenerate.
5734 * config/tc-sh64.c (shmedia_md_assemble): Don't protect
5735 dwarf2_emit_insn call with test on debug_type.
5737 * config/tc-sh64.c (sh64_set_contents_type): Make contents-type
5738 CRT_SH5_ISA32 sticky for 64-bit.
5740 Generate .crange sections when switching ISA mode or emitting
5741 constants in same section as code.
5742 * config/tc-sh64.c: Reformat structure definitions.
5743 (sh64_end_of_assembly, sh64_mix, sh64_shcompact_const_crange): New
5745 (sh64_set_contents_type): Rename from sh64_init_section. Rewrite
5746 to emit a .cranges descriptor when contents type changes. Only
5747 emit error if changing contents type and -no-mix is in effect.
5748 (sh64_emit_crange, sh64_flush_last_crange, sh64_flag_output,
5749 sh64_flush_pending_output, sh64_tc_cons_fix_new): New functions.
5750 (shmedia_md_end): Set sh64_end_of_assembly. Pass
5751 sh64_flush_last_crange over sections.
5752 When checking main symbol of datalabel symbol, check for
5753 STO_SH5_ISA32, not ISA type of section in definition.
5754 (shmedia_frob_file_before_adjust): Check main symbol for
5755 STO_SH5_ISA32; don't check ISA type of section in definition.
5756 (shmedia_frob_section_type): Adjust for .cranges; set section flag
5757 to SHF_SH5_ISA32_MIXED or SHF_SH5_ISA32 according to whether
5758 .cranges entries have been output.
5759 (shmedia_elf_new_section): Just call md_flush_pending_output.
5760 (shmedia_md_assemble): Do not emit a BFD_RELOC_SH_SHMEDIA_CODE
5761 fix. Do not set tc_segment_info_data.in_code for section. Call
5762 sh64_set_contents_type for SHmedia code.
5763 (s_sh64_mode): Do not call sh64_init_section or set seen_insn to
5764 false. Call md_flush_pending_output.
5765 (sh64_target_format): Check that -no-mix and
5766 -shcompact-const-crange are used in sane combination with other
5768 (shmedia_md_pcrel_from_section): Check type of fix for how to
5770 (sh64_consume_datalabel): Check symbol for having STO_SH5_ISA32,
5771 not ISA type of section in definition.
5772 * config/tc-sh64.h (struct sh64_segment_info_type): Rewrite to
5773 hold contents-type state.
5774 (md_flush_pending_output): Redefine to sh64_flush_pending_output.
5775 (sh64_flush_pending_output): Declare.
5776 (TC_CONS_FIX_NEW): Define to sh64_tc_cons_fix_new.
5777 (sh64_tc_cons_fix_new): Declare.
5778 * config/tc-sh.c (sh_elf_cons) [HAVE_SH64]: Unset seen_insn and
5779 call sh64_flag_output.
5780 (md_assemble) [HAVE_SH64]: Do not emit BFD_RELOC_SH_CODE. Just
5781 call sh64_set_contents_type to mark SHcompact code and set
5783 (md_longopts): New options "-no-mix" and
5784 "-shcompact-const-crange".
5785 (md_parse_option): Handle new options.
5786 (md_show_usage): Add blurb for new options.
5787 (md_number_to_chars) [HAVE_SH64]: Call sh64_flag_output.
5789 * config/tc-sh64.c: Delete investigated and obsolete fixme:s.
5790 (sh64_last_insn_frag): New.
5791 (shmedia_md_convert_frag): Use tc_frag_data field of incoming frag
5792 to get frag for insn opcode for generating fixups; do not assume it
5794 (shmedia_build_Mytes): Set sh64_last_insn_frag after growing frag
5796 * config/tc-sh64.h (ELF_TC_SPECIAL_SECTIONS): Define for .cranges
5798 (TC_FRAG_TYPE): Define as fragS *.
5799 (TC_FRAG_INIT): Define to set tc_frag_data to sh64_last_insn_frag.
5800 (sh64_last_insn_frag): Declare.
5801 (sh64_consume_datalabel): Fix typo; check for seginfo != NULL,
5802 not == NULL before dereferencing.
5804 Get rid of BFD section flag and EF_SH64_ABI64.
5805 * config/tc-sh64.c (shmedia_frob_section_type): Use
5806 elf_section_data (sec)->tdata, not a specific BFD section flag, to
5807 communicate the section as containing SHmedia code. Describe why.
5808 * config/tc-sh.c (sh_elf_final_processing): Tweak comment. Set
5809 EF_SH64 regardless of ABI.
5810 * config/tc-sh64.c (shmedia_md_apply_fix): Decapitalize "invalid"
5811 in error message. Handle resolved expressions for
5812 BFD_RELOC_SH_IMMS10, BFD_RELOC_SH_IMMS10BY2,
5813 BFD_RELOC_SH_IMMS10BY4 and BFD_RELOC_64.
5814 (shmedia_check_limits): Handle BFD_RELOC_64.
5815 (sh64_adjust_symtab): Do not decrement the GAS symbol value for
5816 a STO_SH5_ISA32 symbol, only the BFD value.
5818 * config/tc-sh64.c: Call dwarf2_emit_insn, not the defunct
5819 dwarf2_generate_asm_lineno.
5821 Handle PC-relative MOVI expansions with assembler relaxation.
5822 Generate PC-relative relocs from 16-bit PC-relative expressions.
5823 * config/tc-sh64.c (SHMEDIA_MD_PCREL_FROM_FIX): Break out from...
5824 (shmedia_md_pcrel_from_section): ...here.
5825 (shmedia_md_apply_fix): Handle fixups for 16-bit operands that has
5826 turned completely resolved. Adjust relocation type for 16-bit
5827 immediate operands that has turned PC-relative. Adjust back for
5828 MD_PCREL_FROM_SECTION being applied twice.
5829 (shmedia_md_convert_frag): Always emit reloc for expression with
5830 global or weak symbol. Handle relaxation result for PC-relative
5832 (shmedia_md_estimate_size_before_relax): An expression with a weak
5833 or global symbol can not be relaxed. Break out tests for
5834 relaxable symbol into variable sym_relaxable.
5835 <cases C (MOVI_IMM_64, UNDEF_MOVI) and C (MOVI_IMM_32,
5836 UNDEF_MOVI)>: Break out any PC-relative expression and change
5838 (shmedia_build_Mytes): CSE &operands->operands[j] into variable
5840 <case A_IMMS16>: Fix typo for initial minor relaxation type of
5841 MOVI expansion. If X_op_symbol of the immediate expression is
5842 set, make an expression symbol for the argument to frag_var.
5843 * config/tc-sh.c (MOVI_IMM_32_PCREL, MOVI_IMM_64_PCREL): New
5845 (END): Adjust for new relaxations.
5846 (md_relax_table): Add entries for new relaxations.
5848 * config/tc-sh64.c (shmedia_parse_reg): Initialize variable len.
5850 * config/tc-sh64.c (shmedia_md_convert_frag): Correct all MOVI and
5851 SHORI operand offsets in PT/PTA/PTB expansions.
5853 Implement DataLabel semantics.
5854 * config/tc-sh.c (sh_frob_file) [HAVE_SH64]: Call
5855 shmedia_frob_file_before_adjust.
5856 * config/tc-sh64.c [! OBJ_ELF]: Emit #error.
5857 (DATALABEL_SUFFIX): Define.
5858 (shmedia_md_end) <before adjusting STO_SH5_ISA32 symbols>: Walk
5859 symbol list to update "datalabel" symbols to their main symbol
5861 (shmedia_frob_file_before_adjust): New.
5862 (sh64_adjust_symtab): For remaining datalabel symbols, set to
5863 undefined and set STT_DATALABEL.
5864 (sh64_frob_label): Initialize TC symbol field.
5865 (sh64_consume_datalabel): Actually implement semantics. New
5866 parameter operandf, call it instead of expression.
5867 (sh64_exclude_symbol): New.
5868 * config/tc-sh64.h (md_parse_name): Pass on the function operand
5869 to sh64_consume_datalabel.
5870 (tc_symbol_new_hook): Define to tc_frob_symbol.
5871 (TC_SYMFIELD_TYPE): Define to symbolS *.
5872 (tc_frob_symbol): Define to call sh64_exclude_symbol.
5874 * config/tc-sh64.c (shmedia_init_reloc): Tweak comment for default
5876 (shmedia_md_assemble): Call dwarf2_generate_asm_lineno if
5877 generating dwarf2 debug information.
5879 * config/tc-sh64.c (sh64_target_format): Use elf64-sh64l and
5880 elf64-sh64 for the 64-bit ABI.
5881 * config/tc-sh.c (md_show_usage): Tweak usage output for -abi=*
5884 * config/tc-sh.c: Remove conditionalizing on HAVE_SH64 for
5887 * config/tc-sh64.c: Tweak comments, formatting and error messages.
5888 (enum sh64_abi_values): New type.
5889 (enum sh64_isa_values): New type.
5890 (sh64_isa_mode): Replace shmedia_mode. All referers changed.
5891 (seen_shcompact_mode, seen_shmedia_mode): Delete.
5892 (sh64_abi): Replace shmedia_64.
5893 (shmedia_md_convert_frag) <C (MOVI_IMM_64, MOVI_64),
5894 C (MOVI_IMM_32, MOVI_32)>: Correct register number handling.
5895 (s_sh64_mode): Check validity for this target.
5896 (sh64_target_format): Initialize defaults for ISA and ABI.
5897 Fallback to old object format if no SH64 ISA or ABI has been
5899 * config/tc-sh.c (md_parse_option): Check combinations for errors.
5900 (sh_elf_final_processing): Change to have EF_SH64_ABI64 for 64-bit
5901 ABI and EF_SH64 for 32-bit ABI, if SH64 options are specified.
5902 * config/tc-sh64.h: Fix typo in comment.
5904 * config/tc-sh64.c (shmedia_md_estimate_size_before_relax)
5905 <PT fixups for absolute values>: Size will be longest, not
5907 (shmedia_md_convert_frag): Disable PTB-warning machinery. Correct
5908 all MOVI and SHORI operand offsets in PT/PTA/PTB expansions.
5909 * config/tc-sh.c (parse_reg) [HAVE_SH64]: Add local variables l0
5910 and l1 to hold lowercase of two first characters. Change all
5911 remaining TO_LOWER to tolower.
5912 * config/tc-sh64.c (TO_LOWER): Delete.
5913 (shmedia_find_cooked_opcode): Use tolower, not TO_LOWER.
5914 (md_parse_name): Define.
5915 (sh64_consume_datalabel): Declare.
5916 (DOLLAR_DOT): Define.
5917 * config/tc-sh64.c (shmedia_parse_exp): New.
5918 (sh64_consume_datalabel): New; just ignoring datalabel semantics.
5919 (shmedia_parse_reg): Remove const from src
5921 (shmedia_get_operands): Ditto for args parameter and ptr variable.
5922 (shmedia_md_assemble): Ditto for op_end variable.
5923 (shmedia_get_operand): Ditto for ptr parameter and src variable.
5924 Use shmedia_parse_exp, not parse_exp.
5925 * config/tc-sh64.c (shmedia_parse_reg): Add shmedia_arg_type
5926 parameter. All callers changed.
5927 (shmedia_get_operand): Add shmedia_arg_type parameter. All
5929 (shmedia_parse_reg): Put first two character in local variables.
5930 Use tolower, not TO_LOWER. If no register is found and argtype
5931 indicates a control register, scan shmedia_creg_table
5934 * Makefile.am (CPU_TYPES): Add sh64.
5935 (TARGET_CPU_CFILES): Add config/tc-sh64.c.
5936 (TARGET_CPU_HFILES): Add config/tc-sh64.h.
5937 Regenerate dependencies.
5938 * Makefile.in: Regenerate.
5939 * configure.in: Add support for sh64-*-elf*.
5940 * configure: Regenerate.
5941 * config/tc-sh64.h: New.
5942 * config/tc-sh64.c: New.
5943 * config/tc-sh.c (md_pseudo_table) [HAVE_SH64]: New pseudos
5944 .mode, .isa and .uaquad.
5945 [HAVE_SH64] (SH64PCREL16_32, SH64PCREL16_64, SH64PCREL16PT_32,
5946 SH64PCREL16PT_64, MOVI_IMM_32, MOVI_IMM_64): Define.
5947 [HAVE_SH64] (END): Define as 10.
5948 [HAVE_SH64] (UNDEF_SH64PCREL, SH64PCREL16, SH64PCREL32,
5949 SH64PCREL48, SH64PCREL64, UNDEF_MOVI, MOVI_16, MOVI_32, MOVI_48,
5951 [HAVE_SH64] (SH64PCREL16_F, SH64PCREL16_M, SH64PCREL16_LENGTH,
5952 SH64PCREL32_F, SH64PCREL32_M, SH64PCREL32_LENGTH, SH64PCREL48_F,
5953 SH64PCREL48_M, SH64PCREL48_LENGTH, SH64PCREL64_LENGTH,
5954 MOVI_16_LENGTH, MOVI_32_LENGTH, MOVI_48_LENGTH, MOVI_64_LENGTH):
5956 (md_relax_table) [HAVE_SH64]: Provide relaxations for SHmedia.
5957 (md_begin) [HAVE_SH64]: Call shmedia_md_begin.
5958 (parse_reg) [HAVE_SH64]: Parse register names case-insensitive.
5959 (md_assemble) [HAVE_SH64]: Call shmedia_md_assemble if assembling
5960 SHmedia instructions. Handle state-change after switching to
5962 (md_longopts) [HAVE_SH64]: New options --isa=* and --abi=*.
5963 (md_parse_option) [HAVE_SH64]: Parse new options.
5964 (md_show_usage) [HAVE_SH64]: Show usage of new options.
5965 (md_convert_frag) [HAVE_SH64] <default>: Call
5966 shmedia_md_convert_frag instead of abort.
5967 (sh_force_relocation) [HAVE_SH64]: Also force relocation for
5968 BFD_RELOC_SH_SHMEDIA_CODE.
5969 (sh_elf_final_processing) [HAVE_SH64]: Set flags identifying
5970 SHcompact or SHmedia code.
5971 (md_apply_fix) [HAVE_SH64] <default>: Return result from calling
5972 shmedia_md_apply_fix instead of abort.
5973 (md_estimate_size_before_relax) [HAVE_SH64] <default>: Return
5974 result from calling shmedia_md_estimate_size_before_relax instead
5976 (sh_do_align) [HAVE_SH64]: If shmedia_mode, let shmedia_do_align
5978 (tc_gen_reloc) [HAVE_SH64]: For unrecognized relocs, call
5979 shmedia_init_reloc and do nothing more if it returns non-zero.
5980 (sh_finalize) [HAVE_SH64]: Call shmedia_md_end.
5981 * po/POTFILES.in: Regenerate.
5982 * po/gas.pot: Regenerate.
5986 * config/tc-sh.c (parse_at): Install the correct version of
5989 * config/tc-sh.c (md_apply_fix3) <BFD_RELOC_32_PLT_PCREL>: Don't
5990 assume fixP->fx_subsy is non-NULL.
5994 * config/tc-sh.c (parse_at): Set arg type of @(expr,pc) to
5995 A_DISP_PC_ABS, and adjust it by -4.
5996 (get_specific): Accept A_DISP_PC_ABS where A_DISP_PC is
5998 (build_Mytes): Mark PCRELIMM fix-ups as pc-relative only if
5999 the operand type is not A_DISP_PC_ABS.
6003 * config/tc-mmix.c (tc_gen_reloc): Don't try and take the value of
6004 common and weak symbols. Handle common and weak symbols as
6005 undefined symbols with regards to GREG handling and merging.
6006 (mmix_frob_file): Ditto.
6010 * configure.in (hppa-*-netbsd*): New target.
6011 * configure: Regenerate.
6012 * config/tc-hppa.h: Also define WARN_COMMENTS if TE_NetBSD.
6016 * config/tc-v850.c: Add missing prototypes amd use old-style
6017 function definitions.
6018 (AREA_ZDA, AREA_SDA, AREA_TDA): Delete.
6019 (sdata_section tdata_section, zdata_section, sbss_section,
6020 tbss_section, zbss_section, rosdata_section, rozdata_section,
6021 scommon_section, tcommon_section, zcommon_section,
6022 call_table_data_section, call_table_text_section): Delete.
6023 (v850_sdata, v850_tdata, v850_zdata, v850_sbss, v850_tbss,
6024 v850_zbss, v850_bss, v850_rosdata, v850_rozdata,
6025 v850_call_table_data, v850_call_table_text): Delete.
6026 (struct v850_seg_entry): New.
6027 (v850_seg_table): New.
6028 (SDATA_SECTION TDATA_SECTION, ZDATA_SECTION, SBSS_SECTION,
6029 TBSS_SECTION, ZBSS_SECTION, BSS_SECTION, ROSDATA_SECTION,
6030 ROZDATA_SECTION, SCOMMON_SECTION, TCOMMON_SECTION, ZCOMMON_SECTION,
6031 CALL_TABLE_DATA_SECTION, CALL_TABLE_TEXT_SECTION): Define.
6034 (v850_comm): Use do_v850_seg and v850_seg_table. Simplify
6035 recording of alignment.
6036 (md_pseudo_table): Use v850_seg.
6037 (md_begin): Don't init .call_table_data and .call_table_text here.
6038 Set v850_seg_table bss entry.
6042 Support on-demand global register allocation by passing on
6043 base-plus-offset relocs to the linker.
6044 * config/tc-mmix.c: Tweak and fix typos in comments.
6045 (allocate_undefined_gregs_in_linker): New variable.
6046 (OPTION_LINKER_ALLOCATED_GREGS): New option macro.
6047 (md_longopts): Add --linker-allocated-gregs.
6048 (md_parse_option) <case 'x'>: Imply --linker-allocated-gregs.
6049 <case OPTION_LINKER_ALLOCATED_GREGS>: New.
6050 (md_show_usage): Update text for -x. Add text for
6051 --linker-allocated-gregs.
6052 (tc_gen_reloc): Derive default value for addend from val and
6053 baddsy. Use addsec and bfd_is_abs_section in more places. Don't
6054 emit error for BFD_RELOC_MMIX_BASE_PLUS_OFFSET without suitable
6055 GREG if allocate_undefined_gregs_in_linker.
6056 * doc/as.texinfo (Overview) <Target MMIX options>: Add
6057 --linker-allocated-gregs.
6058 * doc/c-mmix.texi (MMIX-Opts): Add blurb about
6059 --linker-allocated-gregs. Mention that it's implied by -x.
6060 (MMIX-Pseudos) <GREG>: Mention when and how a GREG can be omitted.
6061 (MMIX-mmixal): Clarify dated comparison and location of MMIXware.
6063 * config/tc-mmix.h (md_parse_name): Use ISUPPER, not isupper.
6067 * Makefile.am: Run "make dep-am"
6068 * Makefile.in: Regenerate.
6072 * config/tc-or32.c: New file.
6073 * config/tc-or32.h: New file.
6074 * configure.in: Add support for or32 targets.
6075 * configure: Regenerate.
6076 * config/obj-coff.c: Add support for or32 targets.
6077 * config/obj-coff.h: Add support for or32 targets.
6078 * Makefile.am: Add support for or32 targets.
6079 * Makefile.in: Regenerate.
6080 * NEWS: Mention support for OpenRISC.
6081 * doc/Makefile.in: Regenerate.
6082 * po/POTFILES.in: Regenerate.
6083 * po/gas.pot: Regenerate.
6087 * config/tc-sh.c (parse_reg): Fix end-of-word check for is, ix, iy
6092 * config/tc-mips.c (tc_gen_reloc): Arrange for
6093 BFD_RELOC_PCREL_HI16_S relocations to be output relative to
6094 their LO16 parts, even for ELF.
6098 * config/tc-i386.c: Protect definitions of true and false
6103 * config/obj-elf.c (elf_frob_file_before_adjust): Remove symbols
6104 made because of .weak, if they are neither defined nor used in any
6109 * configure: Regenerated.
6113 * doc/Makefile.am (install): Depend on install-info.
6114 * doc/Makefile.in: Regenerate.
6118 * po/fr.po: Updated version
6122 * config/tc-h8300.c (check_operand): Don't print a warning
6123 when a valid 24-bit address is given to a 16-bit address
6128 * config/tc-sh.c (sh_elf_suffix): Removed.
6129 (sh_PIC_related_p, sh_check_fixup, sh_cons_fix_new,
6130 sh_end_of_match, sh_parse_name): New functions.
6131 (sh_elf_cons): Simplify.
6132 (parse_exp): Reject misplaced PIC operands.
6133 (md_undefined_symbol): Simplify.
6134 (sh_fix_adjustable): Let @GOTOFF be adjusted.
6135 (md_apply_fix3): Write @PLT and @GOTOFF addends in place.
6136 (tc_gen_reloc): Move fixp subsy absolute value into addnumber.
6137 Complain if subsy remains at the end.
6138 * config/tc-sh.h (sh_parse_name, sh_cons_fix_new): Declare.
6139 (md_parse_name, TC_CONS_FIX_NEW, O_PIC_reloc): Define.
6143 * config/tc-mn10300.c (xr_registers): Move `pc'...
6144 (other_registers): ... here.
6148 * Makefile.am: Run "make dep-am".
6149 * Makefile.in: Regenerate.
6150 * gas/po/POTFILES.in: Regenerate.
6154 * config/obj-coff.c (obj_coff_init_stab_section): Make the
6155 stabstr_name allocation permanent, as it will be referenced from
6160 * configure.in (ia64-*-netbsd*): New target.
6161 * configure: Regenerate.
6165 * doc/as.texinfo (Overview) <Target ARM options>: Add missing {}
6167 <Detailed description, ARM options>: Ditto.
6168 * doc/c-arm.texi (ARM Options): Ditto.
6172 * tc-arm.c (do_xsc_mia, do_xsc_mar, do_xsc_mra): Renamed from
6173 do_mia, do_mar and do_mra respectively.
6174 (do_mav_*): Renamed from do_c_*.
6175 (mav_reg_required_here, mav_parse_offset): Renamed from
6176 cirrus_reg_required_here and cirrus_parse_offset respectively.
6177 (MAV_MODE?): Renamed from CIRRUS_MODE?.
6182 * tc-arm.c (ARM_EXT_V5J, ARM_ARCH_V5TEJ): Define.
6183 (insns): Add pattern for bxj instruction.
6184 (do_bxj): New function.
6185 (arm_cpus): Add arm926ej.
6186 (arm_archs): Add armv5tej.
6190 * doc/c-arm.texi: Add new fpe options to list of supported flags.
6194 * tc-arm.c (arm_fpus): Add fpe2 and fpe3.
6198 * NEWS: Mention new ARM command-line options and VFP support.
6200 * config/tc-arm.c (ARM_CEXT_XSCALE): Replaces ARM_EXT_XSCALE. All
6202 (ARM_CEXT_MAVERICK): Similarly.
6203 (ARM_ANY): Now means any core instruction.
6204 (CPU_DEFAULT): Default to ARM_ANY.
6205 (uses_apcs_26, atcps, support_interwork, uses_apcs_float)
6206 (pic_code): Declare for all object types. Make type int.
6207 (legacy_cpu, legacy_fpu, mcpu_cpu_opt, mcpu_fpu_opt, march_cpu_opt)
6208 (march_fpu_opt, mfpu_opt): Declare.
6209 (md_longopts): Tidy up conditional definitions.
6210 (arm_opts, arm_cpus, arm_archs, arm_fpus, arm_extensions)
6211 (arm_long_opts): New tables.
6212 (arm_parse_cpu, arm_parse_arch, arm_parse_fpu): New functions.
6213 (arm_parse_extension): New function.
6214 (md_parse_option): Rewrite using new table-driven system.
6215 (md_show_usage): Use new table-driven system.
6216 (md_begin): Calculate cpu_variant from command line option data.
6217 * doc/as.texinfo (ARM ISA options): Docuement new ARM-specific
6218 command-line options.
6219 * doc/c-arm.texi: Likewise.
6223 * as.c (parse_args): Update year.
6227 * config/tc-tic54x.c (encode_address): Add a more informative
6228 warning about incorrect syntax.
6232 * po/gas.pot: Regenerate.
6236 * macro.c (get_any_string): Add no-c-format comment to prevent
6237 confusion when translating string.
6238 * gasp.c (get_any_string): Add no-c-format comment to prevent
6239 confusion when translating string.
6243 * config/tc-mips.c (percent_op): Ensure longer percent_op's are
6244 matched before the shorter ones.
6245 (my_getSmallParser): Fix handling of nested parentheses in
6246 percent_op's. Code cleanup.
6247 (my_getPercentOp): New function, code from my_getSmallParser.
6248 (my_getSmallExpression): Fix handling of closing parentheses.
6249 Code cleanup. Better comments.
6253 po/tr.po: New file: Turkish translation.
6254 configure.in (LINGUAS): Add "tr".
6255 configure: Regenerate.
6259 Support for VFP instructions
6260 * tc-arm.c (CP_WB_OK, CP_NO_WB): New defines.
6261 (cp_address_required_here): New argument wb_ok. When false, do not
6262 accept write-back forms of addressing. Change all callers.
6263 (FPU_VFP_EXT_NONE, FPU_VFP_EXT_V1xD, FPU_VFP_VFP_V1)
6264 (FPU_VFP_EXT_V2): Define.
6265 (FPU_ARCH_VFP, FPU_ARCH_VFP_V1xD, FPU_ARCH_VFP_V1, FPU_ARCH_VFP_V2):
6266 Define in terms of above.
6267 (vfp_dp_reg_pos, vfp_sp_reg_pos, vfp_ldstm_type): New enums.
6268 (vfp_reg): New struct.
6269 (vfp_regs): New array of registers.
6270 (insns): Add VFP instructions.
6271 (sn_table): New array of VFP single-precision register names.
6272 (dn_table): New array of VFP double-precision register names.
6273 (all_reg_maps): Add the new register tables.
6274 (arm_reg_type): Add new values for above. Increase RET_TYPE_MAX.
6275 (vfp_sp_reg_required_here, vfp_dp_reg_required_here, do_vfp_sp_monadic)
6276 (do_vfp_dp_monadic, do_vfp_sp_dyadic, do_vfp_dp_dyadic)
6277 (do_vfp_reg_from_sp, do_vfp_sp_reg2, do_vfp_sp_from_reg)
6278 (do_vfp_reg_from_dp, do_vfp_reg2_from_dp, do_vfp_dp_from_reg)
6279 (do_vfp_dp_from_reg2, vfp_psr_parse, vfp_psr_required_here)
6280 (do_vfp_reg_from_ctrl, do_vfp_ctrl_from_reg, do_vfp_sp_ldst)
6281 (do_vfp_dp_ldst, vfp_sp_reg_list, vfp_dp_reg_list, vfp_sp_ldstm)
6282 (vfp_dp_ldstm, do_vfp_sp_ldstmia, do_vfp_sp_ldstmdb, do_vfp_ldstmia)
6283 (do_vfp_dp_ldstmdb, do_vfp_xp_ldstmia, do_vfp_xp_ldstmdb)
6284 (do_vfp_sp_compare_z, do_vfp_dp_compare_z, do_vfp_dp_sp_cvt)
6285 (do_vfp_sp_dp_cvt): New functions.
6286 (md_begin): Set soft-float flag for appropriate VFP work.
6287 (md_atof): Handle VFP-format doubles.
6288 (md_parse_option): Handle VFP command-line options.
6289 (md_show_usage): Display VFP command-line options.
6293 * tc-arm.c (md_parse_option): Tidy up setting of cpu_variant for
6294 various command line options.
6298 * config/tc-xstormy16.c: (xstormy16_fix_adjustable): Do not fix
6300 (xstormy16_md_apply_fix3): Do not return a value.
6304 * tc-arm.c (md_longopts): On targets that aren't bi-endian, support
6305 the -EL/-EB option that matches the target's endianness.
6306 (md_parse_option): Likewise.
6310 * tc-arm.c (md_longopts): Fix misplaced #endif -- the -oabi option
6311 is not dependent on ARM_BI_ENDIAN.
6315 * tc-arm.c (all error messages): Normalize capitalization of messages.
6317 * tc-arm.h (md_operand): Delete define.
6318 * tc-arm.c (in_my_get_expression): New static variable.
6319 (my_get_expression): Set and clear it.
6320 (md_operand): New function. If called from my_get_expression
6321 put the error in inst.error.
6322 (output_inst): Now takes argument of instruction being assembled.
6323 Print it out with any error message.
6324 (do_ldst, do_ldstv4, thumb_load_store): Fault attempt to use a store
6326 (end_of_line): Don't update inst.error if it is already set.
6330 * tc-arm.c (do_ldst): Fix handling an immediate expression pseudo
6331 op that can be translated into a mvn instruction.
6335 * gas/config/tc-ia64.h (MD_FLAGS_DEFAULT): New Macro for
6336 setting default md.flags.
6337 (SHT_INIT_ARRAY): New elf special section used by HP-UX.
6338 (SHT_FINI_ARRAY): New elf special section used by HP-UX.
6339 * gas/config/tc-ia64.c (setup_unwind_header): Add support
6340 for 32 bit unwind info blocks.
6341 (generate_unwind_image): Add support for different types
6342 of unwind images (32 bits and/or big-endian).
6343 (ia64_init): Use MD_FLAGS_DEFAULT to set md.flags.
6344 (ia64_target_format): Add support for hpux target formats.
6345 (ia64_gen_real_reloc_type): Add support for FUNC_IPLT_RELOC.
6346 (ia64_elf_section_type): Add support for SHT_INIT_ARRAY and
6347 SHT_FINI_ARRAY elf section types.
6351 * tc-arm.c (struct reg_entry): Move before prototypes.
6352 (int_register, cp_register, fp_register): Delete.
6353 (reg_table): Delete. Replaced with ...
6354 (rn_table, cp_table, cn_table, fn_table, mav_mvf_table)
6355 (mav_mvd_table, mav_mvfx_table, mav_mvdx_table, mav_mvax_table)
6356 (mav_dspsc_table): ... one table per register set.
6357 (arm_reg_hsh): Delete.
6358 (struct reg_map): New structure.
6359 (all_reg_maps): New array.
6360 (enum arm_reg_type): New enums.
6361 (build_reg_hsh): New function.
6362 (insert_reg_alias): Use hash table passed by caller. Adjust all
6364 (create_register_alias): New function, split out from ...
6365 (md_assemble): ... here.
6366 (md_begin): Build new register hash tables.
6367 (arm_reg_parse): New argument for the hash table to search. Adjust all
6369 (arm_reg_parse_any): New function.
6370 (co_proc_number): Look up the processor number in the processor hash
6372 (cirrus_regtype): Delete.
6373 (cirrus_register, cirrus_mvf_register, cirrus_mvd_register)
6374 (cirrus_mvfx_register, cirrus_mvdx_register, cirrus_mvax_register)
6375 (ARM_EXT_MAVERICKsc_register): Delete.
6376 (do_c_binops_1, do_c_binops_2, do_c_binops_3): Delete.
6377 (do_c_binops_1[a-o], do_c_binops_2[a-c], do_c_binops_3[a-d]): New
6379 (do_c_triple_4, do_c_triple_5): Delete.
6380 (do_c_triple_4[ab], do_c_triple_5[a-h]): New functions.
6381 (do_c_quad_6): Delete.
6382 (do_c_quad_6[ab]): New functions.
6383 (do_c_binops, do_c_triple, do_c_quad, do_c_shift, do_c_ldst): Rework
6384 arguments to use new register parsing methods.
6385 (cirrus_reg_required_here): Likewise.
6386 (insns): Reclassify cirrus maverick worker functions.
6387 (cirrus_valid_reg): Delete.
6391 * configure.in (sh*le): Set cpu_type=sh and endian=little.
6392 (sh*-*-netbsdelf*): New target.
6393 * configure: Regenerate.
6394 * tc-sh.h: Update copyright years.
6395 (TARGET_FORMAT): Add version for TE_NetBSD.
6399 * read.c (emit_expr): Do not allow 'size' or 'nbytes' to go
6404 * config/tc-m68k.h (md_prepare_relax_scan): Rewrite.
6405 * config/tc-m68k.c (md_relax_table): Add md_prepare_relax_scan comment.
6409 * tc-mips.c (mips_cprestore_valid): New flag.
6410 (mips_frame_reg_valid): New flag.
6411 (macro) [M_JAL_2]: Check both flags.
6412 [M_JAL_A]: Likewise.
6413 (s_cprestore): Set mips_cprestore_valid.
6414 (tc_get_register): If setting mips_frame_reg, set
6415 mips_frame_reg_valid and clear mips_cprestore_valid.
6416 (s_mips_ent): Clear both flags.
6417 (s_mips_end): Clear both flags.
6421 * write.c (write_object_file): Make use of bfd_section_list_remove.
6422 * config/obj-ecoff.c (ecoff_frob_file): Likewise.
6423 * config/tc-mmix.c (mmix_frob_file): Likewise.
6427 * configure.in (i386-*-netbsdelf*): Collapse target into...
6428 (i386-*-netbsd*): ...this. Add support for x86-64.
6429 * configure: Regenerated.
6433 * config/tc-ppc.c (md_parse_option): BookE is not Motorola specific.
6437 * config/tc-m68k.c (md_estimate_size_before_relax): Test for a
6440 For older changes see ChangeLog-0001
6446 version-control: never