3 * configure.in: Make 'mipself' and 'mipsecoff' emulations
4 map to MIPS-specific files, as they used to do before the
10 * config/tc-mips.c (md_parse_option): Don't try to compile
11 ELF-only option code if not ELF.
15 * config/tc-ia64.h (md_elf_section_type): New macro.
16 (ELF_TC_SPECIAL_SECTIONS): Drop .IA_64.unwind and .IA_64.unwind_info
17 (they're now handled via ia64_elf_section_type.
19 * config/tc-ia64.c (unwind): New members saved_text_seg,
20 saved_text_subseg, and force_unwind_entry.
21 (optimize_unw_records): New function to optimize away unnecessary
23 (ia64_elf_section_type): New function.
24 (output_unw_records): Generate unwind info only if the size is
25 non-zero or if it's forced for some other reason (e.g.,
26 handlerdata or a personality routine).
27 (generate_unwind_image): Don't switch back to previous
28 section---stay inside the unwind info section instead so that
29 handlerdata that may follow goes into the right place.
30 (dot_handlerdata): Force generation of unwind entry and save the
31 current active text segment before generating unwind image.
32 (dot_unwentry): Force generation of unwind entry.
33 (dot_personality): Ditto.
34 (dot_endp): Generate unwind table entry only if there is
35 some unwind info or the unwind entry was forced.
37 * config/tc-ia64.c (make_unw_section_name): New macro to form
39 (generate_unwind_image): Add "text_name" argument. Use it to
40 form unwind section name.
41 (dot_handlerdata): Determine current segment (section) name and
42 pass it to generate_unwind_image().
43 (dot_endp): Determine current segment (section) name and use
44 it to determine the appropriate unwind section name.
45 (ia64_md_do_align): Add missing ATTRIBUTE_UNUSED declarations to
46 n, fill, and max arguments.
50 * Makefile.am: Add linux target for S/390.
51 * configure.in: Likewise.
52 * config/tc-s390.c: New file.
53 * config/tc-s390.h: New file.
57 * config/tc-sh.c (md_pseudo_table): Add uaquad. Use s_uacons for
58 2byte, 4byte and 8byte.
62 * config/tc-hppa.c (pa_build_unwind_subspace): Don't call
63 md_number_to_chars with size > sizeof (valueT).
67 * config/tc-ia64.h (TC_RELOC_RTSYM_LOC_FIXUP): Do fixup if
68 there is no relocation.
72 * config/tc-ia64.h (TC_RELOC_RTSYM_LOC_FIXUP): New. Defined.
74 * config/tc-ia64.c (md_parse_option): Only accept the valid
75 ia64 options on "-axxx".
79 * config/tc-ia64.c (errata_nop_necessary_p): Return 0 instead of
80 aborting for invalid operands.
84 * config/tc-hppa.c (fix_new_hppa): Pass in unwind directly rather
85 than via pointer. Update all callers.
86 (UNWIND_LOW32): Define.
87 (UNWIND_HIGH32): Define.
88 (pa_build_unwind_subspace): Use the above macros instead of dumping
89 bitfields directly. Call frag_more once rather than multiple times.
90 (md_assemble): Use UNWIND_LOW32.
92 (pa_procend): Likewise.
93 (process_exit): Use UNWIND_HIGH32.
97 * config/tc-m68hc11.h (LISTING_HEADER): Use m68hc11_listing_header
98 function to select the header according to the cpu.
99 (md_after_pass_hook, md_do_align): Remove.
100 (md_cleanup, m68hc11_cleanup): Remove.
101 (md_pcrel_from_section): Declare.
102 * config/tc-m68hc11.c (build_dbranch_insn): Remove insn_size.
103 (build_jump_insn, build_insn): Likewise.
104 (m68hc11_listing_header): New function.
105 (m68hc11_cleanup): Remove.
109 * config/tc-m68hc11.c (relaxable_symbol): Relax externally visible
110 symbols because there is no support for shared libraries and these
111 symbols can't be overridden (unless they are weak).
115 * dwarf2dbg.c (out_debug_abbrev): Terminate the abbreviations
116 for the compilation unit with a zero byte.
120 * config/tc-hppa.c (pa_ip): Support 12 bit branches to absolute
121 destinations. Correct range check for 17 and 22 bit branches.
125 * config/tc-m68k.c (tc_gen_reloc): Do not abort if tcbit is
126 still set. Issue an error message instead.
127 (md_estimate_size_before_relax): Delete unused variable
128 'buffer_address'. Fixup parentheses around if statement.
132 * as.c: Fix formatting.
134 * messages.c: Likewise.
136 * symbols.c: Likewise.
140 * config/tc-m32r.c (m32r_handle_align): Declare type of fragp.
144 * config/tc-alpha.c: Fix formatting.
148 * config/tc-alpha.c: Fix formatting.
152 * config/tc-alpha.c: Fix formatting.
156 * config/tc-arm.c (ldm_flags): Remove redundant bit from "fa" and
158 (stm_flags): Remove redundant bit from "ed" and "da" flags.
162 * configure.in (cpu_type, arch): Match i386 too.
163 * configure: Rebuilt.
167 * config/tc-i386.c: Fix formatting.
171 * config/tc-hppa.c (tc_gen_reloc): Use SEGREL32 instead of DIR32
172 relocs for .PARISC.unwind section.
174 * config/tc-hppa.c (pa_build_unwind_subspace): Build unwind
175 depending on section flags, not just for .text.
179 * config/tc-ia64.c (ia64_flush_insns): Handle unwind directives
180 not immediately followed by an instruction.
184 * config/tc-m68hc11.c: Fix formatting.
188 * symbols.c (colon): Change 'already defined symbol' from a
189 fatal error to an ordinary error. There is no reason why this
190 error should be fatal.
192 * message.c (as_fatal): Delete output file, if one has been
197 * config/tc-hppa.h (TARGET_FORMAT): Add hppa-linux variants.
201 * config/tc-alpha.c: Fix formatting.
202 * config/tc-arc.c: Likewise.
203 * config/tc-arc.h: Likewise.
204 * config/tc-d10v.c: Likewise.
205 * config/tc-i370.c: Likewise.
206 * config/tc-i386.c: Likewise.
207 * config/tc-i960.c: Likewise.
208 * config/tc-m68k.c: Likewise.
209 * config/tc-ppc.c: Likewise.
210 * config/tc-sparc.c: Likewise.
211 * config/tc-tahoe.c: Likewise.
212 * config/tc-vax.c: Likewise.
214 * config/tc-arc.c: Fix formatting.
216 * config/tc-arc.c: Fix formatting.
220 * config/tc-hppa.c (pa_build_unwind_subspace): Use SEGREL32 for
221 both 32 and 64 bit ELF.
223 * config/tc-hppa.c (pa_ip): Store `a' flag in bit zero of operand
224 and don't bother storing `m' for "ce" completer. Tidy handling of
225 'J' and 'K' operands to suit. Handle '<' and '>' operands.
229 * tc-i386.h (TARGET_MACH): New macro.
230 (i386_mach): Declare.
231 * tc-i386.c (i386_mach): New function.
235 * doc/as.texinfo: Fix spelling and cross-references.
237 * doc/c-arm.texi: Fix typos. Say that `;' is a line separator
238 character for all systems, not just GNU/Linux. Make it explicit
239 that `-k' doesn't affect code generation, just ELF flags.
243 * config/tc-i386.c (md_assemble): Check cpu_flags even for nullary
248 * cgen.c (gas_cgen_finish_insn): Call dwarf2_emit_insn.
252 * as.c (print_args): Update copyright date to 2001.
256 * doc/c-arc.texi: New file.
257 Some sections to be expanded.
261 * config/tc-i386.c (md_longopts): Recognize "--64" only for ELF.
262 (md_parse_option): Always accept "--32".
266 * as.h (TC_ARC): Ensure struc-symbol.h included.
267 * as.c (dwarf2dbg.h): Include to remove implicit declaration
269 * struc-symbol.h (SYMBOLS_NEED_BACKPOINTERS): Define.
270 (TARGET_SYMBOL_FIELDS) added.
272 * doc/Makefile.am (CPU_DOCS): Added c-arc.texi.
273 * doc/c-arc.texi: New file.
274 Some sections to be expanded.
275 * doc/as.texinfo: Update command-line options.
276 Removed outdated text for ARC dependant features, instead include
277 text from above file.
279 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Define local flag.
280 (TARGET_SYMBOL_FIELDS): Alias to previous definition.
283 * config/tc-arc.c: New updated configuration for
284 ARC, including selection of core variants, and extensibility of
285 instructions, registers etc. through directives.
287 * config/tc-arc.c (arc_extinst): Minor corrections for
289 (arc_common) Likewise. Make alignment argument optional for local
290 symbols also, with default of zero.
294 * config/tc-m68hc11.c (md_estimate_size_before_relax): Fix
295 STATE_INDEXED_OFFSET when the symbol is undefined (16-bit offset).
296 (build_indexed_byte): Don't relax indexed byte, use 16-bit offset
297 and fix_new_exp() instead.
298 (md_convert_frag): For indexed post byte use the symbol value
299 rather than the displacement.
300 (md_relax_table): Fix indexed offset relax.
304 * config/tc-m68hc11.c (md_estimate_size_before_relax):Don't
306 (relaxable_symbol): New function.
310 * config/tc-i386.h (TC_RELOC_GLOBAL_OFFSET_TABLE): Removed, it's
315 * config/tc-arm.c (arm_fix_adjustable): Define for OBJ_COFF.
316 * config/tc-arm.h (obj_fix_adjustable): Define for OBJ_COFF
320 * symbols.c (DOLLAR_LABEL_CHAR): New constant - the magic
321 character used to dollar local symbols.
322 (LOCAL_LABEL_CHAR): New constant - the magic character used to
324 (dollar_label_name): Use DOLLAR_LABEL_CHAR.
325 (fb_label_name): Prefix local labels with LOCAL_LABEL_PREFIX,
327 Use LOCAL_LABEL_CHAR.
328 (decode_local_label_name): Skip LOCAL_LABEL_PREFIX.
329 Use DOLLAR_LABEL_CHAR and LOCAL_LABEL_CHAR.
330 (S_IS_LOCAL): Use DOLLAR_LABEL_CHAR and LOCAL_LABEL_CHAR.
334 * config/tc-i386.c (i386_immediate, i386_displacement):
339 * doc/c-i386.texi (i386-Arch): Remove spaces incorrectly inserted
344 * doc/as.texinfo (Bug Reporting): Update email address for
350 * configure.in: Define DEFAULT_ARCH for i386.
351 * tc-i386.c (md_assemble): Return after the error message;
352 move testing for 64bit operands to proper place.
356 * doc/as.texinfo: Document '#' as comment character for i386 and
357 x86_64. Add AMD x86-64 into menu of machine dependent information.
359 * doc/c-i386.texi: Document x86_64 extensions.
363 * tc-i386.c (md_assemble): Handle third byte of the opcode as prefix.
367 * tc-i386.c (cpu_arch): Add Pentium4 and modify sledgehammer entry.
368 * NEWS: Add note about Pentium4 support.
372 * tc-i387.c (pi, pte, pt): Update.
373 (type_names): Add new types.
377 * tc-i386.h (CpuK6, CpuAthlon, CpuSledgehammer, CpuMMX, Cpu3dnow,
378 CpuUnknown): Renumber
379 (CpuP4, CpuSSE2): New.
380 (CpuUnknownFlags): Add CpuP4 and CpuSSE2
384 * config/tc-alpha.c (alpha_force_relocation): Handle vtable
386 (alpha_fix_adjustable): Likewise.
387 (md_apply_fix): Likewise.
391 * listing.c (listing_message): Allocate string only if it is
394 * configure: Rebuild.
398 * doc/internals.texi (Relaxing with a table) <after relaxation>:
399 Point out caveats with generating fixups for the opcode in a frag.
403 * configure.in: Add support for x86_64 and x86_64-*-linux-gnu*
408 * listing.c (calc_hex): Print the variable part only if the
409 fragment type is rs_fill.
413 * doc/internals.texi (tc_conditional_pseudoop,
414 TC_LINKRELAX_FIXUP): Fix typos.
418 * write.c (subsegs_finish): Fix thinko last change -- don't
419 "optimize" the alignment == 0 case.
423 * as.h (rs_align_test): New.
424 * frags.c (NOP_OPCODE): Move default from read.c.
425 (MAX_MEM_FOR_RS_ALIGN_CODE): New default.
426 (frag_align_code): New.
427 * frags.h (frag_align_code): Declare.
428 * read.c (NOP_OPCODE): Remove.
429 (do_align): Use frag_align_code.
430 * write.c (NOP_OPCODE): Remove.
431 (get_recorded_alignment): New.
432 (cvt_frag_to_fill): Handle rs_align_test.
433 (relax_segment): Likewise.
434 (subsegs_finish): Align last subseg in section to the
435 section alignment. Use frag_align_code.
436 * write.h (get_recorded_alignment): Declare.
437 * config/obj-coff.c (size_section): Handle rs_align_test.
438 (fill_section, fixup_mdeps): Likewise.
439 (write_object_file): Use frag_align_code.
441 * config/tc-alpha.c (alpha_align): Use frag_align_code.
442 (alpha_handle_align): New.
443 * config/tc-alpha.h (HANDLE_ALIGN): New.
444 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
446 * config/tc-i386.h (md_do_align): Use frag_align_code.
447 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
449 * config/tc-ia64.c (ia64_md_do_align): Don't do code alignment.
450 (ia64_handle_align): New.
451 * config/tc-ia64.h (HANDLE_ALIGN): New.
452 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
454 * config/tc-m32r.c (m32r_do_align): Remove.
455 (m32r_handle_align): New.
456 (fill_insn): Use frag_align_code.
457 * config/tc-m32r.h (md_do_align): Remove.
458 (HANDLE_ALIGN, MAX_MEM_FOR_RS_ALIGN_CODE): New.
459 * config/tc-m88k.c, config/tc-m88k.h: Similarly.
460 * config/tc-mips.c, config/tc-mips.h: Similarly.
462 * config/tc-sh.c (sh_cons_align): Use rs_align_test.
463 (sh_handle_align): Likewise. Handle rs_align_code.
464 (sh_do_align): Remove.
465 * config/tc-sh.h (md_do_align): Remove.
466 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
468 * config/tc-sparc.c (sparc_cons_align): Use rs_align_test.
469 (sparc_handle_align): Likewise. Handle rs_align_code.
470 * config/tc-sparc.h (md_do_align): Remove.
471 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
475 * config/tc-d10v.c (md_assemble): set prev_seg and prev_subseg
476 when we assemble the first half of a pair.
480 * config/tc-i386.c (reloc): Update the macro for non-bfd
482 (BFD_RELOC_X86_64_GOTPCREL): Set to 0 for non-bfd assembler.
486 * dwarf2dbg.c (dwarf2_finish): Remove #if BFD_ASSEMBLER.
490 * tc-i386.h (i386_target_format): Define even for ELFs.
491 (QWORD_MNEM_SUFFIX): New macro.
492 (CpuK6,CpuAthlon,CpuSledgehammer, Cpu64, CpuNo64, CpuUnknownFlags):
494 (CpuMMX,CpuSSE,Cpu3dnow, CpuUnknown): Renumber.
495 (IgnoreSize, DefaultSize, No_?Suf, FWait, IsString, regKludge, IsPrefix,
497 (Size64, No_qSuf, NoRex64, Rex64): New macros.
498 (Reg64, Imm32S, Imm64, Disp32S, Disp64): New macros.
499 (Imm8, Imm8S, Imm16, Imm32, Imm1, BaseIndex, Disp8, Disp16, Disp32,
500 InOutPortReg,ShiftCount, Control, Debug, Test, FloatReg, FloatAcc,
501 SReg2, SReg3, Acc, JumpAbsolute, RegMMX, RegXMM, EsSeg, InvMem):
503 (Reg, WordReg): Add Reg64.
504 (Imm): Add Imm32S and Imm64.
506 (Disp): Add Disp64 and Disp32S.
507 (AnyMem): Add Disp32S.
508 (RegRex, RegRex64): New macros.
509 (rex_byte): New type.
510 * tc-i386.c (set_16bit_code_flag): Kill.
511 (fits_in_unsigned_long, fits_in_signed_long): New functions.
512 (reloc): New parameter "signed"; support x86_64.
513 (set_code_flag): New.
514 (DEFAULT_ARCH): New macro; default to "i386".
515 (default_arch): New static variable.
516 (struct _i386_insn): New fields Operand_PCrel; rex.
517 (flag_16bit_code): Kill; All tests replaced to "flag_code == CODE_64BIT"
518 (flag_code): New enum and static variable.
519 (use_rela_relocations): New static variable.
520 (flag_code_names): New static variable.
521 (cpu_arch_flags): Default to CpuUnknownFlags|CpuNo64.
522 (cpu_arch): Add "sledgehammer"; Add CPUAthlon to Athlon and CpuK6 to
524 (i386_align_code): Return plain "nop" for x86_64.
525 (mode_from_disp_size): Support Disp32S.
526 (smallest_imm_type): Support Imm32S and Imm64.
527 (offset_in_range): Support size of 8.
528 (set_cpu_arch): Do not clobber to Cpu64/CpuNo64.
529 (md_pseudo_table): Add "code64"; use set_code_flat.
530 (md_begin): Emit sane error message on hash failure.
531 (tc_i386_fix_adjustable): Support x86_64 relocations.
532 (md_assemble): Support QWORD_MNEM_SUFFIX, REX registers,
533 instructions supported on particular arch just partially,
534 output of 64bit immediates, handling of Imm32S and Disp32S type.
535 (i386_immedaite): Support x86_64 relocations; support 64bit constants.
536 (i386_displacement): Likewise.
537 (i386_index_check): Cleanup; support 64bit addresses.
538 (md_apply_fix3): Support x86_64 relocation and rela.
539 (md_longopts): Add "32" and "64".
540 (md_parse_option): Add OPTION_32 and OPTION_64.
541 (i386_target_format): Call even for ELFs; choose between
542 elf64-x86-64 and elf32-i386.
543 (i386_validate_fix): Refuse GOTOFF in 64bit mode.
544 (tc_gen_reloc): Support rela relocations and x86_64.
545 (intel_e09_1): Support QWORD.
549 * config/tc-i386.c (intel_e09_1): Only flag as a memory operand if
550 it's not an offset expression.
551 (intel_e10_1): Ditto. Also, if the operand is an offset expression,
552 keep the braces '[' and ']' in the output string.
553 (intel_e11): Ditto. Also remove comparison intel_parser.op_modifier
554 != FLAT. There is no such op_modifier.
558 * dwarf2dbg.c: If we don't have <limits.h>, try including <sys/param.h>
563 * as.h: Fix formatting.
565 * dwarf2dbg.c: Likewise.
566 * input-scrub.c: Likewise.
571 * configure.in (i386-*-msdosdjgpp): Set bfd_gas to yes.
572 configure: Regenerate.
576 * dwarf2dbg.c: #include <limits.h> only if it exists.
580 * config/tc-hppa.c (pa_ip): Correct CHECK_FIELD typo.
581 (md_apply_fix): Here too.
585 * config/tc-ia64.h (ia64_init): Add prototype.
589 * dwarf2dbg.c: Enabled only if BFD_ASSEMBLER is defined.
591 * read.h (outputting_stabs_line_debug): Change it to int.
592 * stabs.c (outputting_stabs_line_debug): Likewise.
596 * config/obj-bout.c (obj_crawl_symbol_chain): Don't take
597 the address of a function result.
601 * config/tc-ppc.c (md_pseudo_table): Add .file and .loc.
602 (md_assemble): Call dwarf2_emit_insn.
603 (shlib): Fix typo SHILB -> SHLIB.
604 (md_parse_option): Likewise.
605 (ppc_elf_validate_fix): Likewise:
606 * config/tc-ppc.h (DWARF2_LINE_MIN_INSN_LENGTH): New.
610 * cgen.h: Fix formatting.
611 * input-scrub.c: Fix formatting.
612 * macro.c: Fix formatting.
613 * config/tc-mips.c: Fix formatting.
614 * doc/c-mips.texi: Fix formatting.
618 * tc-i386.c (md_assemble): Refuse 's' and 'l' suffixes in the intel
619 mode; convert 'd' suffix to 's' or 'l'; remove all DWORD_MNEM_SUFFIX
621 (intel_e09_1): Convert QWORD to 'l' suffix for FP operations; refuse
623 * tc-i386.h (DWORD_MNEM_SUFFIX): Kill.
626 * i386.h (*_Suf): Remove No_dSuf.
627 (d_suf, wld_Suf,sld_Suf, sldx_Suf, bwld_Suf, d_FP, sld_FP, sldx_FP)
629 (i386_optab): Remove 'd' in the suffixes.
633 * config/tc-i386.c (T_SHORT): Undefine before defining.
637 * config/tc-mips.c: Fix formatting.
641 * config/tc-d10v.c (flag_allow_gstabs_packing): New variable.
642 (md_longopts): New options --gstabs-packing, --no-gstabs-packing.
643 (md_show_usage): Ditto.
644 (md_parse_option): Ditto.
645 (d10v_cleanup): Writes pending instruction only if
646 ! outputting_stabs_line_debug || ! flag_allow_gstabs_packing.
647 Fix compile time warning messages.
649 * doc/c-d10v.texi: Documents new options.
653 * stabs.c (outputting_stabs_line_debug): New variable.
654 (stabs_generate_asm_lineno): Set outputting_stabs_line_debug at
655 function entry and unset at function exit.
657 * read.h (outputting_stabs_line_debug): New extern declaration.
659 * as.c: Include dwarf2dbg.h for definition of dwarf2_finish.
661 * dwarf2dbg.c: Fix compile time warning messages.
665 * config/tc-a29k.c: Fix formatting.
666 * config/tc-alpha.c: Likewise.
667 * config/tc-arm.c: Likewise.
668 * config/tc-cris.c: Likewise.
669 * config/tc-hppa.c: Likewise.
670 * config/tc-i370.c: Likewise.
671 * config/tc-i386.c: Likewise.
672 * config/tc-i860.c: Likewise.
673 * config/tc-i960.c: Likewise.
674 * config/tc-ia64.c: Likewise.
675 * config/tc-m68hc11.c: Likewise.
676 * config/tc-m68k.c: Likewise.
677 * config/tc-m88k.c: Likewise.
678 * config/tc-pj.c: Likewise.
679 * config/tc-ppc.c: Likewise.
680 * config/tc-sh.c: Likewise.
681 * config/tc-sparc.c: Likewise.
682 * config/tc-tahoe.c: Likewise.
683 * config/tc-vax.c: Likewise.
687 * config/tc-mips.c (mips_ip): When calculating offsets,
688 don't accept as constant the difference between the
689 addresses of symbols in two different sections.
691 * config/tc-mips.c (macro_build): Add new 'U' and 'J' operand
693 (validate_mips_insn): Likewise. Also, update 'B' operand
694 specifier to use OP_*_CODE20 constants and delete 'm' operand
696 (mips_ip): Remove 'm' operand specifier, add 'U' and 'J'
697 operand specifiers. Change warning generated by 'B' operand
698 specifier to reflect its new multi-purpose usage.
700 * config/tc-mips.c (mips_set_options): Use ISA_UNKNOWN rather than
701 -1, and update comment.
702 (file_mips_isa): Likewise.
703 (mips_cpu): Use CPU_UNKNOWN rather than -1, and update comment.
704 (ISA_HAS_COPROC_DELAYS, ISA_HAS_64BIT_REGS, gpr_interlocks): Use
705 ISA_* constants rather than hard-coded numbers.
706 (mips_cpu_info): New structure.
707 (mips_cpu_info_table): New table describing CPU and ISA names
709 (mips_cpu_info_from_name, mips_cpu_info_from_isa,
710 mips_cpu_info_from_cpu): New functions.
711 (mips_isa_to_str): New function to get string for ISA name.
712 (mips_cpu_to_str): Convert to use mips_cpu_info_from_cpu, and
714 (md_begin): Redo CPU and ISA selection logic, using
715 mips_cpu_info_from_*. Convert to use ISA_* constants rather
716 than hard-coded numbers.
717 (append_insn, mips_emit_delays, macro, macro2): Convert to use
718 ISA_* constants rather than hard-coded numbers.
719 (mips_ip): Convert to use mips_isa_to_str to get ISA name.
720 (md_longopts): Delete OPTION_NO_MIPS32.
721 (md_parse_option): Convert to use ISA_* constants rather than
722 hard-coded numbers. Make OPTIONS_MIPS32 case treat MIPS32
723 as an ISA. Delete OPTION_NO_MIPS32 case. Convert OPTION_MCPU
724 to use strcasecmp to recognize "default" and to use
725 mips_cpu_info_from_name to get CPU numbers from argument.
726 (md_show_usage): Move -mips32 so it's with the rest of the ISA
727 flags. Change 4Kc, 4Kp and 4Km CPU entries to just be
729 (s_mipsset): Accept ISA value 32.
730 * doc/as.texinfo: Clean up MIPS options summary slightly,
731 remove -no-mips32. Add note about -mips4 and -mips32
732 specifying those ISA levels. Delete -mips32 and -no-mips32
733 cpu flag descriptions.
734 * doc/c-mips.texi: Add -mips32 to list of ISA switches. Clean
735 up the supported CPU switch list, and replace 4Kc, 4Km, and
736 4Kp entries with a single mips32-4k entry. Note that you can
739 * tc-mips.c (ISA_HAS_64BIT_REGS): Add checks for ISA_MIPS5 and
741 (md_longopts, OPTION_MIPS5, OPTION_MIPS64): Add options for
743 (md_parse_option): Add cases for OPTION_MIPS5 and
745 (md_show_usage): Mention -mips5 and -mips64 arguments.
746 (s_mipsset): Add cases for MIPS5 and MIPS64.
747 (mips_cpu_info_table): Add entries for MIPS5 and MIPS64 ISAs
749 * doc/as.texinfo: Mention -mips5 and -mips64 options
751 * doc/c-mips.texi: Likewise. Also update introduction
752 and ".set" usage information.
754 * config/tc-mips.c (md_show_usage): Add "sb1" to the
756 (mips_cpu_info_table): Add SB-1 entries.
757 * doc/c-mips.texi: Add "sb1" to the list of CPUs
758 known to the -mcpu option.
760 * doc/as.texinfo: Correct description of MIPS -mcpu
761 option, by copying some of the text from doc/c-mips.texi.
765 * configure.in (arm-*-rtems*, a29k-*rtems*, h8300-*-rtems*):
767 (sparc*-*-rtemself*, sparc*-*-rtemsaout*): New targets.
768 (sparc*-*-rtems*): Switched from a.out to ELF.
769 * configure: Regenerate.
770 * config.in: Regenerate.
771 * Makefile.in: Regenerate.
772 * aclocal.m4: Regenerate.
773 * po/gas.pot: Regenerate.
777 * config/obj-coff.c (obj_coff_weak): Use S_SET_WEAK if it exists,
778 even in non BFD_ASSEMBLER case.
782 * tc-i386.c (md_assemble): Swap i.disp_relocs when using intel
787 * dwarf2dbg.c: Rewrite from scratch. Queue all debugging output
788 until dwarf2_finish; use relaxation to get cross-fragment offsets;
789 thread multiple subsegments properly; handle multiple code
790 sections properly; emit proper compilation unit info for assembler
793 * as.h (enum _relax_state): Add rs_dwarf2dbg.
794 * dwarf2dbg.h (struct dwarf2_line_info): Remove filename.
795 (dwarf2dbg_estimate_size_before_relax): Declare.
796 (dwarf2dbg_relax_frag, dwarf2dbg_convert_frag): Declare.
797 * write.c: Include dwarf2dbg.h.
798 (cvt_frag_to_fill): Handle rs_dwarf2dbg.
799 (relax_segment): Likewise.
803 * config/tc-sh.c (md_convert_frag) <undefined symbol, conditional
804 jump>: Use as_bad_where instead of as_bad. Tweak error message
805 accordingly. Stabilize frag by updating fix part and resetting
807 <undefined symbol, unconditional jump>: Ditto.
808 (sh_elf_cons): Cast *input_line_pointer to unsigned char when
809 indexing is_end_of_line[].
810 (md_assemble): Initialize size to 0.
811 (md_section_align): Mark parameter seg as unused.
812 (parse_reg): Parse names case-insensitively.
816 * config/obj-aout.h: Fix formatting.
817 * config/obj-bout.h: Likewise.
818 * config/obj-coff.c: Likewise.
819 * config/obj-coff.h: Likewise.
820 * config/obj-elf.h: Likewise.
821 * config/obj-som.h: Likewise.
822 * config/obj-vms.c: Likewise.
823 * config/obj-vms.h: Likewise.
824 * config/tc-h8300.h: Likewise.
825 * config/tc-ns32k.h: Likewise.
826 * config/tc-sparc.h: Likewise.
827 * config/tc-tic54x.h: Likewise.
828 * config/tc-z8k.h: Likewise.
832 * doc/as.1 (COPYING): Mention that the GNU Free Documentation
833 License is present in the sources, but not the output, and
834 also available from the GNU website.
835 (GNU Free Documentation License): Comment out this section.
839 * Makefile.am (CPU_OBJ_VALID): Add case to filter out invalid coff
840 targets. Remove i860 from valid a.out targets.
841 * Makefile.in: Regenerate.
843 * config/tc-cris.c: Include dwarf2dbg.h.
844 (md_pseudo_table): Add .file and .loc.
845 (md_assemble): Call dwarf2_emit_insn if generating ELF.
846 (s_cris_file, s_cris_loc): New.
847 * config/tc-cris.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
848 * Makefile.am: Regenerate dependencies.
849 * Makefile.in: Regenerate.
853 * expr.c (STANDARD_MUL_PRECEDENCE): Correct value.
854 (MRI_MUL_PRECEDENCE): Likewise.
855 (op_rank): Fix a comment typo.
859 * config/tc-m68hc11.c (build_indexed_byte): Print the offset in
861 (get_operand): Fix analysis for movw/movb instructions.
865 * configure.in (xscale-elf): Add target.
866 (xscale-coff): Add target.
867 * configure: Regenerate.
869 * config/tc-arm.c (ARM_EXT_V5E): New ARM architecture
871 (ARM_EXT_XSCALE): New ARM architecture extension.
872 (ARM_LONGMUL): Rename to ARM_EXT_LONGMUL.
873 (ARM_HALFWORD): Rename to ARM_EXT_HALFWORD.
874 (ARM_THUMB): Rename to ARM_EXT_THUMB.
875 (ARM_ARCH_V4): Remove processor from architecture.
876 (ARM_ARCH_3M): New architecutre definition.
877 (ARM_ARCH_V5TE): New architecutre definition.
878 (ARM_ARCH_XSCALE): New architecutre definition.
879 (CPU_DEFAULT): Allow to be defaulted to XScale.
880 (atpcs): New boolean variable.
881 (ldr_flags): Support 'd' flag for double word loads.
882 (str_flags): Support 'd' flag for double word stored.
883 (do_mia): New function.
884 (do_mar): New function.
885 (do_mra): New function.
886 (do_pld): New function.
887 (do_ldrd): New function.
888 (do_blx): New function.
889 (do_bkpt): New function.
890 (do_clz): New function.
891 (do_lstc2): New function.
892 (do_cdp2): New function.
893 (do_t_blx): New function.
894 (do_t_bkpt): New function.
895 (do_smla): New function.
896 (do_smlal): New function.
897 (do_smul): New function.
898 (do_qadd): New function.
899 (do_co_reg2c): New function.
900 (LONGEST_INSN): Redefine to 7.
902 * doc/c-arm.texi: Document -mxscale, -mmarmv5te and -matpcs
903 command line switches.
907 * config/tc-ia64.c (pseudo_func): Add missing initializers.
908 (struct rsrc): Make line unsigned.
909 (gr_values): Add missing initializer.
910 (SLOT_NUM_NOT_SET): Add unsigned cast.
911 (ia64_elf_section_flags, output_vbyte_mem, count_output, dot_radix,
912 dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
913 dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
914 dot_unwentry, dot_altrp, dot_saveg, dot_savef, dot_saveb, dot_savegf,
915 dot_spill, dot_spillreg, dot_spillreg_p, dot_label_state,
916 dot_copy_state, dot_unwabi, dot_personality, dot_proc, dot_body,
917 dot_prologue, dot_endp, dot_regstk, dot_psr, dot_alias, dot_ln,
918 dot_reg_val, dot_entry, dot_mem_offset, ia64_init, mark_resource,
919 md_undefined_symbol, md_apply_fix3, tc_gen_reloc, ia64_md_do_align):
920 Add ATTRIBUTE_UNUSED to unused parameters.
921 (convert_expr_to_ab_reg): Add parens.
922 (convert_expr_to_xy_reg): Add parens. Comment out >= REG_GR test.
923 (dot_prologue): Initialize grsave when declared.
924 (md_pseudo_table): Add missing initializers.
925 (operand_match): Add casts to bfd_vma.
926 (emit_one_bundle): Delete unused local prev. Make required_template
928 (specify_resource): Cast i to unsigned.
929 (note_register_values): Use fprintf_vma.
930 (print_dependency): Likewise.
934 * config/tc-ia64.c (generate_unwind_image): Call record_alignment
935 for unwind info section.
936 (dot_endp): Likewise for unwind section.
938 * config/tc-ia64.c (emit_one_bundle): Pass size of 8 not 4 to
943 * config/tc-sparc.c (md_pseudo_table): Add .file and .loc.
944 (output_insn): Call dwarf2_emit_insn.
945 * config/tc-sparc.h (DWARF2_LINE_MIN_INSN_LENGTH): New.
949 * ehopt.c (eh_frame_code_alignment): New arg `in_seg', update all
950 callers. Don't switch segments. Expect CIE == -1 in .debug_frame.
951 (check_eh_frame): Handle .eh_frame and .debug_frame concurrently.
955 * config/tc-arm.c (md_pseudo_table): Add support for .line and
960 * config/tc-i386.c (md_pseudo_table): Add .file and .loc.
964 * dwarf2dbg.c (dwarf2_gen_line_info): Early out for no line number.
965 * config/obj-elf.h (ECOFF_DEBUGGING) [TC_ALPHA]: Adjust for
966 tri-state definition of alpha_flag_mdebug.
967 * config/tc-alpha.c (alpha_flag_mdebug): Init to -1.
968 (s_alpha_file): Store first .file directive.
970 (md_pseudo_table): Add stabs and stabn.
974 * config/tc-i386.c (md_assemble): Call dwarf2_emit_insn.
978 * as.c (debug_type): Init to DEBUG_UNSPECIFIED.
979 (main): Call dwarf2_finish.
980 * as.h (debug_type): Clarify documentation of the meaning
982 * dwarf2dbg.c (DWARF2_LINE_MIN_INSN_LENGTH): Default to 1.
983 (print_stats): Fix parenthesis problem.
984 (now_subseg_size): New.
985 (dwarf2_finish): Use it. If DEBUG_DWARF2, emit bits for .debug_info.
986 (dwarf2_directive_file): Don't set debug_type.
987 (dwarf2_where): Honor DEBUG_DWARF2 first.
988 (dwarf2_emit_insn): Renamed from dwarf2_generate_asm_lineno;
989 do nothing if not emitting dwarf2 debug info, or no work.
990 * dwarf2dbg.h (dwarf2_emit_insn): Update.
991 * ecoff.c (add_file): Turn on DEBUG_ECOFF only if DEBUG_UNSPECIFIED.
992 (ecoff_new_file): Likewise.
993 * read.c (generate_lineno_debug): Kill ecoff hackery. Update
994 commentary wrt dwarf2.
996 * config/tc-alpha.c (alpha_adjust_symtab_relocs): Add
997 ATTRIBUTE_UNUSED as needed.
998 (emit_insn): Call dwarf2_emit_insn.
1001 (s_alpha_coff_wrapper): Don't handle them.
1002 (md_pseudo_table): Update for .file and .loc.
1003 * config/tc-alpha.h (DWARF2_LINE_MIN_INSN_LENGTH): New.
1005 * config/tc-arm.c (output_inst): Update for dwarf2_emit_insn;
1006 don't protect with debug_type.
1007 * config/tc-hppa.c (md_assemble): Likewise.
1008 * config/tc-m68hc11.c (m68hc11_new_insn): Likewise.
1009 * config/tc-mn10300.c (md_assemble): Likewise.
1010 * config/tc-sh.c (md_assemble): Likewise.
1011 * config/tc-v850.c (md_assemble): Likewise.
1013 * config/tc-arm.c (arm_end_of_source): Remove.
1014 * config/tc-hppa.c (pa_end_of_source): Remove.
1015 * config/tc-m68hc11.c (m68hc11_end_of_source): Remove.
1016 * config/tc-mn10300.c (mn10300_finalize): Remove.
1017 * config/tc-sh.c (sh_finalize): Remove.
1018 * config/tc-v850.c (sh_finalize): Remove.
1020 * config/tc-arm.h (md_end): Remove.
1021 * config/tc-hppa.h (md_end): Remove.
1022 (DWARF2_LINE_MIN_INSN_LENGTH): New.
1023 * config/tc-m68hc11.h (md_end): Remove.
1024 * config/tc-mn10300.h (md_end): Remove.
1025 * config/tc-sh.h (md_end): Remove.
1026 * config/tc-v850.h (md_end): Remove.
1028 * config/tc-ia64.c (emit_one_bundle): Don't protect
1029 dwarf2 bits with debug_type.
1030 (md_assemble): Likewise.
1031 (ia64_end_of_source): Don't call dwarf2_finish.
1035 * config/tc-ia64.c (errata_nop_necessary_p): Abort if general regno
1036 >= 128 instead of > 128. Abort if predicate regno is >= 64 instead of
1041 * config/obj-elf.c (obj_elf_symver): Don't check the missing
1046 * config/tc-tic30.c: Fix formatting.
1047 * config/tc-tic80.c: Likewise.
1048 * config/tc-v850.c: Likewise.
1049 * config/tc-vax.c: Likewise.
1050 * config/tc-w65.c: Likewise.
1051 * config/tc-z8k.c: Likewise.
1055 * config/tc-v850.c: Support dwarf2.
1056 * config/tc-v850.h: Ditto.
1058 * config/tc-v850.c (cons_fix_new_v850): Don't rely on
1059 parse_cons_expression_v850 to initialize hold_cons_reloc.
1063 * tc-ia64.c (struct md): New entries LAST_GROUPS, GROUP_IDX.
1064 (errata_nops_necessary_p): New function.
1065 (emit_one_bundle): Call it. Update the GROUP_IDX field in struct
1070 * config/tc-ia64.c (ia64_target_format): If EF_IA_64_BE not set, then
1071 return little endian bfd formats.
1075 * config/aout_gnu.h: Fix formatting.
1076 * config/atof-vax.c: Likewise.
1077 * config/m68k-parse.h: Likewise.
1078 * config/m88k-opcode.h: Likewise.
1079 * config/obj-elf.c: Likewise.
1080 * config/tc-arm.c: Likewise.
1081 * config/tc-cris.c: Likewise.
1082 * config/tc-i386.c: Likewise.
1083 * config/tc-ia64.c: Likewise.
1084 * config/tc-mn10300.c: Likewise.
1085 * config/te-386bsd.h: Likewise.
1086 * config/te-hppa.h: Likewise.
1087 * config/te-nbsd.h: Likewise.
1088 * config/te-ppcnw.h: Likewise.
1089 * config/te-sparcaout.h: Likewise.
1090 * config/te-tmips.h: Likewise.
1091 * config/vax-inst.h: Likewise.
1092 * config/vms-conf.h: Likewise.
1096 * config/tc-alpha.c (s_alpha_prologue): Preserve visibility bits.
1100 * config/obj-elf.c (elf_frob_symbol): Support
1101 ".symver name,name2@@@nodename".
1102 (elf_frob_file_before_adjust): Likewise.
1104 * doc/as.texinfo: Updated for ".symver name,name2@@@nodename"
1105 and ".symver name,name2@@@nodename".
1110 * config/obj-elf.c (obj_elf_symver): Check missing version
1115 * dwarf2dbg.c (dwarf2_generate_asm_lineno): Use addressT
1116 instead of bfd_vma for non-bfd assemblers.
1120 * itbl-ops.c: Fix comment typos.
1124 * config/tc-ia64.c (struct unw_rec_list): Add slot_frag field.
1125 (struct unwind): Add next_slot_frag field.
1126 (slot_index): New parameters slot_frag and first_frag. Add code
1127 to add in frag sizes when different. Add comments.
1128 (fixup_unw_records): New locals first_frag and last_frag. Pass new
1129 arguments to slot_index.
1130 (emit_one_bundle): Set slot_frag field. Set next_slot_number after
1131 loop end. Set next_slot_frag field.
1135 * doc/as.texinfo (.symver): Updated for versioned symbol
1138 * obj.h (format_ops): Add the frob_file_before_adjust field.
1140 * config/obj-aout.c (aout_format_ops): Set the
1141 frob_file_before_adjust field to 0.
1142 * config/obj-coff.c (coff_format_ops): Likewise.
1143 * config/obj-ecoff.c (ecoff_format_ops): Likewise.
1145 * config/obj-elf.c (obj_elf_symver): Allow duplicated version
1147 (elf_frob_file_before_adjust): New function to remove unneeded
1148 versioned symbols from the symbol table.
1149 (elf_format_ops): Set the frob_file_before_adjust field to
1150 elf_frob_file_before_adjust.
1152 * config/obj-elf.h (obj_frob_file_before_adjust): Defined if
1155 * config/obj-multi.h (obj_frob_file_before_adjust): Defined.
1159 * gas/config/tc-arc.h: Avoid warnings for LITTLE_ENDIAN and
1161 * gas/config/tc-arc.c: Use S_IS_LOCAL to test local symbols.
1162 Fix compile time warning messages.
1166 * stabs.c (generate_asm_file): Increase length of xmalloc'ed
1167 buffer in order to avoid buffer overflows.
1171 * config/tc-ia64.c (md_shortopts, md_parse_option, md_show_usage):
1172 Change M to m for -milp32 or -mlp64 to match gcc.
1173 (dot_endp): Use bytes_per_address instead of 8.
1174 (emit_one_bundle): Use number_to_chars_littleendian instead of
1176 (fix_insn): Likewise.
1177 (ia64_init): New function.
1178 (ia64_target_format): New function.
1179 (md_begin): Set endianness, arch, and machine as appropriate.
1180 * config/tc-ia64.h: (TARGET_BYTES_BIG_ENDIAN, md_number_to_chars):
1181 Make these macros depend on TE_HPUX macro.
1182 (TARGET_FORMAT): Define.
1183 (HOST_SPECIAL_INIT): Define.
1184 * config/te-hpux.h: New file.
1185 * configure.in: Add "ia64-*-hpux*" target to configure.
1186 * configure: Regenerate.
1190 * as.c: Fix formatting.
1191 * dwarf2dbg.c: Likewise.
1192 * input-file.c: Likewise.
1193 * input-file.h: Likewise.
1194 * input-scrub.c: Likewise.
1195 * itbl-ops.c: Likewise.
1196 * listing.c: Likewise.
1197 * macro.h: Likewise.
1198 * messages.c: Likewise.
1200 * subsegs.c: Likewise.
1201 * subsegs.h: Likewise.
1202 * write.c: Likewise.
1206 * doc/as.texinfo: Add GNU Free Documentation License.
1207 * doc/gasp.texi: Add GNU Free Documentation License.
1208 * doc/as.1: Add GNU Free Documentation License.
1212 * config/tc-arm.c: Add include of "dwarf2dbg.h"
1216 * configure.in: Recognise i[3456]86-chaosdev-storm-chaos.
1217 * configure: Regenerate.
1221 * read.c (original_case_string): New global variable.
1222 (read_a_source_file): Copy opcode string into
1223 original_case_string if clobbering the case of the opcode.
1224 * read.h: Export the definition of original_case_string.
1225 * config/tc-arm.c (md_assembler): When parsing a .req
1226 directive use the original opcode string, not the case
1231 * config/tc-mn10300.c (debug_line): Remove this static
1233 (md_assemble): Call dwarf2_generate_asm_lineno instead of
1234 dwarf2_where and dwarf2_gen_line_info.
1238 * config/tc-arm.c (do_msr): Improve error message.
1242 * config/tc-mn10300.c (md_apply_fix3): Use valuep if fully resolved
1243 or pc-relative, else use fx_offset.
1247 * config/tc-ia64.c (struct md): New field tag_fixups.
1248 (ia64_flush_insns): Handle tag_fixups. Error if dangling
1249 qualifying predicate.
1250 (emit_one_bundle): Delete spurious multiplication by one. Handle
1252 (ia64_start_line): Error if dangling qualifying predicate.
1253 (defining_tag): New static variable.
1254 (ia64_unrecognized_line, case '['): Parse tags.
1255 (ia64_frob_label): Create tag_fixups.
1256 (md_assemble): Reset md.qp.X_op after using it.
1260 * config/tc-sh.c (md_apply_fix [BFD_RELOC_SH_PCDISP12BY2]): Allow 4094.
1264 * tc-ia64.c (extra_goodness): Only prefer F in slot 1 and B in slot 2.
1268 * expr.c: Fix formatting.
1269 * flonum-copy.c: Likewise.
1270 * flonum.h: Likewise.
1276 * as.h (OPTION_MD_BASE): Bump to 190.
1277 * as.c (parse_args) <std_longopts>: Add comment about the need to
1278 check OPTION_MD_BASE in as.h.
1280 * config/tc-sh.c (md_apply_fix): For ELF, do not "adjust back" VAL
1285 * configure.in (emulations): Add m68hc12.
1286 * configure: Regenerate.
1287 * po/gas.pot: Regenerate.
1291 * config/tc-arm.c (psrs): Remove lowercase versions of spsr* and
1293 (arm_psr_parse): Handle lowercase CPSR and SPSR.
1297 * dwarf2out.c (dwarf2_generate_asm_lineno): New function: Generate
1298 a DWARF2 line number information sequence.
1300 *dwarf2out.h: Add prototype for dwarf2_generate_asm_lineno.
1302 * read.c (generate_lineno_debug): Update comment describing why
1303 DWARF2 line number debug information is not generated
1304 automatically by this function.
1306 * doc/as.texinfo: Note that --gdwarf2 only works on some targets,
1309 * config/tc-arm.h (md_end): Define.
1310 (DWARF2_LINE_MIN_INSN_LENGTH): Define.
1312 * config/tc-arm.c (output_inst): Call dwarf2_generate_asm_lineno
1313 if generating DWARF2 line numbers.
1314 (arm_end_of_source): New function. Call dwarf2_finish if
1317 * config/tc-hppa.c (md_assemble): Use dwarf2_generate_asm_lineno.
1318 * config/tc-m68hc11.c (m68hc11_new_insn): Use dwarf2_generate_asm_lineno.
1319 * config/tc-sh.c (md_assemble): Use dwarf2_generate_asm_lineno.
1323 * tc-i386.c: Fix prototype declarations for functions taking no
1328 * tc-i386.c (i386_operand_modifier): Remove.
1329 (build_displacement_string): Remove.
1330 (i386_parse_seg): Remove.
1331 (i386_intel_memory_operand): Remove.
1332 (i386_intel_operand): Re-write using recursive descent parser based
1333 on MASM documentation.
1334 (struct intel_parser_s): New structure.
1335 (intel_parser): New static variable.
1336 (struct intel_token): New structure.
1337 (cur_token, prev_token): New static variables.
1350 (intel_match_token): New function.
1351 (intel_get_token): New function.
1352 (intel_putback_token): New function.
1353 (intel_expr): New function.
1354 (intel_e05): New function.
1355 (intel_e05_1): New function.
1356 (intel_e06): New function.
1357 (intel_e06_1): New function.
1358 (intel_e09): New function.
1359 (intel_e09_1): New function.
1360 (intel_e10): New function.
1361 (intel_e10_1): New function.
1362 (intel_e11): New function.
1366 * config/tc-sparc.c (sparc_ip): Fix a bug which caused v9_arg_p
1367 instructions to loose any special insn->architecture mask.
1369 * config/tc-sparc.c (v9a_asr_table): Add v9b ASRs.
1370 (sparc_md_end, sparc_arch_types, sparc_arch,
1371 sparc_elf_final_processing): Handle v8plusb and v9b architectures.
1372 (sparc_ip): Handle siam mode operands. Support v9b ASRs (and
1373 request v9b architecture if they are used).
1377 * config/tc-m68k.c: Fix the previous misapplied patch.
1381 * config/tc-m68k.h (RELAX_RELOC_*): New definitions for both
1382 BFD_ASSEMBLER and !BFD_ASSEMBLER.
1383 * config/tc-m68k.c (md_convert_frag_1): Use them instead of
1388 * debug.c: Fix formatting.
1389 * depend.c: Likewise.
1390 * dwarf2dbg.c: Likewise.
1391 * dwarf2dbg.h: Likewise.
1392 * ecoff.c: Likewise.
1395 * flonum-konst.c: Likewise.
1396 * frags.h: Likewise.
1400 * as.c: New option OPTION_TARGET_HELP. Prints all target specific
1402 * doc/as.texinfo: Added notes about this new option.
1406 * config/tc-sh.c (JREG): Remove.
1407 (md_convert_frag): Remove #if 0:d code using JREG.
1411 * config/tc-i386.c (i386_operand_modifier): Only match
1412 modifiers SHORT and FLAT if they are followed by a space.
1413 (parse_register): When `allow_naked_reg' is set, do not confuse
1414 identifiers that start with a register name with a register.
1418 * app.c: Fix formatting.
1421 * bit_fix.h: Likewise.
1428 * config/obj-elf.c (elf_frob_symbol): Revert 2000-10-07 change.
1432 * config/tc-hppa.c (md_apply_fix): Remove plainly wrong assert.
1433 Re-arrange function a little and improve error message.
1435 * write.c (write_relocs): Fix a comment.
1437 * config/obj-elf.c (elf_frob_symbol): Make section syms global on
1442 * config/tc-ia64.c (resources_match): Handle IA64_RS_PRr.
1446 * config/tc-i386.c: Delete some useless comments, reformat others.
1448 * config/tc-i386.h (TC_FIX_ADJUSTABLE): Add check to cover
1449 non-global syms in linkonce sections.
1453 * config/tc-ia64.c (operand_match): Don't use // style comments.
1454 * config/tc-i370.c: Likewise.
1458 Changes to handle varying register prefix and user symbol prefix.
1459 * config/tc-cris.c (SYNTAX_RELAX_REG_PREFIX,
1460 SYNTAX_ENFORCE_REG_PREFIX, SYNTAX_USER_SYM_LEADING_UNDERSCORE,
1461 SYNTAX_USER_SYM_NO_LEADING_UNDERSCORE, REGISTER_PREFIX_CHAR): New.
1462 (s_syntax, cris_force_reg_prefix, cris_relax_reg_prefix,
1463 cris_sym_leading_underscore, cris_sym_no_leading_underscore): New.
1464 (demand_register_prefix): New variable.
1465 (md_pseudo_table): New pseudo ".syntax".
1466 (md_longopts): New options --no-underscore and --underscore.
1467 (cris_target_format): Return elf32-us-cris or elf32-cris depending
1468 on symbols_have_leading_underscore.
1469 (get_gen_reg): Accept or require REGISTER_PREFIX_CHAR.
1470 (get_spec_reg): Ditto.
1471 (cris_number_to_imm) <case BFD_RELOC_VTABLE_ENTRY>: Remove FIXME.
1473 (md_parse_option) <case 'h' 'H'>: Deprecate; add reference to
1475 <case OPTION_NO_US, case OPTION_US>: New.
1476 (md_show_usage): Be brief and reformat to match continuation of
1478 * po/gas.pot: Regenerate.
1482 * config/tc-hppa.c (hppa_force_relocation): If OBJ_SOM, don't
1483 force relocs for 12 bit branches.
1484 (md_apply_fix): Similarly, adjust logic here.
1488 * config/tc-hppa.c (md_apply_fix): Add fmt assertion. Don't
1489 adjust for external and weak syms as we will use a reloc. Allow
1490 for +8 offset when calculating limits of branches.
1491 (hppa_fix_adjustable): Undo 2000-09-23 change.
1492 (hppa_force_relocation): Likewise. Add fx_addsy assertion.
1493 Correct distance calculation.
1494 (tc_gen_reloc): Print the file name and line number if we can't
1498 * config/tc-hppa.c (nonzero_dibits): Define.
1499 (arg_reloc_stub_needed): Check each arg and return value
1500 separately for zero case.
1501 (pa_align): Declare argument `bytes'.
1505 * config/tc-cris.c: Fix formatting.
1506 * config/tc-d10v.h: Likewise.
1507 * config/tc-d30v.c: Likewise.
1508 * config/tc-d30v.h: Likewise.
1509 * config/tc-fr30.c: Likewise.
1510 * config/tc-fr30.h: Likewise.
1511 * config/tc-m68k.c: Likewise.
1512 * config/tc-m68k.h: Likewise.
1513 * config/tc-pj.h: Likewise.
1514 * config/tc-ppc.c: Likewise.
1515 * config/tc-ppc.h: Likewise.
1516 * config/tc-sh.c: Likewise.
1517 * config/tc-sh.h: Likewise.
1518 * config/tc-sparc.c: Likewise.
1519 * config/tc-v850.h: Likewise.
1520 * config/tc-vax.h: Likewise.
1521 * config/tc-w65.h: Likewise.
1522 * config/tc-z8k.h: Likewise.
1526 * config/tc-hppa.c (hppa_fix_adjustable): Do the external and weak
1527 checks only for ELF.
1528 (hppa_force_relocation): Likewise.
1532 * config/tc-ia64.c (dv_sem): Add "stop".
1533 (specify_resource, case IA64_RS_PR): Only handles regs 1 to 15 now.
1534 (specify_resource, case IA64_RS_PRr): New for regs 16 to 62.
1535 (specify_resource, case IA64_RS_PR63): Reorder (note == 7) test to
1537 (mark_resources): Check IA64_RS_PRr.
1541 * config/tc-m68k.c (md_relax_table, m68k_ip, md_convert_frag_1,
1542 md_estimate_size_before_relax): Redesign and clean up the
1543 relaxation mechanism.
1547 * config/tc-ns32k.c: Fix formatting.
1548 * config/tc-ns32k.h: Likewise.
1552 * config/tc-m32r.c: Fix formatting.
1553 * config/tc-m32r.h: Likewise.
1554 * config/tc-m68851.h: Likewise.
1555 * config/tc-m68hc11.c: Likewise.
1556 * config/tc-m68hc11.h: Likewise.
1557 * config/tc-m88k.c: Likewise.
1558 * config/tc-mcore.c: Likewise.
1559 * config/tc-mcore.h: Likewise.
1560 * config/tc-mips.c: Likewise.
1561 * config/tc-mips.h: Likewise.
1562 * config/tc-mn10200.h: Likewise.
1563 * config/tc-mn10300.h: Likewise.
1564 * config/tc-tahoe.c: Likewise.
1565 * config/tc-tahoe.h: Likewise.
1569 * config/tc-vax.c (synthetic_votstrs): Remove jbssi and jbcci.
1570 Likewise in relaxation description comments.
1574 * config/tc-hppa.h (TC_FORCE_RELOCATION_SECTION): Allow
1575 subtraction of two syms without emitting a relocation.
1578 * config/tc-hppa.c (hppa_force_relocation): Force relocations for
1579 global or weak symbols.
1583 * config/tc-h8300.h: Fix formatting.
1584 * config/tc-h8500.c: Likewise.
1585 * config/tc-h8500.h: Likewise.
1586 * config/tc-hppa.h: Likewise.
1587 * config/tc-i370.h: Likewise.
1588 * config/tc-i386.h: Likewise.
1589 * config/tc-i860.c: Likewise.
1590 * config/tc-i860.h: Likewise.
1591 * config/tc-i960.h: Likewise.
1592 * config/tc-ia64.c: Likewise.
1593 * config/tc-ia64.h: Likewise.
1597 * config/tc-a29k.c: Fix formatting.
1598 * config/tc-alpha.c: Likewise.
1599 * config/tc-arc.c: Likewise.
1600 * config/tc-arc.h: Likewise.
1601 * config/tc-arm.c: Likewise.
1602 * config/tc-arm.h: Likewise.
1603 * config/tc-avr.c: Likewise.
1604 * config/tc-avr.h: Likewise.
1605 * config/tc-tic30.c: Likewise.
1606 * config/tc-tic30.h: Likewise.
1607 * config/tc-tic54x.c: Likewise.
1608 * config/tc-tic54x.h: Likewise.
1609 * config/tc-tic80.c: Likewise.
1610 * config/tc-tic80.h: Likewise.
1614 * config/tc-ia64.c (specify_resource): For PR%/PR63, note types of
1615 parallel comparisons for later use.
1616 (struct rsrc): Add parallel comparison type.
1617 (resources_match): Skip special cases of PR usage (non-conflicting
1622 * config/obj-ecoff.c: Fix formatting.
1623 * config/obj-elf.c: Likewise.
1624 * config/obj-elf.h: Likewise.
1625 * config/obj-evax.h: Likewise.
1626 * config/obj-generic.h: Likewise.
1627 * config/obj-hp300.c: Likewise.
1628 * config/obj-hp300.h: Likewise.
1629 * config/obj-ieee.h: Likewise.
1630 * config/obj-vms.c: Likewise.
1631 * config/obj-vms.h: Likewise.
1635 * config/tc-mips.c (md_begin): Recognize 4Kc, 4Km and 4Kp processors.
1636 (md_parse_option): Ditto.
1637 (md_longopts): Add -mips32 option.
1638 (md_show_usage): Document new options.
1639 (mips_ip): Assemble sdbbp 20 bit 'm' args for MIPS32.
1640 (mips_ip): Assemble mfc0 with a sub-selection code.
1641 (validate_mips_insn): Handle 'H' (OP_*_SEL) and 'm' (OP_*_CODE20).
1642 (mips_cpu_to_str): New function.
1643 (mips_ip): Use mips_cpu_to_str instead of printing numeric cpu value.
1644 Use CPU_* defines instead of hardcoded numbers.
1646 * doc/as.texinfo: Document new options.
1647 * doc/c-mips.texi: Ditto.
1651 * as.h: Fix formatting.
1652 * asintl.h: Likewise.
1653 * bit_fix.h: Likewise.
1654 * config/obj-aout.c: Likewise.
1655 * config/obj-aout.h: Likewise.
1656 * config/obj-bout.c: Likewise.
1657 * config/obj-bout.h: Likewise.
1658 * config/obj-coff.c: Likewise.
1659 * config/obj-coff.h: Likewise.
1660 * dwarf2dbg.h: Likewise.
1662 * flonum.h: Likewise.
1663 * frags.h: Likewise.
1664 * itbl-ops.h: Likewise.
1665 * macro.h: Likewise.
1668 * struc-symbol.h: Likewise.
1669 * subsegs.h: Likewise.
1670 * symbols.h: Likewise.
1672 * write.h: Likewise.
1676 * bignum-copy.c: Fix formatting.
1677 * config/tc-i370.c: Likewise.
1678 * config/tc-i960.c: Likewise.
1679 * config/tc-m68k.c: Likewise.
1680 * ehopt.c: Likewise.
1681 * flonum-copy.c: Likewise.
1682 * flonum-konst.c: Likewise.
1683 * flonum-mult.c: Likewise.
1684 * literal.c: Likewise.
1687 * stabs.c: Likewise.
1688 * subsegs.c: Likewise.
1692 * configure.in (arm*-*-uclinux*): New target.
1693 * configure: Regenerate.
1697 * input-file.c: Fix formatting.
1698 * itbl-ops.c: Likewise.
1699 * messages.c: Likewise.
1703 * config/tc-arm.c (md_apply_fix3): Correct handling of ADRL when
1708 * configure.in (AC_ISC_POSIX): Put after AC_CANONICAL_SYSTEM.
1709 * configure: Rebuild.
1713 * atof-generic.c: Fix formatting.
1714 * config/tc-mips.c: Likewise.
1715 * config/tc-vax.c: Likewise.
1716 * input-scrub.c: Likewise.
1720 * config/tc-sh.h (TARGET_FORMAT): Use sh-linux targets.
1721 * configure.in (sh-*-linux*): Added.
1722 * configure: Rebuilt.
1726 * config/tc-hppa.c: Fix formatting.
1728 * ecoff.c: Fix formatting.
1732 * configure: Rebuilt with new libtool.m4.
1736 * cgen.c: Fix formatting.
1737 * config/tc-ia64.c: Likewise.
1741 * aclocal.m4: Regenerate.
1742 * config.in: Regenerate.
1743 * configure: Regenerate.
1744 * po/gas.pot: Regenerate.
1748 * config/tc-cris.c: Correct comment typos.
1752 * config/tc-mn10300.c: Cleanup.
1753 (md_pcrel_from): Enable.
1757 * expr.c (operand): Fix a comment typo.
1758 * write.c (write_relocs): Fix a signed/unsigned warning.
1760 * config/tc-hppa.c (fudge_reg_expressions): New
1761 (hppa_force_reg_syms_absolute): New.
1762 (pa_equ): Allow reg_section expressions.
1763 * config/tc-hppa.c (md_optimize_expr): Define.
1764 (hppa_force_reg_syms_absolute): Prototype.
1766 * config/tc-hppa.c (pa_11_fp_reg_struct): Delete.
1767 (pa_parse_number): Pass in arg to select fp reg parsing.
1768 Return 1 to indicate format checks pass. If strict, then only
1769 accept a register or register symbol. Return value in...
1770 (pa_number): New static for pa_parse_number.
1771 (FP_REG_BASE): Define.
1772 (FP_REG_RSEL): Define.
1773 (pre_defined_registers): Apply FP_REG_BASE and FP_REG_RSEL as
1774 appropriate. White space changes.
1775 (need_pa11_opcode): Don't bother passing any params, get them from
1777 (pa_ip): Modify all calls to pa_parse_number and need_pa11_opcode.
1778 Remove extraneous check in case 'Q'.
1779 (pa_equ): Modify call to pa_parse_number to do strict parsing. If
1780 reg, set section of resulting symbol to reg_section.
1781 (pa_parse_space_stmt): Modify call to pa_parse_number.
1782 (pa_space): Likewise.
1784 * config/tc-hppa.c: (md_apply_fix): Handle vtable relocs.
1785 (hppa_force_relocation): Handle vtable relocs.
1786 (pa_vtable_entry): New.
1787 (pa_vtable_inherit): New.
1788 (md_pseudo_table): Add entries for vtable pseudos.
1789 (hppa_fix_adjustable): Reject reduction of R_PARISC_GNU_VTINHERIT
1790 and R_PARISC_GNU_VTENTRY relocs. Reject reduction of relocs
1792 (tc_gen_reloc): Remove ELF_ARG_RELOC_INSN code.
1793 (pa_type_args): Don't call symbol_get_bfdsym multiple times.
1794 Set STT_PARISC_MILLICODE for OBJ_ELF when encountering a
1796 * config/obj-elf.c (obj_elf_type): Allow md_elf_symbol_type to
1797 specify a symbol type.
1799 * config/tc-hppa.h: Reorganize file a little, grouping OBJ_ELF
1800 dependent things together.
1801 (md_elf_symbol_type): Define.
1803 * config/tc-hppa.c (fix_new_hppa): Elide "$PIC_pcrel$0" pseudo
1805 * config/tc-hppa.h (tc_frob_symbol): Elide "$PIC_pcrel$0" here too.
1807 * config/obj-elf.h (obj_elf_vtable_inherit): Declare.
1808 (obj_elf_vtable_entry): Declare.
1810 * config/obj-elf.c (obj_elf_vtable_inherit): Return struct fix *
1811 and export function.
1812 (obj_elf_vtable_entry): Similarly.
1813 (elf_pseudo_table): Fix the damage with a cast.
1817 * config/tc-ia64.c (emit_one_bundle): Stop collecting insns
1818 for template selection when a label is needed.
1822 * config/tc-ia64.c: Fix formatting.
1826 * configure.in: Increase version number to 2.10.91.
1827 * configure: Regenerate.
1828 * aclocal.m4: Regenerate.
1829 * config.in: Regenerate.
1830 * po/gas.pot: Regenerate.
1831 * Makefile.in: Regenerate.
1835 * config/tc-sh.h [OBJ_ELF] (TC_FIX_ADJUSTABLE): Define.
1836 * config/tc-sh.c (md_apply_fix): Map 32-bit relocations that
1837 become PC-relative to BFD_RELOC_32_PCREL. Reject 16- or 8-bit
1839 (sh_obj_adjustable): Return 1 for PC-relative offsets used in
1844 * config/tc-sh.h (DIFF_EXPR_OK, GLOBAL_OFFSET_TABLE_NAME,
1845 TC_RELOC_GLOBAL_OFFSET_TABLE, TC_RELOC_RTSYM_LOC_FIXUP): Define.
1846 * config/tc-sh.c (sh_elf_cons, sh_elf_suffix): New functions.
1847 [OBJ_ELF] (md_pseudo_table) <long, int, word, short>: Use them.
1848 (GOT_symbol): New variable.
1849 (md_undefined_symbol): Set it.
1853 * config/tc-ia64.c (match): Don't inline.
1854 (extra_goodness): New.
1855 (md_begin): Prefer nop.f and nop.b for best_template.
1859 * as.c: Fix formatting.
1861 * frags.c: Likewise.
1862 * macro.c: Likewise.
1866 * config/tc-mn10300.c: Cleanup and fix warnings.
1867 (md_pseudo_table): Add initializers.
1868 (md_show_usage): Cleanup.
1869 (md_parse_option): Fix warnings.
1870 (md_undefined_symbol): Fix warnings.
1871 (md_conver_frag): Fix warnings.
1872 (tc_gen_reloc): Fix warnings.
1873 (md_apply_fix3): Fix warnings.
1874 (check_operand): Fix warnings.
1878 * acinclude.m4: Include libtool and gettext macros from the
1880 * aclocal.m4, configure: Rebuilt.
1884 * config/tc-ppc.c (md_parse_option): Recognize -m405.
1888 * listing.c: Fix formatting.
1892 * app.c: Fix a comment typo. Fix formatting.
1896 * tc-vax.c (md_convert_frag): Correctly calculate the pc relative
1897 offset of the target destination for jmp instructions.
1898 (md_assemble): Change mode to VAX_ABSOLUTE_MODE as per comments.
1902 * NEWS: Mention support for CRIS.
1906 * config/tc-avr.h (TC_IMPLICIT_LCOMM_ALIGNMENT): New macros.
1907 Sets `.lcomm' alignment to zero.
1911 * config/tc-i386.h <OBJ_MAYBE_ELF, OBJ_MAYBE_COFF>
1912 (TC_FIX_ADJUSTABLE): Define.
1916 * config/tc-ia64.c (output_unw_records): Set U & E flags only if
1917 unwind.personality_routine is set.
1921 * write.c (TC_FIX_ADJUSTABLE): Remove the duplicate.
1925 * config/tc-i386.h (TC_FIX_ADJUSTABLE): Do *NOT* define if target
1930 * config.in (STRICTCOFF): New for strict COFF.
1932 * configure.in: Define STRICTCOFF for i386-*-msdosdjgpp*,
1933 i386-*-go32* and i386-go32-rtems*.
1934 * configure: Rebuilt.
1936 * config/obj-coff.c (obj_coff_endef): Follow the historical
1937 behavior if STRICTCOFF is not defined.
1939 * doc/internals.texi: Document STRICTCOFF.
1943 * write.c (TC_FIX_ADJUSTABLE): Define to 1, if not defined.
1944 (fixup_segment) Use it instead of TC_DONT_FIX_NON_ADJUSTABLE.
1945 * config/tc-i386.h (TC_DONT_FIX_NON_ADJUSTABLE): Remove.
1946 <OBJ_ELF, OBJ_COFF, TE_PE> (TC_FIX_ADJUSTABLE): Define.
1947 * config/tc-arm.h (TC_DONT_FIX_NON_ADJUSTABLE): Remove.
1948 <OBJ_ELF> (TC_FIX_ADJUSTABLE): Define.
1949 * config/tc-i960.h, config/tc-m68k.h, config/tc-v850.h:
1954 * config/tc-mn10300.c: (md_apply_fix): New function.
1955 (mn10300_force_relocation): New function.
1956 (mn10300_fix_adjustable): New function.
1958 * config/tc-mn10300.h: (TC_FORCE_RELOCATION): Define.
1959 (TC_HANDLES_FX_DONE): Define.
1960 (obj_fix_adjustable): Define.
1961 (MD_APPLY_FIX3): Define.
1962 (TC_LINKRELAX_FIXUP): Define.
1964 * write.c: (TC_LINKRELAX_FIXUP): Define if not
1966 (fixup_segment): Use TC_LINKRELAX_FIXUP.
1968 * doc/internals.texi: Document TC_LINKRELAX_FIXUP.
1972 * config/tc-i860.c (md_apply_fix3): Do not insert the immediate
1973 if the fixup resulted in a relocation.
1977 * config/tc-arm.c (decode_shift): Replace as_tsktsk with as_warn.
1978 Make reference to first element of shift_names explicit.
1982 * write.c (fixup_segment) [TC_DONT_FIX_NON_ADJUSTABLE]: Use
1983 obj_fix_adjustable() and tc_fix_adjustable() to tell whether to
1984 add a symbol's address. Removed all target-specific #ifdefs that
1985 used to accomplished the same.
1986 * config/tc-v850.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
1987 * config/tc-m68k.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
1988 * config/tc-arm.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
1989 * config/tc-i960.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
1990 * config/tc-i386.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
1994 * dwarf2dbg.c: Fix formatting.
1998 * config/tc-arm.c (decode_shift): Allow illegal shifts by zero
1999 to be recoded as logical shift lefts by zero.
2003 * config/tc-ia64.c (specify_resource, case IA64_RS_GR): Handle
2004 postincrement modified registers. Handle IA64_OPND_R3_2 addl
2006 (note_register_values): Handle IA64_OPND_R3_2 operands.
2010 * config/tc-i860.c (md_operand): Silly typo fixed.
2014 * config/tc-arm.c (struct asm_shift): Delete.
2016 (enum asm_shift_index): New.
2017 (struct asm_shift_properties): New.
2018 (struct asm_shift_name): New.
2019 (shift_properties[]); New.
2020 (shift_names[]); New.
2022 (decode_shift): Use new structures.
2023 Issue a warning is "ROR #0" is used.
2024 Issue a warning if "ASR #0" or "LSR #0" is used.
2026 (md_begin): Initialise arm_shift_hsh table from new
2027 asm_shift_name array.
2031 * config/tc-sparc.c: Kill all warnings.
2032 (md_parse_option): Set -32/-64 for -xarch=, allow all -A archs
2034 (md_show_usage): Update usage text.
2038 * config/tc-arm.c (do_bx): Warn about "bx px" not being very
2043 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Defined.
2045 * config/tc-sh.c (md_assemble): Changed so debug_type
2046 test performed for ppi_assemble
2047 * config/tc-sh.c: Included dwarf2dbg.h.
2048 (debug_line): Defined.
2049 (md_assemble): Generates dwarf2 line info.
2050 (sh_finalize): New function. Finalize dwarf2 info.
2051 (assemble_ppi): Returns size of code generated.
2052 (build_Mytes): Returns size of code generated.
2053 (md_pseudo_table): Added "file" and "loc" psuedo ops.
2054 * config/tc-sh.h (md_end): Defined.
2055 (sh_finalize): Declared.
2059 * config/tc-sh.c (md_apply_fix) [BFD_RELOC_32, BFD_RELOC_16]: Use
2064 * config/tc-arm.c (do_bx): Allow "bx pc".
2068 * config/tc-ia64.c (md_longopts): Add -mconstant-gp and -mauto-pic.
2069 (md_parse_option): Add OPTION_MCONSTANT_GP and OPTION_MAUTO_PIC.
2070 (md_begin): Change assignment to md.flag to OR in the new bit.
2074 * config/obj-coff.c (obj_coff_endef) [BFD_ASSEMBLER]: Set the debug
2075 flag for storage types C_ARG, C_REGPARM, C_FIELD, C_MOS, C_MOE,
2080 * NEWS: Mention i860 support.
2084 * config/tc-mn10300.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
2088 * doc/c-arm.texi (ARM Directives): Fix warnings from makeinfo.
2092 * doc/c-i860.texi (Opcodes for i860): Remove braces from @item
2097 * expr.c: Fix formatting.
2098 * config/obj-bout.c: Likewise.
2102 * doc/c-i860.texi: Flesh out the i860 section more.
2106 * symbols.c: Fix formatting.
2111 * tc-i386.c (md_assemble): Skip suffix check if the opcode modifier
2112 has the IgnoreSize bit set.
2117 * configure.in: Use elf on Unixware 7 (i586-sco-sysv5uw7.1.0)
2118 * configure: Regenerate.
2122 * hash.c: Fix formatting.
2127 * config/tc-i860.h: Rework completely for BFD_ASSEMBLER.
2128 (i860_fix_info): New enum.
2129 (MD_APPLY_FIX3): Define.
2130 (WORKING_DOT_WORD): Define.
2131 (TC_HANDLES_FX_DONE): Define.
2132 (DIFF_EXPR_OK): Define.
2133 (LISTING_HEADER): Define.
2134 (TARGET_FORMAT): Select target format based on endian flag.
2135 (TARGET_BYTES_BIG_ENDIAN): Default to little endian.
2136 (target_big_endian): Add external declaration.
2138 * config/tc-i860.c: All existing code reworked completely. Other
2139 new code shown below.
2140 (SYNTAX_SVR4): Define.
2141 (target_warn_expand): New variable.
2142 (md_shortopts): Declare and define (-Qy, -Qn, and -V options).
2143 (md_longopts): Declare and define with new options (-EL, -EB,
2145 (md_show_usage): New function.
2146 (md_operand): New function.
2147 (obtain_reloc_for_imm16): New function.
2148 (md_apply_fix3): New function.
2149 (tc_gen_reloc): New function.
2153 * config/tc-m68hc11.c (build_jump_insn): Make sure the
2154 2 bytes of the jump address are in the same frag.
2155 (find): Accept 68hc12 register indirect modes.
2157 * NEWS: Mention 68HC11 & 68HC12 support.
2161 * config/tc-ia64.c (unwind): Add prologue_mask member.
2162 (dot_vframe): Elide psp_gr record if it overlaps prologue_gr.
2163 (dot_save): Likewise for pfs_gr, rp_gr, and preds_gr.
2164 (dot_body): Clear unwind.prologue_mask.
2165 (dot_prologue): Set it. Accept a register second argument.
2169 * config/atof-ieee.c: Fix formatting.
2170 * config/atof-tahoe.c: Likewise.
2174 * config/tc-sparc.c (md_begin): Fix typo in recent formatting
2177 * doc/as.texinfo (Pseudo Ops): Update to include descriptions
2178 of .popsection, .previous, .pushsection, .subsection,
2179 .version, .vtable_entry, .vtable_inherit and .weak.
2183 * config/tc-cris.c: Fix formatting.
2184 * config/tc-i386.c: Likewise.
2185 * config/tc-sparc.c (sparc_ip): Simplify the code.
2189 * config/tc-cris.c: Rearrange code for readability.
2190 * config/tc-d10v.c: Fix formatting.
2191 * config/tc-m32r.c: Likewise.
2192 * config/tc-sparc.c: Likewise.
2196 * config/tc-ia64.c (emit_one_bundle): Call ia64_free_opcode
2197 before ia64_find_opcode.
2198 (md_assemble): Likewise.
2202 * config/tc-arm.c (do_mrs): Fix skip of 'cpsr_all' flag.
2203 Undo some formatting fixes.
2207 * config/obj-som.c: Fix formatting.
2208 * config/obj-ieee.c: Likewise.
2209 * config/tc-arm.c: Likewise.
2210 * config/tc-v850.c: Likewise.
2214 * doc/c-m68k.texi (section M680x0 Options): Turn into a table
2215 index by command line option.
2219 * doc/c-m68k.texi (@cindex @samp{--pcrel}): Rewrite option description.
2220 (@node M68K-Branch): Rewrite to match the reality.
2224 * doc/c-i860.texi: New file.
2225 * doc/Makefile.am (CPU_DOCS): Add c-i860.texi.
2226 * doc/Makefile.in: Regenerate.
2227 * doc/all.texi: Add I860 as relevant architecture.
2228 * doc/as.texinfo: Include i860 dependent file c-i860.texi.
2232 * config/tc-d30v.c: Fix formatting.
2236 * config/tc-m68k.c (flag_keep_pcrel, OPTION_PCREL): Add --pcrel option.
2237 (md_convert_frag_1, md_estimate_size_before_relax): When making DBcc
2238 long emit a long branch if available instead of an absolute jump, never
2239 emit absolute jumps for anything with --pcrel.
2241 * doc/c-m68k.texi: Document new command line option.
2245 * config/tc-avr.c: Use PARAMS macro in function declarations.
2246 Don't declare md_pcrel_from_section (already in tc-avr.h).
2247 (avr_operands): Use AVR_UNDEF_P and AVR_SKIP_P macros.
2248 (avr_operand): Don't set (unsigned) op_mask to -1.
2252 * configure.in: Add bits for i860-stardent-{sysv4, elf}*.
2253 * configure: Regenerated.
2254 * config/obj-elf.c (obj_elf_type): Recognize a fifth type
2255 of operand to the .type directive (.e.g, "type").
2259 * as.h (warn_comment, found_comment, found_comment_file): Declare.
2260 * app.c (do_scrub_chars): Record where first comment found.
2261 * read.c (read_a_source_file): Init found_comment on entry, and
2262 notify whether comments found on exit.
2263 * config/tc-hppa.c (md_shortopts): Add "c".
2264 (md_longopts): Add warn-comment.
2265 (md_parse_option): Handle it.
2266 (md_show_usage): Show available options.
2267 * config/tc-hppa.h (WARN_COMMENTS): Define if TE_LINUX
2271 * config/tc-mn10300.c (md_convert_frag): Fix printfs.
2272 (tc_gen_reloc): Add cast when assigning bfd_abs_symbol to
2274 (md_estimate_size_before_relax): Don't fall off end of function.
2278 * config/tc-avr.c: Fix formatting.
2279 * config/tc-ns32k.c: Likewise.
2283 * config/tc-d10v.c (find_opcode): Remove extraneous `='.
2287 * config/tc-d10v.c: Fix formatting.
2288 * config/tc-z8k.c: Likewise.
2289 * config/tc-sparc.c: Likewise.
2293 * cgen.c (queue_fixup): Declare opinfo.
2294 (gas_cgen_parse_operand): Mark unused parameters with ATTRIBUTE_UNUSED.
2295 (gas_cgen_md_operand): Ditto.
2296 (gas_cgen_md_apply_fix3): Ditto.
2300 * config/obj-coff.c (obj_frob_symbol): Don't merge
2301 labels. Don't merge if the symbol isn't constant. Return
2302 immediately if a symbol is merged.
2306 * frags.c (frag_align): Correct absolute section alignment.
2310 * config/obj-coff.c (obj_frob_symbol): revert previous change,
2311 it breaks linking against DLLs.
2315 * configure.in: Add CRIS support.
2316 * configure: Regenerate.
2317 * Makefile.am: (CPU_TYPES): Add cris.
2318 (CPU_OBJ_VALID) [aout]: Add cris.
2319 (MULTI_CPU_TYPES): Add cris.
2320 (MULTI_CPU_OBJ_VALID) [aout]: Add cris.
2321 [coff]: Only i386 and mips are valid.
2322 (TARGET_CPU_CFILES): Add config/tc-cris.c.
2323 (TARGET_CPU_HFILES): Add config/tc-cris.h.
2324 (MULTI_CFILES): Add config/e-crisaout.c and config/e-criself.c.
2325 Regenerate dependencies.
2326 * Makefile.in: Regenerate.
2327 * aclocal.m4: Regenerate.
2328 * as.c: Declare crisaout, criself.
2329 * config/tc-cris.h, config/tc-cris.c: New.
2330 * config/e-criself.c, config/e-crisaout.c: New.
2331 * po/POTFILES.in, po/gas.pot: Regenerate.
2335 * read.c: Fix formatting.
2336 * write.c: Fix formatting.
2340 * sb.c: Include <stdlib.h> if exists for abort ().
2344 * config/tc-hppa.c (hppa_fix_adjustable): Correct LR%/RR% comment.
2345 (arg_reloc_stub_needed): #ifdef OBJ_SOM, not #ifdef SOM.
2346 (pa_type_args): Same here.
2350 * config/obj-coff.c (obj_frob_symbol): Don't merge labels. Don't
2351 merge if the symbol isn't constant. Don't call S_SET_EXTERNAL if
2352 the storage class is already set.
2356 * config/tc-m68hc11.c: Fix formatting.
2357 * config/tc-mn10200.c: Likewise.
2358 * config/tc-mn10300.c: Likewise.
2359 * config/tc-pj.c: Likewise.
2360 * config/tc-tic80.c: Likewise.
2361 * config/tc-w65.c: Likewise.
2365 * expr.c (operand): Permit %bin literals if LITERAL_PREFIXPERCENT_BIN
2370 * doc/c-mips.texi (MIPS Opts): Remove erroneous space after
2375 * config/tc-hppa.c (hppa_fix_adjustable): Use the same checks for
2376 ELF as are used for SOM (except the 32-bit reloc one) to decide
2377 whether a symbol can be reduced to a section symbol. Expand on
2378 the comment for symbols involved in LR% and RR% expressions.
2382 * config/tc-mips.c (mips_disable_float_construction): New
2383 static variable. Set to true if doubles should not be
2384 constructed by loading two single width fp registers with
2385 halves of the value.
2386 (mips_ip): Test mips_disable_float_construction.
2387 (md_longopts): Add command line switches --construct-floats
2388 and --no-construct-floats.
2389 (md_parse_option): Parse new command line options.
2390 (md_show_usage): Describe new command line options.
2392 * doc/c-mips.texi: Document new command line options.
2396 * configure.in: Remove the test /usr/dde for mips-*-sysv4*MP*
2397 * configure: Regenerate.
2401 * configure.in (DEFAULT_EMULATION setting): Revert part of
2402 2000-07-01 change that set te_multi=multi unless set to tmips.
2403 * configure: Regenerate.
2407 * config/obj-coff.c (coff_frob_section): Add padding to the last
2408 section when aligning it increases its size.
2412 * config/tc-tic54x.c: Fix formatting.
2416 * config/tc-h8500.c: Fix formatting.
2417 * config/tc-tic54x.c: Fix formatting.
2421 * config/tc-hppa.c (md_apply_fix): Check fmt 12 and 22 pc-rel
2422 displacements correctly.
2424 * read.h (s_abort): Add ATTRIBUTE_NORETURN.
2428 * hash.c (hash_insert): Add cast to obstruct_alloc to fix
2435 * config/te-hppalinux64.h: Add a new emulation.
2436 * configure.in (emulations): Add configure bits to support the
2437 64-bit Linux/parisc target.
2438 * configure: Regenerate.
2440 * config/tc-hppa.c (md_parse_option): Support `-V' for ELF.
2441 (md_shortopts): Add `V' for ELF.
2443 * config/tc-hppa.c (arg_reloc_stub_needed): Define as zero except
2444 when SOM or ELF_ARG_RELOC are defined.
2445 (pa_type_args): Only set symbol_arg_reloc_info when SOM or
2446 ELF_ARG_RELOC are defined.
2447 (pa_stringer_aux): Don't pa_check_current_space_and_subspace here..
2448 (pa_stringer): ..Do it here instead. Fix comment typos.
2449 (hppa_force_relocation): Cast enums to int before comparing with
2453 * config/tc-hppa.c: Removed unneeded libbfd.h to fix macro
2454 redifinition warning.
2455 (md_apply_fix): Added cast from enum to int for fixP->fx_r_type.
2456 (hppa_force_relocation): ditto
2457 (md_apply_fix): Added cast to buf to fix warnings.
2459 * config/tc-hppa.h (pa_define_label, parse_cons_expression_hppa,
2460 cons_fix_new_hppa, hppa_force_relocation): Prototype.
2462 * config/tc-hppa.c (reloc_type): It's an enum for OBJ_ELF.
2463 (R_N0SEL, R_N1SEL): Define only for OBJ_SOM.
2464 (tc_gen_reloc): Make `code' a reloc_type and `codes' a
2465 reloc_type** to avoid warnings in switch.
2466 (md_apply_fix): Make insn, val signed. Zap buf_wd and read insn a
2467 little earlier instead.
2469 * config/tc-hppa.c (symbol_arg_reloc_info): Define for both som
2471 (pa_type_args): Use symbol_arg_reloc_info.
2472 (struct pa_it): Make arg_reloc unsigned int.
2473 (struct hppa_fix_struct): Likewise for fx_arg_reloc.
2474 (pa_text, pa_data, pa_comm): Don't compile for TE_LINUX.
2475 (pa_code): Delete. pa_text duplicates this function.
2476 (md_pseudo_table): Call obj_elf_text for ".code" if TE_LINUX.
2477 (fix_new_hppa): Argument offset is offsetT, arg_reloc is unsigned
2479 (cons_fix_new_hppa): Actually change selector to e_fsel when
2480 warning about assuming so.
2481 (tc_gen_reloc): More example elf arg reloc code.
2482 (md_apply_fix): Use arg_reloc_stub_needed for elf too.
2483 (hppa_force_relocation): Likewise.
2485 * config/tc-hppa.h: Use TARGET_ARCH_SIZE to select target include
2487 (pa_end_of_source): Prototype.
2488 (hppa_fix_adjustable): Prototype.
2489 (LABELS_WITHOUT_COLONS): Move it..
2490 * config/te-hppa.h: To here.
2492 * config/te-hppa64.h: New file.
2494 * config/tc-hppa.c: Use TARGET_ARCH_SIZE to select target reloc
2496 (md_apply_fix): Pass stdoutput to bfd_hppa_insn2fmt. Handle
2497 format -10, -16, 16 relocs.
2498 (hppa_elf_mark_end_of_function): Test for null
2499 last_call_info->start_symbol
2501 * config/tc-hppa.c (pa_ip): In case 'V', pass `strict' to
2502 CHECK_FIELD, not INSERT_FIELD_AND_CONTINUE. Don't pass opcode to
2503 re_assesmble_* functions. Delete extraneous statements. Fix
2505 (md_apply_fix): Compare against 1048575 in case 21. Don't pass
2506 insn to re_assemble_*.
2510 * doc/internals.texi (Expressions): Fix typo.
2514 * config/tc-sh.c: Fix formatting.
2515 * config/tc-tic54x.c: Fix formatting.
2516 * depend.c: Fix formatting.
2517 * flonum-konst.c: Likewise.
2518 * flonum-mult.c: Likewise.
2522 * config/tc-sh.c: Fix comments.
2523 * config/obj-vms.c: Fix comments.
2524 * config/tc-a29k.c: Likewise.
2525 * config/tc-alpha.c: Likewise.
2526 * config/tc-h8300.c: Likewise.
2527 * config/tc-h8500.c: Likewise.
2528 * config/tc-i370.c: Likewise.
2529 * config/tc-ia64.c: Likewise.
2530 * config/tc-m68hc11.c: Likewise.
2531 * config/tc-m68k.c: Likewise.
2532 * config/tc-mips.c: Likewise.
2533 * config/tc-ns32k.c: Likewise.
2534 * config/tc-ppc.c: Likewise.
2535 * config/tc-z8k.c: Likewise.
2539 * config/obj-coff.c (TC_COFF_SECTION_DEFAULT_ATTRIBUTES): New.
2540 Default to '(SEC_LOAD | SEC_DATA)'.
2541 (obj_coff_section) [BFD_ASSEMBLER]: Use it.
2543 * doc/internals.texi (CPU Backend): Describe
2544 TC_COFF_SECTION_DEFAULT_ATTRIBUTES.
2548 * as.c (parse_args): NULL terminate the long option list.
2552 * config/tc-d30v.h: Include "write.h" for fixS.
2553 (d30v_start_line, md_pcrel_from_section): Add function prototypes.
2557 * config/tc-d30v.c (write_2_short): Further changes to warning
2558 messages produced when combining EITHER_BUT_PREFER_MU attributed
2567 * config/tc-arm.c (psrs): Accept combinations of flags.
2571 * config/tc-avr.c: Change _ () to _() around all strings marked
2572 for translation (exception from the usual coding style).
2573 (avr_opt): New struct variable, how the new switches are set.
2574 (OPTION_MMCU): Define as 'm' and actually use.
2575 (md_longopts): Add -mall-opcodes, -mno-skip-bug, -mno-wrap.
2576 (show_mcu_list): New function, display the list of known MCUs.
2577 (md_show_usage): Document the new switches. Call show_mcu_list.
2578 (avr_set_arch): Change 'm' to OPTION_MMCU.
2579 (md_parse_option): Call show_mcu_list if unknown MCU specified.
2580 Handle the new switches.
2581 (avr_operands): Disable warnings for undefined combinations of
2582 operands if -mall-opcodes. Disable warnings for skipping two-word
2583 instructions if enhanced core or -mno-skip-bug.
2584 (avr_operand): Accept all addressing modes on avr1 if -mall-opcodes.
2585 (md_apply_fix3): Reject 8K wrap if >8K or -mno-wrap.
2586 (md_assemble): Accept opcodes not supported by MCU if -mall-opcodes.
2587 (avr_ldi_expression): Warn about implicit lo8().
2588 * config/tc-avr.h (md_pcrel_from_section): Add prototype.
2592 * configure.in: Add entry for mips-*-sysv4*MP*
2593 * configure: Rebuild
2594 * config/tc-mips.c (mips_target_format): Return elf32-tradbigmips or
2595 elf32-tradlittlemips for traditional mips targets.
2596 * config/tc-mips.c (md_estimate_size_before_relax): Duplicate the
2597 test for Link Once sections as in adjust_reloc_syms.
2598 * config/te-tmips.h: New file for traditional mips targets. Define
2603 * config/obj-coff.c (obj_coff_setcion) [BFD_ASSEMBLER]: If the
2604 flags argument is not present, don't change an existing section's
2605 section's attributes. If the flags argument is present, warn if the
2606 attributes don't match the section's current attributes. When
2607 long section names are supported, set SEC_LINK_ONCE and
2608 SEC_LINK_DUPLICATES_DISCARD for a new .gnu.linkonce section.
2612 * config/obj-aout.c (obj_aout_type): Do not ignore for undefined
2613 symbols; create them.
2617 * write.c (set_segment_vma): New: Set vma and lma for a segment.
2618 (write_object_file) [BFD_ASSEMBLER && OBJ_COFF && TE_GO32]: Use it.
2622 * config/tc-mips.c (mips_ip): handle "(foo-.-4)" type of
2623 expressions. Ignore the problem when handling 16 bit signed
2624 immediates, because the assembler will take care of the relocation
2629 * config/tc-d30v.c (write_2_short): Do not allow opcodes with
2630 the EITHER_BUT_PREFER_MU attribute to be combined into a reverse
2631 sequential order, and emit warning messages if the input source
2632 code contains constructs like that, or parallel constructs
2633 containing such opcodes.
2637 * config/tc-avr.c (mcu_types): Rename avr4 to avr5, add avr4.
2638 Add more MCU types for avr4 and avr5. Replace at94k{10,20,40}
2639 with just at94k. Change AVR_ISA_85xx back to AVR_ISA_2xxx.
2640 (md_show_usage): Update usage message.
2641 (md_parse_option): Allow redefinition of MCU type within the
2642 same avr[1-5] bfd machine type. Show both old and new MCU type
2643 in the error message.
2644 (md_apply_fix3): Support 8K wrap if AVR_ISA_MEGA is not set.
2645 Simplify 8K wrap code.
2649 * config/obj-aout.c: Remove all uses of DEFUN.
2650 * config/obj-ieee.c: Likewise.
2651 * config/tc-sh.c: Fix comment typos.
2652 * config/tc-tahoe.c: Likewise.
2653 * config/tc-vax.c: Likewise.
2654 * config/tc-w65.c: Likewise.
2655 * config/tc-z8k.c: Likewise.
2656 * config/tc-h8300.c (build_bytes): Assemble ldmac correctly.
2660 * config/tc-i386.c (md_estimate_size_before_relax): Revert
2661 more changes from Sept 1999
2662 (tc_i386_fix_adjustable): ditto
2663 (md_apply_fix3): ditto
2667 * cgen.c (expr_jmp_buf_p): New validity flag for expr_jmp_buf.
2668 (gas_cgen_parse_operand): Set it around expression() call.
2669 (gas_cgen_md_operand): Test for it before longjmp().
2673 * config/tc-h8500.c: Remove all uses of DEFUN.
2674 * config/tc-sh.c: Likewise.
2675 * config/tc-w65.c: Likewise.
2676 * config/tc-z8k.c: Likewise.
2678 * config/tc-h8500.c: Fix typos in comments.
2682 * expr.c (operand): Permit $hex literals if LITERAL_PREFIXDOLLAR_HEX
2687 * expr.c (operand): Do not as_bad() if RELAX_PAREN_GROUPING is
2688 defined. Fix error message for `[' grouping.
2692 * config/tc-h8300.c: Fix formatting and comment typos.
2696 * config/tc-ia64.c (note_register_values): Move premature QP
2697 notation clearing into the appropriate place.
2701 * dep-in.sed: Escape literal `.'s on patterns. Trim off `../'
2702 first before anything else. Add bin-bugs.h, emul.h and progress.h
2703 Sort list of files as for $(OBJS) in Makefile.am.
2705 * Makefile.am (DEP): grep for leading `/' in DEPA, and fail if we
2706 find one. Remake dependencies.
2707 ($(OBJS)): Add bin-bugs.h, emul.h, and progress.h Sort the list.
2708 * Makefile.in: Regenerate.
2709 * doc/Makefile.in: Regenerate.
2711 * config/tc-i386.c (i386_displacement): Don't assume a constant
2712 displacement is necessarily 16 bits when in 16 bit code mode.
2713 (md_assemble): Instead size the displacement here after we know
2714 for sure that a .code16gcc operand hasn't automatically added
2715 operand size prefixes.
2719 * Makefile.am: Rebuild dependency.
2720 * Makefile.in: Rebuild.
2724 * config/tc-h8300.c (parse_reg): Make the function static.
2725 (parse_exp): Likewise.
2729 * config/tc-i386.c (tc_i386_fix_adjustable): Revert change from
2730 Sept 1999; RVA relocs need to be treated more like DIR32 relocs
2731 for cygwin import libraries to work properly.
2735 * Makefile.am: Rebuild dependency.
2736 * Makefile.in: Rebuild.
2737 * configure: Likewise.
2738 * doc/Makefile.in: Likewise.
2742 * doc/internals.texi (CPU backend): Add @itemx for
2743 TC_START_LABEL_WITHOUT_COLON.
2744 * doc/c-tic54x.texi: New.
2745 * doc/as.texinfo: Add tic54x features and include primary tic54x
2747 * doc/all.texi: Add C54X.
2748 * doc/Makefile.am (CPU_DOCS): Add c-tic54x.texi.
2749 * doc/Makefile.in: Regenerate.
2750 * configure.in: Add tic54x and define LIBM for tic54x.
2751 * configure: Regenrate.
2752 * config/tc-tic54x.[ch]: New.
2753 * config/obj-coff.h: Add tic54x.
2754 * Makefile.am: (CPU_TYPES): Add tic54x.
2755 (TARGET_CPU_CFILES): Add 'tc-tic54x.c'.
2756 (TARGET_CPU_HFILES): Add 'tc-tic54x.h'.
2757 (as_new_LDADD): Add $(LIBM).
2758 * Makefile.in: Regenerate.
2762 * doc/Makefile.am (CPU_DOCS): Added 68hc11 file.
2763 * doc/c-m68hc11.texi: Document 68HC11 and 68HC12 port.
2764 * doc/as.texinfo: Likewise.
2766 * configure, Makefile.in: Regenerate.
2767 * configure.in (emulations): Recognize m6811 and m6812.
2768 * Makefile.am (CPU_TYPES, TARGET_CPU_CFILES, TARGET_CPU_HFILES):
2769 Added files for 68hc11 and 68hc12 assembler.
2770 * config/tc-m68hc11.c: Assembler for 68hc11 and 68hc12.
2771 * config/tc-m68hc11.h: Header definition for that assembler.
2775 * symbols.c (resolve_symbol_value): Use bfd_octets_per_byte
2776 instead of OCTETS_PER_BYTE.
2778 * config/tc-v850.c: Fix compile time warnings.
2779 * config/tc-ppc.c: Fix compile time warnings.
2783 * configure.in: Don't emulate i386-pc-pe-coff with i386coff.
2784 * configure: Rebuild.
2788 * config/obj-coff.c (obj_coff_weak): Typo fix: Change BFD_ASSEMLER
2793 * config/tc-mips.c (md_parse_option): Accept RM5200,RM5230,
2794 RM5231, RM5261, RM5721 and RM7000 as r5000 cpu variants.
2796 * doc/c-mips.texi: Document newly accepted cpu variants.
2800 * config/tc-mips.h: Remove definition of ONLY_STANDARD_ESCAPES.
2804 * macro.c (getstring): Make it possible to escape the quote
2809 * config/tc-hppa.c (pa_export): Weak symbols can be global.
2813 * configure: Regenerate.
2817 * app.c (do_scrub_begin): Don't default lex[';'] as a line
2819 * doc/internals.texi (line_separator_chars): Semicolon is no
2820 longer a default. Mention null and newline as defaults.
2822 * read.c (is_end_of_line): Remove ifdef TC_HPPA.
2824 * config/tc-i386.h (line_separator_chars): Explicitly mention `;'
2825 * config/tc-i860.h (line_separator_chars): Likewise.
2826 * config/tc-h8300.c (line_separator_chars): Likewise.
2827 * config/tc-i960.c (line_separator_chars): Likewise.
2828 * config/tc-m68k.c (line_separator_chars): Likewise.
2829 * config/tc-mips.c (line_separator_chars): Likewise.
2830 * config/tc-ns32k.c (line_separator_chars): Likewise.
2831 * config/tc-sparc.c (line_separator_chars): Likewise.
2832 * config/tc-vax.c (line_separator_chars): Likewise.
2834 * config/tc-h8300.c (comment_chars): Use string initialiser.
2835 * config/tc-i960.c (line_comment_chars): Likewise.
2836 * config/tc-z8k.c (comment_chars, line_comment_chars,
2837 line_separator_chars): Likewise.
2839 * config/tc-arm.c (line_separator_chars): Always use `;', not just
2844 * config/tc-arm.c (cons_fix_new_arm): Assign correct reloc value
2849 * configure.in (VERSION): Update to show this is the CVS mainline.
2853 * config/tc-alpha.c (md_undefined_symbol): Properly understand that
2854 $at is the integer register $r28, vs. both $r28 and the floating
2855 point register $f28.
2859 * config/tc-ia64.c (generate_unwind_image): Call ia64_flush_insns.
2860 (dot_endp): Don't call ia64_flush_insns.
2861 (emit_one_bundle): Don't delete prologue/body records from
2862 unwind_record list in first loop. Rewrite second loop to account for
2867 * config/tc-ia64.c: Add missing prototypes.
2868 (generate_unwind_image): Cast argument to output_unw_records call.
2872 * config/tc-avr.c (avr_operand): fix the formatting of the comment.
2876 * config/tc-avr.c (AVR_ISA_???): moved to include/opcode/avr.h
2877 (REGISTER_P): likewise.
2878 (avr_opcodes): uses include/opcode/avr.h
2879 (avr_operand): enable ld r,Z or st r,Z for at90s1200.
2883 * read.c (is_end_of_line): No ';' for TC_HPPA. Add missing
2888 * read.c (is_end_of_line): Put back `;'.
2892 * config/tc-i386.c (md_shortopts): Remove 'm', add 'q' to non-elf.
2896 * expr.c (operand): Test is_end_of_line outside switch to catch
2897 line separator chars that are also operators.
2898 (operator): Return O_illegal for line separator chars.
2900 * read.c (is_end_of_line): Use 1 instead of 99. Don't set `;'
2901 entry (or `!' entry for TC_HPPA).
2903 * config/tc-arm.c (my_get_float_expression): Cast to unsigned char
2904 before indexing is_end_of_line. Remove redundant check for '\0'.
2906 * config/tc-h8500.c (md_assemble): Likewise.
2907 * config/tc-mcore.c (md_assemble): Likewise.
2908 * config/tc-tic30.c (tic30_find_parallel_insn): Likewise.
2911 * config/tc-m88k.c (s_bss): Cast to unsigned char before indexing
2913 * config/tc-mcore.c (mcore_cons): Likewise.
2914 (mcore_float_cons): Likewise.
2915 (mcore_stringer): Likewise.
2916 * config/tc-tic30.c (tic30_find_parallel_insn): Likewise.
2920 * config/tc-arm.c (do_mrs): Allow SPSR_BIT to be set correctly.
2924 * config/tc-sh.c: Fix compile time warning messages.
2926 * config/tc-mips.c: Fix compile time warning messages.
2930 * doc/as.texinfo: Update copyright dates.
2931 (Local Labels): Delete misplaced mention of ARM.
2932 * NEWS: Mention ARM ELF support.
2936 * config/tc-mn10300.c (md_assemble): Copy size to real_size before
2937 it is modified, and use the real_size to compute the frag address
2938 for dwarf2 line info.
2942 * Makefile.am (DEP, DEP1, dep, dep-in, dep-am): Use a better sed
2943 line-matching scheme to cope with automake moving variables around.
2944 ($(TARG_CPU_O)): Remove dependency on TARG_CPU_DEP_@target_cpu_type@
2945 * Makefile.in: Regenerate.
2949 * config/tc-sparc.c (sparc_relax): New.
2950 (md_longopts): Add -relax and -no-relax options.
2951 (md_parse_options, md_show_usage): Likewise.
2952 (md_apply_fix3): Optimize tail call into branch always if possible.
2956 * config/tc-d10v.c (write_2_short, parallel_ok, md_assemble,
2957 d10v_cleanup) implement Mitsubishi's newly explained branch-packing
2958 rules, with warning when a GAS statement specifies a packing that
2959 will result in an instruction being squashed.
2960 Added typdef packing_type and enumerals, changed various integer
2961 literals to use the enumerals.
2965 * config/tc-ia64.c (dot_restorereg_p): New function.
2966 (md_pseudo_table): Add restorereg.p.
2967 (output_X3_format): Fix typo: record type should be UNW_X3, not UNW_X1.
2968 (output_X4_format): Fix typo: record type should be UNW_X4, not UNW_X2.
2970 * config/tc-ia64.h (unw_record_type): Add unwabi.
2971 (unw_r_record): Rename member MASK to GRMASK. Add sub-structure
2972 called MASK with members for imask, and the masks produced by
2973 fr_mem, gr_mem, br_mem, and frgr_mem.
2974 (unw_p_record): Add members ABI and CONTEXT.
2975 (unw_x_record): Add member AB.
2976 * config/tc-ia64.c (enum reg_symbol): Add REG_PSP and REG_PRIUNAT
2977 as pseudo-register for use during unwind info generation.
2978 (AR_PFS, AR_LC): New macros.
2979 (enum pseudo_type): Add PSEUDO_FUNC_REG to permit declaring registers
2980 whose names start with an at sign (as in "@priunat").
2981 (pseudo_func): Add "svr4", "hpux", "nt" constants and "priunat"
2983 (unwind_list, unwind_tail, current_unwind_entry, proc_start,
2984 proc_end, unwind_info, personality_routine): Consolidate into
2985 "unwind" structure to reduce offset-table use. Add member
2986 NEXT_SLOT_NUMBER to track the slot number for the next instruction
2988 (output_R1_format, output_R3_format, output_P3_format,
2989 output_P6_format): Initialize R with zero to reduce compiler warnings.
2990 (output_P7_format): Ditto. Add `default' branch to switch
2991 statement to reduce compiler warnings.
2992 (output_P8_format, output_B1_format, output_B4_format): Ditto.
2993 (output_P4_format): Rename 2nd & 3rd arg to IMASK and IMASK_SIZE.
2994 (format_ab_reg): Rename from format_a_b_reg. Merge A and B args
2995 into single argument.
2996 (output_X1_format, output_X3_format): Initialize R with zero to reduce
2997 compiler warnings. Merge A and B args into single argument.
2998 (output_X2_format, output_X4_format): Remove unused variable R. Merge
2999 A and B args into single argument.
3000 (free_record): Removed (wasn't used).
3001 (free_list_records): Also free imasks in prologue records.
3002 (output_prologue, output_prologue_gr): Initialize mask bits to zero.
3003 (output_spill_mask): Remove.
3004 (output_unwabi): New function.
3005 (output_epilogue, output_label_state, output_copy_state): Call
3007 (output_spill_psprel, output_spill_sprel, output_spill_psprel_p,
3008 output_spill_sprel_p, output_spill_reg, output_spill_reg_p): Add AB
3010 (process_one_record): New locals FR_MASK and GR_MASK. Ignore
3011 gr_mem, fr_mem, br_mem, and frgr_mem records and instead emit them
3012 as part of handling the prologue records. Emit region's imask if
3013 we have one. Handle unwabi, epilogue, label_state, copy_state,
3014 spill_psprel, spill_sprel, spill_reg, spill_psprel_p,
3015 spill_sprel_p, and spill_reg_p records.
3016 (set_imask, count_bits, slot_index): New function.
3017 (fixup_unw_records): Fix region size computation. Handle
3018 epilogue, spill_reg, spill_sprel, spill_psprel, spill_reg_p,
3019 spill_sprel_p, and spill_psprel_p records. Merge mask bits of
3020 frgr_mem, fr_mem, gr_mem, br_mem on a per-region basis and
3021 set_imask accordingly. Update imask for gr_gr, and br_gr records.
3022 (convert_expr_to_ab_reg, convert_expr_to_xy_reg): New function.
3023 (dot_save): Use manifest constants for applicaton registers.
3024 Handle REG_PR and REG_PRIUNAT.
3025 (dot_restore): Don't just ignore it.
3026 (dot_restorereg): New function..
3027 (generate_unwind_image): Ensure unwind info is a multiple of eight
3028 bytes, not just four bytes.
3029 (dot_handlerdata, dot_unwentry): Demand empty rest of line.
3030 (dot_altrp): Don't just ignore it.
3031 (dot_savemem): New function. Replaces dot_savesp() and
3032 dot_savepsp(). Use manifest constants for applicaton registers.
3033 Handle REG_PR and REG_PRIUNAT.
3034 (dot_savef): Simplify.
3035 (dot_saveb): Support generation of br_gr.
3036 (dot_spillreg, dot_spillmem, dot_spillreg_p, dot_spillmem_p,
3037 dot_label_state, dot_copy_state): New function.
3038 (dot_unwabi): Don't just ignore it.
3039 (md_pseudo_table): Add restorereg, spillreg, spillsp, spillpsp,
3040 spillreg.p, spillsp.p, spillpsp, label_state, copy_state,
3041 unwabi, vframesp, and vframepsp. Fix typo alprp->altrp.
3042 (emit_one_bundle): Set slot number for prologue/body records
3043 *before* emitting the first insn.
3044 (emit_one_bundle): Set UNWIND.NEXT_SLOT_NUMBER.
3045 (md_begin): Declare "psp" pseudo-register.
3046 (md_operand): Handle PSEUDO_FUNC_REG. Fix printing of error message
3047 so we don't get segfault.
3048 (output_psp_sprel): Output sp/psp relative offsets as 4-byte word
3049 counts as required per SW Conventions manual
3050 (output_rp_psprel, output_rp_sprel, output_pfs_psprel,output_pfs_sprel,
3051 output_preds_psprel, output_preds_sprel, output_spill_base,
3052 output_unat_psprel, output_unat_sprel, output_lc_psprel,
3053 output_lc_sprel, output_fpsr_psprel, output_fpsr_sprel,
3054 output_priunat_psprel, output_priunat_sprel, output_bsp_psprel,
3055 output_bsp_sprel, output_bspstore_psprel, output_bspstore_sprel,
3056 output_rnat_psprel, output_rnat_sprel, output_spill_psprel,
3057 output_spill_sprel, output_spill_psprel_p, output_spill_sprel_p):Ditto.
3058 (dot_vframe): Implement.
3059 (dot_vframesp, dot_vframepsp): New function.
3063 * configure.in (i386-*-freebsd a.out entry): Quote properly.
3064 * configure: Regenerate.
3068 * config/tc-i386.c (md_assemble): Pass jump reloc in fr_var...
3069 (md_estimate_size_before_relax): so we can use it here instead of
3070 old kludges. Localise vars to blocks. Comment.
3072 * frags.c (frag_new): Update fr_var comments.
3073 * frags.h (struct frag): Ditto.
3077 * config/tc-ia64.c (FUNC_PC_RELATIVE): New.
3078 (pseudo_func): Add pcrel.
3079 (operand_match): Handle IA64_OPND_TGT64.
3080 (build_insn): Likewise.
3081 (md_begin): Initialize pseudo_func[FUNC_PC_RELATIVE].
3082 (ia64_gen_real_reloc_type): Handle FUNC_PC_RELATIVE.
3083 (fix_insn): Handle all three 64-bit relocation types.
3087 * obj.h (struct format_ops): New members begin, app_file,
3088 s_set_other, s_set_desc, s_get_type, s_set_type,
3089 separate_stab_sections, init_stab_section.
3091 * config/obj-multi.h: Update GPL notice to v2.
3093 (obj_app_file): New.
3094 (S_SET_SIZE): Test s_set_size for NULL before calling.
3095 (S_SET_ALIGN): Similar for s_set_align.
3100 (SEPARATE_STAB_SECTIONS): New.
3101 (INIT_STAB_SECTION): New.
3102 (EMIT_SECTION_SYMBOLS): New.
3103 (AOUT_STABS) [OBJ_MAYBE_AOUT]: Define.
3105 * config/obj-elf.h: Update GPL notice to v2.
3106 Mention that this file is included from obj-multi.h.
3107 (obj_begin): Wrap definition in ifndef.
3108 (elf_file_symbol): Constify declaration.
3109 (obj_app_file): Ditto.
3110 (SEPARATE_STAB_SECTIONS, INIT_STAB_SECTION, OBJ_PROCESS_STAB):
3111 Wrap in ifndef SEPARATE_STAB_SECTIONS.
3113 * config/obj-elf.c (elf_s_set_other): New.
3114 (elf_file_symbol): Constify argument.
3115 (elf_separate_stab_sections): New.
3116 (elf_init_stab_section): New.
3117 (elf_format_ops): Add new members. Remove comma at end.
3119 * config/obj-ecoff.c (ecoff_separate_stab_sections): New.
3120 (ecoff_format_ops): Add new fields. Remove comma at end.
3121 Mention inconsistency for emit_section_symbols.
3123 * config/obj-coff.h (c_dot_file_symbol): Constify declaration.
3125 * config/obj-coff.c (c_dot_file_symbol): Constify argument.
3126 (coff_separate_stab_sections): New.
3127 (coff_format_ops): Add new members.
3129 * config/obj-aout.c (obj_aout_sec_sym_ok_for_reloc): New.
3130 (obj_aout_s_set_other): New.
3131 (obj_aout_s_set_desc): New.
3132 (obj_aout_s_get_type): New.
3133 (obj_aout_s_set_type): New.
3134 (obj_aout_separate_stab_sections): New.
3135 (aout_format_ops): New members added. Use obj_aout_process_stab,
3136 not 0. Use obj_aout_sec_sym_ok_for_reloc, not 0.
3137 (obj_aout_frob_symbol): Add ATTRIBUTE_UNUSED to args as
3139 (obj_aout_line, obj_aout_weak, obj_aout_type): Ditto.
3143 * config/tc-i386.c (tc_i386_fix_adjustable): Prevent adjustment
3144 for OBJ_MAYBE_ELF too. Use S_IS_EXTERNAL instead of S_IS_EXTERN.
3145 (md_estimate_size_before_relax): Ensure jumps to weak and
3146 externally visible symbols are relocatable.
3150 * stabs.c (aout_process_stab): Make global.
3151 (s_desc): Add ATTRIBUTE_UNUSED to args as appropriate.
3152 * read.h (aout_process_stab): Declare.
3154 * configure.in (EMULATIONS) [i386aout, i386coff, i386elf]:
3155 Generalize to *aout, *coff *elf.
3156 * configure: Regenerated.
3158 * doc/internals.texi (Object format backend): Say
3159 SEPARATE_STAB_SECTIONS needs to be nonzero, not just defined.
3161 * Makefile.am (TARG_ENV_HFILES): Delete te-multi.h.
3162 * Makefile.in: Regenerated.
3166 * cgen.h (GAS_CGEN_MAX_FIXUPS): Check if already defined.
3170 * config/tc-hppa.c (md_apply_fix): Mask out immediate bits of
3171 instruction to reflect change in re_assemble_*.
3175 * configure.in (hppa-*-hpux11*): If the cpu is hppa*64*, then
3176 build PA64 ELF tools.
3177 * configure: Rebuilt.
3181 * Makefile.am: Regenerate dependencies.
3182 * Makefile.in: Regenerate.
3186 * config/tc-arm.c (struct asm_psr): Add boolean field
3187 distinguishing between CSPR and SPSR. Rename 'number' field
3189 (psrs): Rearrange contents to match new asm_psr structure.
3190 (arm_psr_parse): Move next to psr_required_here. Make it
3191 return an asm_psr structure.
3192 (psr_required_here): Use asm_psr structure returned by
3194 (do_msr): Reorganise to allow psr_required_here to be called
3196 (md_undefined_name): Mark 'name' parameter as unused, since
3197 the COFF target does not use it.
3201 * config/te-386bsd.h: Clean up comments to adhere to the GNU coding
3203 * config/te-aux.h: Likewise.
3204 * config/te-dpx2.h: Likewise.
3205 * config/te-go32.h: Likewise.
3206 * config/te-hp300.h: Likewise.
3207 * config/te-hppa.h: Likewise.
3208 * config/te-i386aix.h: Likewise.
3209 * config/te-ic960.h: Likewise.
3210 * config/te-interix.h: Likewise.
3211 * config/te-nbsd532.h: Likewise.
3212 * config/te-pc532mach.h: Likewise.
3213 * config/te-ppcnw.h: Likewise.
3214 * config/te-psos.h: Likewise.
3215 * config/te-sparcaout.h: Likewise.
3216 * config/te-sun3.h: Likewise.
3217 * config/te-sysv32.h: Likewise.
3221 * config/tc-h8300.c (do_a_fix_imm): Don't rely on `short' being 16
3222 bits. Instead explicitly mask and sign extend. Do the 8 bit mask
3223 and sign extend without an if statement.
3224 (build_bytes): Likewise.
3228 * config/tc-h8300.c (do_a_fix_imm): Output a reloc for no
3229 X_add_symbol L_32 case.
3233 * config/te-freebsd.h: New file.
3237 * asintl.h (gettext, dgettext, dcgettext, textdomain,
3238 bindtextdomain): Replace defines with those from intl/libgettext.h
3239 to quieten gcc warnings.
3241 * NEWS: Mention x86 .arch and -q.
3243 * config/tc-i386.c (quiet_warnings): New.
3244 (md_assemble): Use quiet_warnings.
3245 (md_parse_option): Set quiet_warnings from -q.
3246 (md_show_usage): Mention -q, delete -m.
3247 (flag_do_long_jump): Delete.
3248 (md_parse_option): Remove -m.
3249 (md_show_usage): Remove -m.
3250 (md_create_long_jump): Remove useless flag_do_long_jump code.
3252 * as.c (parse_args): In case OPTION_DEFSYM, use a valueT to hold
3253 the symbol value, and use bfd_scan_vma if BFD_ASSEMBLER.
3258 * doc/c-i386.texi (i386-Arch): New section.
3259 (i386-Syntax): Mention .intel_syntax and .att_syntax.
3261 * config/tc-i386.c (cpu_arch_name, cpu_arch_flags): New.
3262 (smallest_imm_type): Use smallest opcode for shift by one if cpu
3263 architecture has been given and is not 486.
3264 (set_cpu_arch): New.
3265 (md_pseudo_table): Add .arch.
3266 (md_assemble): Warn if cpu architecture has been given and an
3267 unsupported instruction.
3269 * config/tc-i386.h (SMALLEST_DISP_TYPE): Delete.
3270 Move operand_types bit defines after relevant template field.
3271 (template): Add cpu_flags.
3277 * config/tc-mn10300.h (md_end): Define.
3278 (mn10300_finalize): Declare.
3279 * config/tc-mn10300.c: Include dwarf2dbg.h.
3280 (debug_line): Define.
3281 (md_assemble): Generate dwarf2 line info.
3282 (mn10300_finalize): New function. Finalize dwarf2 info.
3286 * config/tc-mips.c (md_estimate_size_before_relax): Use the
3287 external version of the relocation for weak symbols.
3291 * config/tc-ia64.c (output_P7_format, case mem_stack_f): Output fixed
3292 frame size in units of 16 bytes, as required per SW Conventions manual.
3293 (output_unw_records): Output info-block header as a dword to get
3298 * as.h: #include "file", not <file> on files from ../include.
3299 (as_abort, as_fatal): Add ATTRIBUTE_NORETURN.
3300 * config/tc-m68k.c (m68k_ip): Fix signed/unsigned warnings.
3301 (md_convert_frag): Add ATTRIBUTE_UNUSED.
3302 (tc_coff_symbol_emit_hook): Ditto.
3303 (OPTCOUNT): Cast to int to avoid compiler warning.
3304 (md_begin): Fix signed/unsigned warnings.
3308 * config/tc-m68k.c (md_convert_frag_1): Abort if we end up in the
3309 ABRANCH LONG case for a conditional branch on a 68000.
3310 (md_estimate_size_before_relax): Likewise. Also handle
3311 flag_short_refs correctly for ABRANCH, BCC68000, and DBCC.
3312 (m68k-ip: case ABSL): Relax absolute references to 16-bit
3313 PC-relative on all CPUs.
3314 (md_estimate_size_before_relax): Likewise.
3318 * as.c (parse_args): Just mention current year in printed
3323 * config/tc-ppc.c (pre_defined_registers): Add entries for vector
3325 (md_parse_option): Recognize -m7400.
3329 * config/atof-ieee.c (gen_to_words): When adding carry back in,
3330 don't permit lp to become less than the words array.
3334 config/tc-mcore.c (md_apply_fix3): BFD_RELOC_MCORE_PCREL_IMM11BY2
3335 Fix little-endian case.
3339 * as.c (parse_args): Update copyright.
3343 * gas/config/tc-i386.h (SUB_SEGMENT_ALIGN): If TE_GO32, return 4
3344 for the .bss section too.
3348 * configure.in: Set em=linux for hppa-*-linux.
3349 * configure: Regenerate.
3350 * doc/Makefile.in: Regenerate with correct automake.
3352 * frags.c (frag_grow): Sanity check chunk_size.
3354 * config/obj-elf.h: #include "bfd.h" not <bfd.h>
3355 * config/obj-som.h: Likewise.
3356 * config/obj-ieee.h: Likewise.
3358 * config/tc-hppa.h: Test BFD_ARCH_SIZE, not BFD64.
3360 * config/tc-hppa.c (log2): Only compile when OBJ_SOM.
3361 (md_pseudo_table): Fully initialise OBJ_ELF cases.
3362 (fix_new_hppa): Add ATTRIBUTE_UNUSED to args as appropriate.
3363 (pa_ip): low_sign_unext now returns via function value. Use
3364 re_assemble_* instead of dis_assemble_* and
3365 INSERT_FIELD_AND_CONTINUE combination. Don't call sign_unext
3367 (md_convert_frag): Add ATTRIBUTE_UNUSED to args as appropriate.
3368 (md_section_align, md_parse_option, md_show_usage,
3369 md_undefined_symbol, pa_align, pa_block, pa_brtab, pa_try,
3370 pa_callinfo, pa_code, pa_comm, pa_end, pa_enter, pa_entry,
3371 pa_exit, pa_export, pa_import, pa_label, pa_leave, pa_level,
3372 pa_origin, pa_param, pa_proc, pa_procend, pa_space, pa_spnum,
3373 pa_version, pa_compiler, pa_copyright, pa_data, pa_fill, pa_lsym,
3375 (md_apply_fix): Change type of new_val to offsetT. Delete w1, w2,
3376 w, resulti. Add insn, val. Move bfd_get_32 and bfd_put_32
3377 outside of switch. Correct mask and shifting errors in case 10
3378 and case -11. In case 21, compare against signed range to suit
3379 hppa_field_adjust changes. In case 12, use re_assemble_12. In
3380 case 17 and case 22, use offsetT variable to properly check range.
3381 Use re_assemble_* here too.
3382 (evaluate_absolute): Change type of value to offsetT. Call
3383 hppa_field_adjust to do the work for us.
3384 (pa_parse_cmpb_64_cmpltr): Delete save_s.
3385 (pa_parse_cmpib_64_cmpltr): Ditto.
3386 (pa_build_unwind_subspace): Delete unused var subseg. Change type
3387 of i to unsigned int.
3388 (pa_type_args): Conditionally declare symbol if OBJ_SOM.
3389 (pa_end_of_source): Return type is void.
3393 * macro.c (macro_expand_body): Don't prepend macro number with zeroes.
3397 * config/tc-avr.c: ATTRIBUTE_UNUSED added to the necessary places.
3398 More comments added.
3399 (md_begin): Removed "construct symbols for each register name".
3400 Because register names conflicts with GCC generated function
3402 (avr_operand): Now constant numbers can be used as a register
3403 identifiers (0 as r0, 31 as r31).
3404 (md_assemble): use skip_space () before parsing instruction
3409 * configure.in: Set bfd_gas=yes on i386-*-pe and i386-*-nt* to
3410 ensure all pe targets use bfd. Remove unnecessary bfd_gas=yes on
3411 arm-*-netbsd* and arm-*-wince as this is set for all arm*.
3412 * configure: Regenerate.
3416 * as.h: Correctly check GCC version.
3420 * doc/as.1: Fix unbalanced brackets.
3422 * config/tc-i386.c (comment_chars): Don't use '/' as comment start if
3424 (line_comment_chars): Set to '/' if TE_FreeBSD.
3428 * configure.in: Configury support for PA64 (currently disabled).
3429 * configure: Rebuilt.
3433 * config/tc-mips.c (s_change_sec): Use record_alignment, not
3434 bfd_set_section_alignment.
3438 * config/tc-i386.c (offset_in_range): Ensure shift counts are less
3443 * doc/c-arm.texi (ARM Directives): Document behaviour of .align 0.
3444 * doc/as.texinfo (Align): Include arm and strongarm in list of
3445 targets that have the second form of the behaviour of the .align
3450 * obj-som.c: Terminate obj_pseudo_table.
3454 * as.c (parse_args): Allow md_parse_option to override -a listing
3456 * config/obj-coff.c (add_lineno): Change type of offset parameter
3457 from "int" to "bfd_vma."
3458 * config/tc-ppc.c (md_pseudo_table): Add "llong" and "machine."
3459 (ppc_mach, ppc_subseg_align, ppc_target_format): New.
3460 (ppc_change_csect): Align correctly for XCOFF64.
3461 (ppc_machine): New function, which discards "ppc_machine" line.
3462 (ppc_tc): Cons for 8 when code is 64 bit.
3463 (md_apply_fix3): Don't check operand->insert. Handle 64 bit
3465 (md_parse_option): Handle -a64 and -a32.
3467 * config/tc-ppc.h (TARGET_MACH): Define.
3468 (TARGET_FORMAT): Move to function.
3469 (SUB_SEGMENT_ALIGN): Use ppc_subseg_align.
3473 * config/tc-avr.c: New AVR_ISA_ defined.
3474 (md_assemble): Handle opcodes with optional operands (lpm,elpm).
3475 (avr_operand): Handle 'a', 'v' and 'z' constraint letters needed
3476 for `fmul', `movw' and `lpm R,Z' instructions.
3477 (avr_operands): Warn if current opcode is a two-word instruction
3478 and previous opcode was cpse/sbic/sbis/sbrc/sbrs.
3479 (avr_opcodes): New commands added.
3480 (REGISTER_P): Check 'a' and 'v' constraint letters.
3481 (mcu_types): New MCU added.
3485 * config/tc-ia64.c (pseudo_func[]): Add new "nat" entry equivalent
3487 (operand_match): Conditionally insert default bit values for IMMU9.
3491 * configure.in: Add NetBSD/sparc ELF and NetBSD/sparc64 support.
3492 * configure: Rebuilt.
3497 * config/tc-hppa.c (md_apply_fix): Handle new PA2.0 formats.
3499 * config/tc-hppa.c (CHECK_ALIGN): New macro.
3500 Added handling of new operand types l,y,&,fe,fE,fx.
3508 * Makefile.am (CPU_TYPES): Add ia64.
3509 (TARGET_CPU_CFILES): Add config/tc-ia64.c.
3510 (TARGET_CPU_HFILES): Add config/tc-ia64.h.
3511 * Makefile.in: Rebuild.
3512 * app.c (do_scrub_chars): Handle DOUBLESLASH_COMMENTS.
3513 * configure: Rebuild.
3514 * configure.in: Recognize ia64 as cpu type. Set bfd_gas.
3515 (ia64-*-elf*, ia64-*-linux-gnu*): New targets.
3516 * expr.c (expr): Handle md_optimize_expr.
3517 * read.c (LEX_HASH): Add comment.
3518 * config/tc-ia64.c, config/tc-ia64.h: New files.
3522 * config/tc-d30v.c (write_2_short): Disregard opcode1->ecc when
3523 bundling a non-delayed branch type instruction.
3527 * config/tc-mn10300.c (HAVE_AM30): Define.
3528 (md_assemble): Use it.
3532 * config/obj-elf.c (obj_elf_change_section): Check for changed
3535 * Makefile.am: (CPU_MULTI_VALID): Remove.
3536 (MULTI_CPU_TYPES): Define.
3537 (MULTI_CPU_OBJ_VALID): Define.
3538 (DEPTC): Use the above.
3539 (DEPOBJ): Same here.
3541 Regenerate dependencies.
3542 * Makefile.in: Regenerate.
3546 * Makefile.am (YACC, LEX): Get them from configure.
3550 * config/tc-i386.c (offset_in_range): Use addressT instead of
3551 bfd_vma for non-bfd assemblers.
3555 * config/tc-i386.c (offset_in_range): Sign extend val so BFD64
3556 doesn't give spurious errors.
3560 * as.h (SEEK_SET): Define if undefined.
3564 * config/tc-arm.c (md_apply_fix3): Don't use UL suffix on
3565 constants, and don't assume offsetT is 32 bits.
3569 * config/tc-d10v.h: Include "write.h" to get definition of fixS.
3570 (md_pcrel_from_section): Add prototype.
3571 (d10v_fix_adjustable): Add prototype.
3572 (d10v_force_relocation): Replace 'struct fix' with 'fixS'.
3574 * config/tc-d10v.c (md_apply_fix3): Add paren around &&.
3578 * config/tc-d10v.c (find_opcode): Correctly calculate position of
3579 symbol in frag chain.
3583 * config/tc-i386.c (fits_in_signed_byte): Change arg to offsetT.
3584 (fits_in_unsigned_byte, fits_in_unsigned_word): Ditto.
3585 (fits_in_signed_word, smallest_imm_type): Ditto.
3586 (md_assemble): Use an offsetT var to hold offsetT values, not a
3588 (offset_in_range): New.
3589 (md_assemble): Use it.
3590 (md_convert_frag): Change type of target_address, opcode_address,
3591 and displacement_from_opcode_start to offsetT.
3592 (md_create_short_jump): Change type of offset to offsetT.
3593 (md_create_long_jump): Ditto.
3594 (md_apply_fix3): Use -4, not 0xfffffffc for BFD_RELOC_386_PLT32.
3595 (md_chars_to_number): Remove.
3596 (output_invalid): Remove duplicate prototype.
3600 * Makefile.am (CPU_TYPES): Add 'avr'.
3601 (TARGET_CPU_CFILES): Add 'tc-avr.c'.
3602 (TARGET_CPU_HFILES): Add 'tc-avr.h'.
3604 * Makefile.in: Regenerate.
3606 * doc/as.texinfo: Add M32R documentation.
3610 * configure.in (WARN_CFLAGS): Set to -W -Wall by default. Add
3611 --enable-build-warnings option.
3612 * Makefile.am (AM_CFLAGS, WARN_CFLAGS): Add definitions.
3613 * Makefile.in, configure: Re-generate.
3617 * config/tc-arm.c (md_apply_fix3): Treat BFD_RELOC_ARM_PCREL_BLX
3618 in the same way as BFD_RELOC_ARM_PCREL_BRANCH, and
3619 BFD_RELOC_THUMB_PCREL_BLX lie BFD_RELOC_THUMB_PCREL_BRANCH.
3620 (tc_gen_reloc): Accept BFD_RELOC_{ARM|THUMB}_PCREL_BLX.
3621 (arm_force_relocation): Force relocations for
3622 BFD_RELOC_{ARM|THUMB}_PCREL_BLX as well.
3626 * config/tc-sh.c (get_operands): There's no third operand if the
3627 first operand is an immediate.
3631 * config/tc-sh.c (immediate): Delete.
3632 (sh_operand_info): Add immediate member.
3633 (parse_reg): Use A_PC for pc.
3634 (parse_exp): Add second argument 'op'. All callers changed.
3635 (parse_at): Expect pc to be coded as A_PC.
3636 Use immediate field in *op.
3637 (insert): Add fourth argument 'op'. All callers changed.
3638 (build_relax): Add second argument 'op'. All callers changed.
3639 (insert_loop_bounds): New function.
3640 (build_Mytes): Remove DISP_4.
3641 Split IMM_[48]{,BY[24]} into IMM[01]_[48]{,BY[24]}. Add REPEAT.
3642 (assemble_ppi): Use immediate field in *operand.
3643 (sh_force_relocation): Handle BFD_RELOC_SH_LOOP_{START,END}.
3644 (md_apply_fix): Likewise.
3645 (tc_gen_reloc): Likewise. Check for a pcrel BFD_RELOC_SH_LABEL.
3649 * config/tc-sparc.c (sparc_ip): Avoid string pasting.
3653 * internals.texi (CPU backend): Document
3654 TC_CHECK_ADJUSTED_BROKEN_DOT_WORD.
3658 * po/gas.pot: Regenerate.
3660 * as.c (show_usage): Restore translated part of bug string.
3661 * gasp.c (show_usage): Likewise.
3663 * Makefile.am (MKDEP): Use gcc -MM rather than ../mkdep.
3664 (DEP): Quote when passing vars to sub-make. Use "mv -f" rather
3665 than move-if-change.
3666 (DEP1): Modify for "gcc -MM".
3670 (CLEANFILES): Add DEPTCA, DEPOBJA, DEP2a, DEPA.
3671 Update dependencies.
3672 * Makefile.in: Regenerate.
3676 * config/tc-mn10300.c (md_pseudo_table): Use constant names.
3677 (md_begin): Likewise.
3678 (HAVE_AM33): New macro.
3679 (md_assemble): Use it. Match r_regs and xr_regs only if
3684 * as.h: #include "bin-bugs.h"
3685 * as.c (show_usage): Use REPORT_BUGS_TO.
3686 * gasp.c: #include "bin-bugs.h"
3687 (show_usage): Use REPORT_BUGS_TO.
3689 * config/tc-sparc.c (md_show_usage): Add a trailing newline.
3693 * write.c (write_object_file) [! WORKING_DOT_WORD]: If defined,
3694 call TC_CHECK_ADJUSTED_BROKEN_DOT_WORD for each word after
3699 * config/tc-i386.c (i386_immediate): Don't assume a constant
3700 immediate is necessarily 16 bits when in 16 bit code mode.
3701 (md_assemble): Instead set guess_suffix here after we have checked
3706 * config/tc-d30v.c (check_range): Allow signed or unsigned 32-bit
3707 quantities. Correct right shift sign extension.
3708 (build_insn): Make `number' unsigned long. Mask top 6 bits of
3709 32-bit value when shifting into place.
3713 * app.c: Add ATTRIBUTE_UNUSED as needed.
3714 * config/tc-ppc.c: Likewise.
3715 (ppc_size): Make unsigned long.
3716 (ppc_insert_operand): Add casts to avoid warnings.
3720 * config/tc-d10v.h (md_flush_pending_output): Define.
3724 * config/tc-sh.h (SEG_NAME): New macro: return the name of a
3725 segment. Works for both BFD_ASSEMBLER and others.
3726 (SUB_SEGMENT_ALIGN): Use SEG_NAME.
3730 * config/tc-arm.c (tinsns): Add "bal" instruction pattern.
3734 * listing.c (LISTING_LHS_WIDTH): Default depends on
3736 (LISTING_LHS_WIDTH_SECOND): Default to LISTING_LHS_WIDTH.
3740 * config/tc-sh.c (md_show_usage): Use backslash before newline in
3745 * config/tc-avr.h (TC_HANDLES_FX_DONE): Define.
3747 * config/tc-avr.c (mcu_types): Add missing initialiser.
3748 (md_pcrel_from_section): Add prototype.
3749 (avr_operand): Remove redundant test of unsigned < 0.
3750 (avr_cons_fix_new): Ensure exp_mod_pm zero on function exit.
3754 * config/tc-avr.c: New file for AVR support.
3755 * config/tc-avr.h: Likewise.
3756 * configure.in: Add AVR support.
3757 * configure: Regenerate.
3761 * gasp.c (macro_op): Add new argument to check_macro call.
3762 Macro structure definitions moved to macro.h
3763 * sb.h: Add argument to prototype for input_scrub_include_sb.
3764 * input-scrub.c (input_scrub_include_sb): Allow disabling of sb
3765 nesting checks with an additional flag.
3766 (struct input_save): Add flag to indicate whether current sb
3767 should be checked for proper macro/conditional nesting.
3768 (input_scrub_push/pop): Save/restore nest check flag.
3769 (input_scrub_next_buffer): Ditto. Also call end of macro hook if
3771 * macro.c (check_macro): Allow caller to retrieve parsed macro
3772 information if a pointer is provided. This information may be
3773 used by the new macro hooks.
3774 * macro.h: Update prototype for check_macro. Macro struct
3775 definitions moved here from macro.c/gasp.c.
3776 * read.c (read_a_source_file): Add parameter to check_macro call,
3777 and pass macro info to the macro hook, if defined.
3778 (input_scrub_insert_line): New. Allow insertion of a line of
3779 characters into the input stream.
3780 (input_scrub_insert_file): New. Allow insertion of an arbitrary
3781 file into the input stream.
3782 (s_include): Use input_scrub_insert_file.
3783 * internals.texi: Document new macro hooks.
3784 * as.h: New prototypes added.
3788 * config/tc-i386.c: Don't start any as_bad or as_warn message with
3789 an initial capital letter.
3790 (i386_index_check): Reindent.
3794 * config/tc-arm.c (md_apply_fix3): Fix bug detecting overflow of pc
3799 * config/tc-arm.c (do_t_adr): Flag "adr Rd,label"
3800 instruction operand bad if Rd > 7 when generating
3801 thumb instructions. Prevents for example,
3802 "adr r12,label" from silently failing and generating
3803 the wrong instruction.
3807 * config/tc-arm.c (md_apply_fix3): Handle same-section relocations
3808 that have a destingation >= 0x400000.
3809 Fix compile time warning messages.
3813 * config/tc-sh.c (md_begin): When encountering insn that are
3814 not supported by the current arch, only change the name if
3815 its contents are the same as prev_name.
3816 (get_specific): If the the architecture doesn't match, fail.
3820 * config/tc-sh.c (IDENT_CHAR): Define.
3821 (parse_reg): Use it instead of isalnum. Put r[0..7]_bank operand
3822 matching back where it came from.
3826 * config/tc-sh.c (md_show_usage): Add description of -dsp.
3830 * config/tc-sh.c (parse_reg): Match r[0..7]_bank operands before
3835 * config/tc-h8300.c: Add ATTRIBUTE_UNUSED as appropriate.
3839 * expr.c (operand) [case 'f']: When testing if '0f' can start a
3840 floating-point-number, make sure 'f' is in FLT_CHARS.
3844 * read.c (TC_IMPLICIT_LCOMM_ALIGNMENT): New default-definition.
3845 (s_lcomm_internal): Use it.
3846 * doc/internals.texi (CPU backend): Document it.
3847 * config/obj-evax.h (TC_IMPLICIT_LCOMM_ALIGNMENT): Set to 2**3
3852 * config/tc-mips.c (mips_ip): Don't put stuff in .rodata
3855 * config/tc-mips.c (SWITCH_TABLE): The ELF embedded-pic
3856 implementation doesn't have special handling for switch
3858 (macro_build): Allow for code in sections other than .text.
3860 (mips_ip): Likewise.
3861 (md_apply_fix): Do pc-relative relocation madness for MIPS ELF.
3862 Don't perform relocs if we will be outputting them.
3863 (tc_gen_reloc): For ELF, just use fx_addnumber for pc-relative
3864 relocations. Allow BFD_RELOC_16_PCREL_S2 relocs when
3869 * config/tc-m32r.c (m32r_fix_adjustable): Look up the
3870 relocation type based on the entry in the fixup structure.
3871 Put S_IS_EXTERN processing back in.
3875 * Makefile.am (install-exec-tooldir): Depend on
3876 install-exec-bindir for parallel make.
3877 * Makefile.in: Regenerated.
3881 * config/tc-m32r.c (struct md_longopts): Add -m32r command line
3883 (md_parse_option): Parse -m32r command line switch - disable m32rx
3885 (md_show_usage): Document new option.
3887 * doc/Makefile.am (CPU_DOCS): Add c-m32r.texi.
3888 * doc/Makefile.in: Regenerate.
3889 * doc/c-m32r.texi (M32R-Opts): Document new command line switch.
3893 * config/tc-d30v.c (check_range): Remove code that incorrectly
3894 sign extended values where bits < 32.
3898 * configure.in: Support --enable-targets=all on ia32.
3899 * configure: Regenerated.
3903 * gasp.c (do_align): Remove bogus check of alignment value.
3907 * config/obj-elf.c (elf_frob_symbol): Remove code which when
3908 TC_PPC was defined forced the type of a symbol with no other type
3913 * doc/internals.texi (CPU backend): Mention that
3914 line_separator_chars do not break up comments. Fix typos for
3915 LEX_AT and LEX_NAME descriptions. Document operands for
3916 TC_EQUAL_IN_INSN, md_operand and md_section_align. Correct
3917 description of md_create_short_jump usage. Document argument for
3918 md_undefined_symbol.
3922 * config/tc-sparc.c (OPTION_UNDECLARED_REGS): New option.
3923 (md_parse_option): Handle it.
3924 (md_show_usage): Document it.
3928 * config/tc-alpha.c (md_assemble): Accept `1' and `9' in an
3929 opcode, for the instruction `pal19'. From Andrea Arcangeli
3934 * config/tc-i386.c (i386_immediate): Move constant operand sizing
3936 (md_assemble): To here, before template operands are matched.
3937 Also ensure a constant immediate is sign extended when we know the
3938 size is at most 16 bits. This is to catch cases like "add
3939 $0xffc0,%ax" where we don't know the size, and thus that the
3940 immediate can be represented as Imm8S until after parsing the
3942 (i386_displacement): Similarly sign extend 16 bit constant
3944 (md_assemble): Relax 16-bit jump constant range check to suit sign
3945 extended displacements.
3949 * doc/c-mips.texi (MIPS Opts): Fix typo in last patch.
3953 * config/tc-i386.c (md_assemble): Don't swap intersegment jmp and
3954 call operands when intel_syntax.
3955 (intel_float_operand): Return 2 for "fi...".
3956 (i386_operand_modifier): Change "DWORD PTR" test to suit above.
3957 Return SHORT_MNEM_SUFFIX for "WORD PTR" when "fi...". Revert
3958 earlier "SHORT" change.
3959 (md_assemble): When determining suffix from Regs, exclude
3964 * configure: Add arm-wince, mips-pe and sh-pe targets.
3965 * configure: Regenerate.
3967 * config/obj-coff.h (COFF_WITH_PE): Define for mips-pe and
3969 (TARGET_FORMAT): Set to "pe-shl" for the sh-pe target and to
3970 "pe-mips" for the mips-pe target.
3972 * config/tc-arm.c (insns): Change displacement encoded in BL
3973 and B instructions if the target port is arm-wince.
3974 (do_ldst): Do not bias the relocation offset if the target
3976 (md_pcrel_from): Add in missing relocation offset bias if the
3977 target os arm-wince.
3979 * config/tc-mips.c (mips_target_format): Support COFF flavour.
3980 (md_begin): Disable -G support for mips-pe target.
3981 (md_apply_fix): Treat BFD_RELOC_RVA reloc as BFD_RELOC_32.
3982 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Add support for
3985 * config/tc-sh.c (md_begin): sh-pe target is little endian.
3986 * config/tc-sh.h (SUB_SEGMENT_ALIGN): If using a BFD
3987 assembler, just set the alignment to 4.
3989 * config/te-wince-pe.h: New file for WinCE targets. Define
3994 * config/tc-i386.c (md_assemble): Swap segments too for intel mode
3995 string instructions.
3996 (i386_operand_modifier): Set i.suffix = WORD_MNEM_SUFFIX for SHORT.
3997 (i386_intel_memory_operand): After finding a segment override,
3998 check again for no `[' before looking for a displacement. Bomb if
3999 more than one displacement rather than silently discarding the
4000 second and subsequent ones. Free strings malloc'd by
4001 build_displacement_string.
4005 * config/obj-som.c (obj_pseudo_table): Add "weak".
4006 (obj_som_weak): New routine.
4010 * config/tc-i386.c (union i386_op): New.
4011 (struct _i386_insn): Delete disps[], imms[], regs[]. Add op[].
4012 Throughout file replace occurences of disps[n], imms[n], regs[n]
4013 with equivalent op[n].disps, op[n].imms, op[n].regs. Simplify
4014 intel mode operand swapping. Add assert in regKludge and
4015 fake_zero_displacement code. Test i.types[n] when outputting
4016 displacements and immediates. Combine output of Disp16 with
4018 (md_assemble): Don't try to fix broken UNIXWARE_COMPAT opcodes
4019 when in intel mode by (not) reversing fsub and fdiv operands
4020 before the template search. This fails for single operand
4021 shorthand forms of the instruction, and if UNIXWARE_COMPAT is
4022 undefined. Instead fix the base_opcode after we've found the
4023 template. Move base_opcode xor with found_reverse_match from
4024 opcode output code to before this fix so we test for the correct
4026 (md_assemble): Don't use strcmp when deciding to ignore the suffix
4027 check in intel mode. Instead compare opcodes.
4029 * config/tc-i386.h (TC_RELOC): Delete.
4030 * config/tc-i386.c (TC_RELOC): Delete. Replace usage of TC_RELOC
4031 with equivalent call to reloc.
4033 * as.h (flag_m68k_mri): Move declaration after target include, and
4034 only declare when TC_M68K defined. Define as zero otherwise.
4035 (LABELS_WITHOUT_COLONS, NO_PSEUDO_DOT): If undefined, define as 0.
4036 * app.c (scrub_m68k_mri): Declare only when TC_M68K defined.
4037 Define as zero otherwise.
4038 (do_scrub_begin): Use m68k_mri parameter only when TC_M68K defined.
4039 (struct app_save): Declare scrub_m68k_mri only when TC_M68K.
4040 (app_push, app_pop): Save scrub_m68k_mri only when TC_M68K.
4041 (do_scrub_chars): Use LABELS_WITHOUT_COLONS directly rather than
4042 testing whether defined.
4043 * cond.c (ignore_input): Use NO_PSEUDO_DOT directly.
4044 * expr.c (operand): #ifdef unused case labels when TC_M68K undefined.
4045 * read.c: Use LABELS_WITHOUT_COLONS and NO_PSEUDO_DOT directly
4046 rather than testing whether defined.
4047 (s_mri): Set flag_m68k_mri only when TC_M68K defined.
4048 (parse_mri_cons): Declare and use only when TC_M68K.
4049 * config/tc-hppa.h (LABELS_WITHOUT_COLONS): Define as 1.
4050 * config/tc-m68k.h (NO_PSEUDO_DOT): Define as 1.
4051 * config/tc-m88k.h (NO_PSEUDO_DOT): Define as 1.
4053 * NEWS: Mention IBM 370 support.
4057 * config/tc-i386.c (md_assemble): When swapping operands for
4058 intel_syntax, assume everything that's not Imm or Disp is a
4063 * config/tc-i370.c, config/tc-i370.h: New files.
4064 * Makefile.am: Add support for Linux/IBM 370.
4065 * configure.in: Likewise.
4066 * app.c (do_scrub_begin): Don't lex single quote when TC_I370.
4067 * config/obj-elf.c: Include elf/i370.h
4068 (obj_elf_section): Don't do anything special for flag_mri if TC_I370.
4070 * Makefile.in: Regenerate.
4071 * configure: Regenerate.
4073 * doc/c-i370.texi: New file.
4074 * doc/all.texi: Include it.
4075 * doc/as.texinfo: And here.
4076 * doc/Makefile.am(CPU_DOCS): Add c-i370.texi.
4077 * doc/Makefile.in: Regenerate.
4081 * config/tc-d30v.c (parallel_ok): Use FLAG_NOT_WITH_ADDSUBppp to
4082 determine if an instruction can be used in parallel with an ADDppp
4083 or SUBppp instruction.
4087 * doc/c-mips.texi (MIPS Opts): Document -mgp32 and -mgp64.
4091 * config/tc-mips.c (mips_gp32): New variable.
4092 (macro_build) Use mips_gp32.
4094 (md_longopts): Add "-mgp32" and "-mgp64".
4095 (md_parse_option): Add OPTION_GP32 and OPTION_GP64.
4097 (OPTION_M7900): Change offset
4098 (OPTION_NO_M7900): Ditto.
4102 * config/obj-coff.c (add_lineno): Accept non-positive lineno with
4103 warning, and bump it to 1.
4108 * dwarf2dbg.c (print_stats): Add cast to force printf argument to
4113 * config/tc-mips.c (MF_HILO_INSN): Define.
4114 (mips_7000_hilo_fix): Declare.
4115 (append_insn): Conditionally insert nops after an mfhi/mflo insn.
4116 (md_parse_option): Check for 7000_HILO_FIX options.
4117 (OPTION_M7000_HILO_FIX): Define.
4118 (OPTION_NO_M7000_HILO_FIX): Define.
4119 * doc/c-mips.texi (-mfix7000): Describe.
4123 * listing.c (print_lines): Remove unused variable `end'.
4125 * config/tc-i386.c (md_assemble): Use `reloc()' to select reloc
4126 type for JumpInterSegment output. Use enum bfd_reloc_code_real for
4127 reloc_type when BFD_ASSEMBLER.
4128 (md_estimate_size_before_relax): Use enum bfd_reloc_code_real for
4129 reloc_type when BFD_ASSEMBLER. Move common code out of switch
4130 statement and quell signed vs. unsigned comparison warning.
4134 * config/tc-d10v.c (find_opcode): Add a symbol's value to
4135 the computed frag offset, rather than overwriting it.
4139 * config/tc-sh.c ("elf/sh.h"): Include.
4140 (sh_dsp, valid_arch, reg_x, reg_y, reg_efg): New static variables.
4141 (md.begin): Initialize target_arch.
4142 Only include opcodes in has table that match selected architecture.
4143 (parse_reg): Recognize register names for sh-dsp.
4144 (parse_at): Recognize post-modify addressing.
4145 (get_operands): The leading space is now optional.
4146 (get_specific): Remove FDREG_N support. Add support for sh-dsp
4147 arguments. Update valid_arch.
4148 (build_Mytes): Add support for SDT_REG_N.
4149 (find_cooked_opcode): New function, broken out of md_assemble.
4150 (assemble_ppi, sh_elf_final_processing): New functions.
4151 (md_assemble): Use find_cooked_opcode and assemble_ppi.
4152 (md_longopts, md_parse_option): New option: -dsp.
4153 * config/tc-sh.h (elf_tc_final_processing): Define.
4154 (sh_elf_final_processing): Declare.
4158 * config/tc-hppa.c (pa_build_unwind_subspace): Use subseg_new to create
4159 the unwinder subspace. Save the current seg/subseg before creating
4164 * config/tc-mcore.c (INST_BYTE0): Redefine to handle big and
4165 little endian targets.
4166 (INST_BYTE1): Redefine to handle big and little endian
4168 (cpu_type): New type: Select between M340 and M210.
4169 (parse_psrmod): New function: Parse the PSRCLR and PSRSET
4170 instructions of the M340.
4171 (md_assemble): Add support for the MULSH and OPSR classes of
4173 (md_atof): Add support for little endian targets.
4174 (md_parse_option): Add support for -EL, -EB and -mcpu command
4176 (md_convert_frag): Add support for little endian targets.
4177 (md_apply_fix3): Add support for little endian targets.
4178 (md_number_to_chars): Add support for little endian targets.
4182 * read.c (read_a_source_file): If TC_START_LABEL_WITHOUT_COLON is
4183 defined, use it to verify the symbol just read should be a label.
4187 * app.c (do_scrub_chars): Handle "||" for parallel instructions
4188 when DOUBLEBAR_PARALLEL is defined. Avoid stripping whitespace
4189 around colons when KEEP_WHITE_AROUND_COLON is defined.
4190 * doc/internals.texi (CPU backend): Document DOUBLEBAR_PARALLEL
4191 and KEEP_WHITE_AROUND_COLON.
4195 * read.c (s_rept): Call do_repeat, which abstracts the repeat
4197 (do_repeat): New. Abstract repeat logic so that a "break" can be
4199 (end_repeat): New. Provide support for a "break" out of the
4201 * read.h: Add prototypes for new functions.
4205 * doc/internals.texi: Document NUMBERS_WITH_SUFFIX macro.
4206 * as.h: Provide a default NUMBERS_WITH_SUFFIX definition (zero).
4207 * expr.c: Handle numbers with suffixes if NUMBERS_WITH_SUFFIX is
4212 * read.c: Added elseif to directives table.
4213 * read.h: Added prototype for s_elseif.
4214 * doc/as.texinfo: Added description for elseif.
4215 * cond.c (s_elseif): New function
4219 * listing.c (print_lines): Remove conditionals causing bug in
4224 * as.h: Define OCTETS_PER_BYTE and OCTETS_PER_BYTE_POWER
4226 * frags.c (frag_new): Calculate fr_fix in octets
4227 (frag_now_fix) Return offset as target address offset (bytes).
4228 (frag_now_fix_octets) New - Return offset in octets (8-bit
4230 * frags.h: Added prototype for frag_now_fix_octets().
4231 Distinguish between octets and bytes in field descriptions.
4232 * listing.c (calc_hex): Account for octets vs bytes when
4233 printing addresses/offsets.
4234 (print_lines) Ditto. Also, if LISTING_WORD_SIZE is not 1, and
4235 target is little-endian, print the octets in a word in big-endian
4236 order so that the display looks like a proper hexadecimal number,
4237 instead of having the octets reversed.
4238 * read.c (do_align): When recording alignment, alignment power
4239 should be in terms of target bytes (minimum addressible unit)
4241 (do_org) Convert ORG target address (byte) argument into an
4242 octet offset when generating a variable fragment.
4243 * symbols.c (resolve_symbol_value): Symbol final value
4244 converted to a target address offset (bytes) from its octet offset.
4245 * config/obj-coff.c (coff_frob_symbol): Symbol target address
4246 offset (bytes) is adjusted by the frag offset (octets) converted
4248 (coff_frob_section) Section alignment power is in terms of bytes;
4249 convert it to an octet alignment power when calculating size (and
4250 size mask) in octets. Don't modify the section size in order to
4251 "align" it for TI COFF, since that format has a different method
4252 for storing alignment information.
4256 * stabs.c (generate_asm_file): Escape backslashes in stabs file
4257 entries, matching the way GCC generates them. If not escaped, the
4258 filename is encoded incorrectly.
4262 * config/tc-arm.c (reg_table): Add support for ATPCS register
4267 * config/obj-coff.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Don't define if
4269 * config/tc-ppc.h [OBJ_XCOFF] (OBJ_COPY_SYMBOL_ATTRIBUTES):
4271 * config/tc-ppc.c (ppc_fix_adjustable): Don't look at the frag
4272 of a symbol when we really care about its value.
4276 * config/tc-mcore.c (md_assemble): Give warning message if
4277 operands passes to instruction are more than the spec.
4281 * config/tc-arm.c (armadjust_symtab): If the assembler is in
4282 Thumb mode but the label seen was not declared as '.thumb_func'
4283 then set the ST_INFO type to STT_ARM_16BIT mode. This allows
4284 correct disassembly of Thumb code bounded by non function labels.
4288 * Makefile.am (MULTI_CFILES): Add config/e-i386aout.c
4289 Add dependencies for e-i386aout.o. Fix 2 comment lines.
4291 * Makefile.in: Same here.
4294 * configure.in: Set bfd_gas for i386-aout when primary target
4295 is bfd. Handle i386aout emulation. Don't use te_file=multi, as
4296 we may need the primary te_file. Remove incorrect comment.
4298 * configure: Regenerate.
4300 * config/e-i386aout.c: New file.
4302 * as.c (USE_EMULATIONS): Move to before print_version_id.
4303 (struct emulation): Add i386aout.
4304 (show_usage): Split text strings. Reformat -a text. Add --em
4308 * obj.h (struct format_ops): Add s_get_other and s_get_desc.
4309 (aout_format_ops): New.
4312 * read.c (s_lcomm_internal): Rewrite OBJ_AOUT,OBJ_BOUT
4313 preprocessor conditional and add aout USE_EMULATIONS tests.
4314 (read_a_source_file): Don't pass error strings to printf as
4318 * gasp.c (exp_get_abs): Don't pass error strings to printf as
4320 (do_data): Same here.
4321 (process_file): And here.
4324 * symbols.c (colon): Rewrite "already defined" fatal message
4325 code for aout with USE_EMULATIONS.
4328 * config/obj-aout.c (OBJ_HEADER): Define.
4329 (obj_pseudo_table): Rename to aout_pseudo_table. Init all
4331 (obj_aout_frob_symbol): Expand S_GET_DESC, S_GET_TYPE,
4332 S_GET_OTHER, S_SET_TYPE macros since we don't need obj-multi
4334 (obj_aout_type): Expand S_SET_OTHER here too.
4335 (obj_read_begin_hook): Remove.
4336 (aout_pop_insert): New.
4337 (obj_aout_s_get_other): New.
4338 (obj_aout_s_get_desc): New.
4339 (aout_format_ops): New.
4342 * config/obj-aout.h (obj_pop_insert): Define so non-multi usage
4343 gets aout_pseudo_table.
4344 (aout_pseudo_table): Declare.
4345 (obj_read_begin_hook): Define.
4348 * config/obj-coff.c (obj_pseudo_table): Rename to
4350 (coff_pop_insert): Use coff_pseudo_table.
4351 (coff_sec_sym_ok_for_reloc): Remove.
4352 (coff_format_ops): Add 0 entries for s_get_size, s_set_size,
4353 and comment all zero entries and remove #if 0 code.
4356 * config/obj-coff.h (obj_pop_insert): Define.
4357 (coff_pseudo_table): Declare.
4360 * config/obj-ecoff.c (ecoff_format_ops): Add 0 entries for
4361 s_get_size, s_set_size. Comment all zero entries.
4364 * config/obj-elf.c (elf_s_get_other): New function.
4365 (obj_read_begin_hook): Rename to elf_obj_read_begin_hook.
4366 (obj_symbol_new_hook): Rename to elf_obj_symbol_new_hook.
4367 (elf_format_ops): Add elf_s_get_other, 0 s_get_size entry, and
4369 (obj_elf_parse_section_letters): Don't pass error strings to
4370 printf as format arg.
4373 * config/obj-elf.h (ECOFF_DEBUGGING): Define when
4375 (elf_s_get_other): Declare.
4376 (S_GET_OTHER) Define as elf_s_get_other if not already
4378 (S_SET_OTHER): Only define when not already defined.
4379 (elf_obj_read_begin_hook): Declare.
4380 (obj_read_begin_hook): Define.
4381 (elf_obj_symbol_new_hook): Declare.
4382 (obj_symbol_new_hook): Define.
4385 * config/obj-multi.h: Add copyright header and protect against
4386 multiple inclusion. Add * to all function pointers.
4387 (OBJ_HEADER): If defined, include it rather than other defines
4389 (obj_frob_file_after_relocs): Test for NULL.
4390 (obj_symbol_new_hook): Here too.
4391 (obj_sec_sym_ok_for_reloc): And here.
4392 (S_GET_OTHER): Define.
4393 (S_GET_DESC): Define.
4394 (ECOFF_DEBUGGING): Remove as it's done in obj-elf.h
4395 (OBJ_MAYBE_ELF): Update comment.
4397 * config/tc-i386.c (i386_immediate): Add OBJ_MAYBE_AOUT to
4398 OBJ_AOUT preprocessor conditional and handle emulation by
4399 testing OUTPUT_FLAVOR.
4400 (i386_displacement): Here too.
4401 (md_section_align): Similarly here.
4402 (i386_target_format): Conditionally compile when more than one
4403 of OBJ_MAYBE_{ELF,COFF,AOUT} defined. Add aout case.
4404 (i386_immediate): Fix error message for aout BFD_ASSMBLER.
4405 (i386_displacement): Here too.
4408 * config/tc-i386.h (AOUT_TARGET_FORMAT): Define for each TE_*.
4409 Define TARGET_FORMAT for aout only when not multi.
4412 * config/te-multi.h: Delete file as it's identical to te-generic.h
4416 * config/tc-i386.h (DWORD_MNEM_SUFFIX): Delete.
4417 * config/tc-i386.c (DWORD_MNEM_SUFFIX): Rename all occurrences to
4420 * config/tc-i386.h (INTEL_DWORD_MNEM_SUFFIX): Rename to
4422 * config/tc-i386.c (INTEL_DWORD_MNEM_SUFFIX): Here too. Fix some
4427 * config/tc-mips.c (mips_do_align): New function.
4428 * config/tc-mips.h (md_do_align): Define.
4432 * doc/c-arm.texi (ARM Options): Fix typo.
4433 (ARM-Chars): Correct description of `#'. Mention that `;' is a
4434 line separator for Linux.
4435 * doc/as.texinfo (Comments): Mention the ARM.
4439 * configure.in (arm*-*-conix*): New target.
4440 (arm*-*-linux-gnu*): Match instead of arm-*-linux* and
4442 * configure: Regenerate.
4446 * config/obj-elf.c (elf_pseudo_table): Define visibility pseudos.
4447 (obj_elf_visibility): New function.
4449 * doc/as.texinfo (Visibility): New node: document visibility
4452 For older changes see ChangeLog-9899