3 * config/tc-ia64.c (errata_nop_necessary_p): Prototype.
4 (make_unw_section_name): Constify local variables.
6 * config/tc-ia64.c (has_suffix_p): New.
7 (note_register_values): Use it instead of strstr.
11 * config/tc-hppa.c (pa_ip): Handle new 'c' mode completers,
16 * config/tc-sparc.h (TC_PARSE_CONS_EXPRESSION): Define.
17 (sparc_cons): Provide prototype.
18 * config/tc-sparc.c (tc_gen_reloc): Handle BFD_RELOC_*_PCREL and
19 BFD_RELOC_SPARC_PLT{32,64}. Enumerate for which relocs
20 reloc->addend = fixp->fx_addnumber shouldn't be done instead of
21 enumarating for which pc relative ones it should be done.
22 (sparc_cons_special_reloc): New variable.
23 (sparc_cons): New function.
24 (cons_fix_new_sparc): Use sparc_cons_special_reloc.
29 * config/tc-mips.c (file_mips_gp32): Initialize to invalid value.
30 (file_mips_fp32): Likewise.
31 (md_begin): Compatibility handling for -mipsN option.
35 * config/tc-sh.c (parse_at): Reject @(r0) and @(r0,).
39 * config/tc-ppc.c (md_parse_option): Make -maltivec default
40 to generating PowerPC instructions.
44 * configure.in (mips-dec-netbsd*): Delete.
45 * configure: Regenerate.
47 * configure.in (arm-*-netbsdelf*): Add target.
48 * configure: Regenerate.
52 * config/tc-mips.h (TC_HANDLE_FX_DONE): Remove redundant
57 * configure.in (assign object format): Bug fix for setting
59 * configure: Regenerate.
63 * configure.in (m68k-*-netbsdelf*): New target.
64 (m68k-*-netbsd*): Also include ELF support.
65 (m68k-*-netbsdaout*): New alias for m68*-*-netbsd*.
66 * configure: Regenerate.
70 * gasp.c (main): Protoype.
74 * config/tc-alpha.c (alpha_handle_align): Encode unop with RB as $sp.
78 * cgen.c: Add prototype for queue_fixup.
79 (gas_cgen_parse_operand): Move initilisastion of errmsg to avoid
80 possible longjmp corruption.
81 * cgen.h: Add prototype for gas_cgen_md_operand.
85 * config/obj-elf.c (obj_elf_init_stab_section): References are
86 kept to section name strings. Don't alloca them!
90 * config/tc-d10v.c (get_operands): Mark OPERAND_PLUS after
91 OPERAND_ATSIGN as O_absent.
96 * configure.in: Add support for xstormy16.
97 * configure: Regenerated.
98 * Makefile.am: Add support for xstormy16.
99 * Makefile.in: Regenerated.
100 * config/tc-xstormy16.c: New file.
101 * config/tc-xstormy16.h: New file.
105 * tc-arm.c (do_arit, do_cmp, do_mov, do_ldst, do_ldstt, do_ldmstm)
106 (do_branch, do_swi, do_adr, do_adrl, do_empty, do_mul, do_mla)
107 (do_swap, do_msr, do_mrs, do_mull, do_ldstv4, do_bx, do_blx)
108 (do_bkpt, do_clz, do_lstc2, do_cdp2, do_co_reg2, do_smla, do_smlal)
109 (do_smul, do_qadd, do_pld, do_ldrd, do_co_reg2c, do_cdp, do_lstc)
110 (do_co_reg, do_fpa_ctrl, do_fpa_ldst, do_fpa_ldmstm, do_fpa_monadic)
111 (do_fpa_dyadic, do_fpa_cmp, do_fpa_from_reg, do_fpa_to_reg, do_mia)
112 (do_mar, do_mra, do_c_binops, do_c_binops_1, do_c_binops_2)
113 (do_c_binops_3, do_c_triple, do_c_triple_4, do_c_triple_5, do_c_quad)
114 (do_c_quad_6, do_c_dspsc, do_c_dspsc_1, do_c_dspsc_2, do_c_shift)
115 (do_c_shift_1, do_c_shift_2, do_c_ldst, do_c_ldst_1, do_c_ldst_2)
116 (do_c_ldst_3, do_c_ldst_4, do_branch25): Delete redundant argument,
118 (struct asm_opcode): Adjust parms field accordingly.
119 (md_assemble): Don't pass dummy second argument when calling worker
121 (build_arm_ops_hsh): Add prototype
126 * tc-arm.c (struct asm_opcode): Delete comp_suffix and flags. Add
127 cond_offset. Rename variants->variant.
128 (insns): Adjust for new format. Explicitly code each variant that
129 takes flags. Remove temporary instructions.
130 (struct arm_it): Remove redundant field suffix.
131 (s_flag, ldr_flags, str_flags, byte_flag, cmp_flags, ldm_flags)
132 (stm_flags, lfm_flags, sfm_flags, round_flags, fix_flags, except_flag)
134 (struct asm_flg): Delete.
135 (LONGEST_INST): Delete.
136 (V4_STR_BIT): Define.
137 (struct thumb_opcode): Rename variants->variant.
138 (do_empty): Renamed from do_nop.
139 (ldst_extend): Delete argument hwse. Split code for half-word and
140 signed byte instructions to ...
141 (ldst_extend_v4): ... here.
142 (ld_mode_required_here): Use ldst_extend_v4.
143 (do_ldrd): Simplify now that this is only called for ldrd. No
144 need to test for XScale, which was wrong anyway. Don't reject r12
145 as a target register. Add test that ldrd doesn't update an index
147 (do_pld): Don't allow post-indexed or write-back addressing modes.
148 Adjust call to ldst_extend.
149 (do_adr): Split code for adrl to ...
151 (do_cmp): No need to fold in COND_BIT.
152 (do_ldst): Simplify. Split code for ldrt/strt into do_ldstt. Split
153 code to handle half-word and signed byte instructions to ...
154 (do_ldstv4): ... here.
155 (do_ldstt): New function. Handle load/store with translate.
156 (do_ldmstm): Write feature modification bits directly into
158 (do_fpa_ldst): Remove suffix handling code.
159 (do_fpa_dyadic, do_fpa_monadic, do_fpa_from_reg): Likewise.
160 (do_fpa_ldmstm): Type of access is now held in inst.instruction.
161 (build_arm_ops_hsh): New function.
162 (md_begin): Call it. Don't build the ARM opcode directly.
163 (md_assemble): Simplify ARM instruction handling.
167 * tc-sh.c (parse_at): Corrected incorrect op->type selection due
172 * config/tc-d10v.c (write_2_short): Don't skip dummy fixups, so
173 that we can tell which operand refers to the insn put in the L
174 container and mark it as such, so that the relocation type can be
179 * config/tc-mips.c (mips_cpreturn_offset): Better comment.
180 (load_register): Better error message. Cast away signedness
181 mismatches. Add casts needed for varargs.
182 (load_address): Replace checks of HAVE_64BIT_ADDRESS with dbl.
183 Remove superfluous casts.
184 (macro): Cast away signedness mismatches. Remove superfluous casts.
185 (s_cpload): Fix wrong comment.
186 (s_mips_weakext): Standardize output message.
187 (get_number): Likewise.
191 * config/tc-mips.c (append_insn): Add jump address range overflow
196 * config/tc-mips.c (md_parse_option): Fix comment. Allow -mabi option
198 (show): Document -mabi option.
202 * config/tc-arm.c (md_parse_option): Only clear cpu part when
203 specifying 'xscale' cpu (don't change the fpu part).
207 * config/tc-d10v.c (find_opcode): Reject SP operand if
208 OPERAND_NOSP flag is present.
212 * config/tc-mips.c (tc_gen_reloc): One missed BFD_RELOC_MIPS_GPREL.
216 * config/tc-mips.c (macro_build): Replace BFD_RELOC_MIPS_GPREL
217 by BFD_RELOC_GPREL16.
218 (load_address): Likewise.
220 (md_apply_fix): Likewise. Replace BFD_RELOC_MIPS_GPREL32 by
222 (s_gpword): Replace BFD_RELOC_MIPS_GPREL32 by BFD_RELOC_GPREL32.
223 (tc_gen_reloc): Replace BFD_RELOC_MIPS_GPREL by BFD_RELOC_GPREL16.
224 *config/tc-mips.h: Replace BFD_RELOC_MIPS_GPREL by BFD_RELOC_GPREL16.
228 * config/tc-d10v.c (get_operands): Emit OPERAND_PLUS for
231 * config/tc-d10v.c (find_opcode): Reject non-SP operand if
236 * tc-sh.c (parse_at): Removed case-sensitivity of index register
241 * write.c (adjust_reloc_syms): Mark SEC_MERGE symbols as used
242 in reloc if it has non-zero addend.
243 * config/tc-alpha.c (tc_gen_reloc): Reinstall SEC_MERGE check.
244 * config/tc-sparc.c (md_apply_fix3): Likewise.
248 * as.c (parse_args): Call md_after_parse_args if defined.
249 * config/tc-ia64.h (md_after_parse_args): Define.
250 * config/tc-ia64.c (ia64_after_parse_args): Reject --gstabs.
251 * doc/internals.texi (CPU backend): Document md_after_parse_args.
255 * config/tc-alpha.c (tc_gen_reloc): Remove SEC_MERGE test.
256 * write.c (adjust_reloc_syms): Don't handle relocs against
257 SEC_MERGE section symbols specially.
258 (fixup_segment): Likewise.
262 * config/tc-mips.c (mips_need_elf_addend_fixup): New, extracted from...
263 (md_apply_fix3): ...here. Don't prevent the symbol value being
264 subtracted twice from GPREL addends.
265 (tc_gen_reloc): Add the symbol value to a GPREL addend if it was
266 subtracted by the previous function.
270 * config/tc-avr.c (md_apply_fix3): Fix typo introduced by
271 md_apply_fix3 change.
275 * config/tc-m88k.c (md_apply_fix3): Match local variable `val' to
276 usage after md_apply_fix3 cleanups.
280 * doc/c-mmix.texi: Use texinfo 4 features.
281 (MMIX-mmixal): Fix typo.
285 * config/tc-cris.c (md_apply_fix3): Cast value, not pointer, in
290 * stabs.c (stabs_generate_asm_lineno): Remember file and line number
291 from one call to the next, and eliminate consecutive duplicates
292 (thereby emitting only one line symbol per source line).
293 * dwarf2dbg.c (dwarf2_gen_line_info): Ditto.
297 * config/tc-m68k.c (md_apply_fix3): Change val back to a signed type.
301 * config/tc-hppa.c (md_apply_fix3): Fix a typo.
305 * config/tc-alpha.c (md_apply_fix3): Fix a typo.
309 * write.c (fixup_segment): Remove references to md_apply_fix and
311 * cgen.c, config/obj-coff.c, config/tc-*.c: Update all occurances
312 of md_apply_fix to md_apply_fix3.
313 Make all md_apply_fix3 functions void.
314 * cgen.h, config/obj-coff.h, tc-*.h: Remove all definitions of
315 MD_APPLY_FIX3 and TC_HANDLE_FX_DONE.
316 * doc/internals.texi: Update references to md_apply_fix3.
317 * tc.h: Update prototype for md_apply_fix3.
321 * config/tc-i386.h (md_end): Define.
322 (i386_elf_emit_arch_note): Declare.
323 (CpuUnknown): Delete.
324 * config/tc-i386.c (default_arch): Constify.
325 (smallest_imm_type): Remove CpuUnknown test.
326 (md_assemble): Don't bother checking cpu_arch_flags non-zero.
327 (i386_elf_emit_arch_note): New function.
329 * po/POTFILES.in: Regenerate.
333 * config/tc-mips.c (mips_ip): Re-allow %hi() op for non-ELF assembler.
337 * frags.c (frag_grow): Revert last change.
339 * dwarf2dbg.c (get_frag_fix): Align last frag size.
341 * config/tc-ppc.c (ppc_insert_operand): Pass (ppc_cpu | ppc_size)
343 (md_assemble): Likewise.
347 * frags.c (frag_grow): Use frag_make_room() to grow the
352 * config/tc-sparc.c (md_apply_fix3): Do not remove the symbol's
353 value from the addend for fixups against local symbols in
354 SEC_MERGE sections - it was not added in, in the first place.
358 * write.c (relax_segment): Convert symbol address into an octet
359 offset prior to adding to the frag address, which is an octet
360 offset. Add comments to that effect.
361 * config/tc-tic54x.c: Fix bugs causing tests to fail.
362 * config/tc-tic54x.h: Ditch PARAMS macro.
366 * config/obj-elf.c (obj_elf_version): Ensure terminating NUL is
367 put in note section. Use sizeof instead of hard-coded constants.
371 * config/tc-mips.c (support_64bit_objects): Define for OBJ_ELF only.
372 (md_longopts): Allow OPTION_MABI for ELF compilation only. RE-allow
373 OPTION_GP32, OPTION_GP64, OPTION_FP32 for non-ELF compilation.
374 Sort options a bit more logical.
375 (md_parse_option): Allow OPTION_32, OPTION_N32, OPTION_N64,
376 OPTION_MABI only for elf targets.
380 * config/tc-mips.c (set_at): Add cast needed for varargs.
381 (load_register): Likewise.
382 (macro): Likewise. Some code reformatting.
383 (macro2): Add cast needed for varargs.
384 (mips16_macro): Likewise.
388 * config/tc-mips.c (md_assemble): Remove superflous casts.
389 (append_insn): Likewise.
390 (mips16_macro_build): Likewise.
392 (mips16_ip): Likewise.
393 (s_cpload): Likewise.
394 (mips_relax_frag): Likewise.
398 * config/tc-mips.c (mips_abi_level): Move in front of
400 (mips_set_options): Add members gp32, fp32, abi.
401 (file_mips_gp32): New flag.
402 (file_mips_fp32): New flag.
403 (mips_opts): Initialize the new members.
406 (HAVE_32BIT_GPRS): Use the new values from mips_opts.
407 (HAVE_32BIT_FPRS): Likewise.
408 (HAVE_NEWABI): Likewise.
409 (HAVE_64BIT_OBJECTS): Likewise.
410 (md_begin): Likewise. Save default (file) values.
411 (md_parse_option): Use the new values from mips_opts.
412 (s_mipsset): Likewise. Fix logic to keep the ABI selection if
413 possible. Let .set mipsN work together with .set push/pop.
414 Enhance error messages.
415 (mips_elf_final_processing): Use file_mips_* for header processing.
419 * config/tc-avr.c (mcu_types): Update for new devices.
423 * doc/Makefile.am (POD2MAN): Use 'GNU Development Tools' for
425 * doc/Makefile.in: Rebuild.
426 * doc/as.texinfo: Do not put man SEEALSO in document;
427 Use @command for commands, @option for options; Reorganize usage
428 to clearly identify target specific options.
432 * config/tc-mips.c (mips_force_relocation): Remove duplicate code.
436 * config/tc-mips.c (my_getSmallParser): Fix small parser bug.
440 * config/tc-s390.c (tc_s390_force_relocation): Force all relocations
441 that need the global offset table.
442 (md_apply_fix3): Add code to undo opertions done in fixup_segment
443 triggered by TC_FORCE_RELOCATION.
444 * config/tc-s390.h (TC_FORCE_RELOCATION): Replace macro code with
445 a call to function tc_s390_force_relocation.
446 (TC_FIX_ADJUSTABLE): Define.
450 * configure.in (mips-*-netbsd*): Add support for target.
451 * configure: Regenerate.
455 * configure.in (ALL_LINGUAS): Add 'fr'.
456 * configure: Regernate.
457 * po/fr.po: New file. Obtained from the translation project web
462 * tc-arm.c: Re-arrange prototypes by architecture.
463 (insns): Re-arrange instructions by archtitecture. Pld instruction
465 (tinsns): blx and bkpt are part of ARMv5T.
466 (do_fp_{ctrl,ldst,ldstm,dyadic,monadic,cmp,from_reg,to_reg}): Rename
467 to do_fpa_*. All callers changed.
469 * tc-arm.c (insns): Add two temporary instructions to handle
474 * config/tc-m68hc11.c (build_jump_insn): Allocate worst case storage
475 for bra/bsr and use frag_variant(), this ensure that the possible
476 16-bit BFD_RELOC_16 will be in the same frag.
480 * config/tc-mips.c (HAVE_32BIT_ADDRESSES): If compiling embedded
481 PIC code, assume pointers the same size as GPRs.
482 (macro): In M_LA_AB handling for embedded PIC code, support
483 "la $treg,foo-bar($breg)". In load/store handling
484 (label ld_st) support "<op> $treg,<sym>-<local_sym>($breg)"
485 which is used by the compiler for switch statements.
486 In load/store double multi-instruction macro handling
487 (label ldd_std) add a comment that no special handling
488 is currently done for embedded PIC.
489 (mips_ip): In 'o' (16-bit offset) case, only accept 16
494 General cleanup of feature definitions.
495 * tc-arm.c (ARM_EXT_LONGMUL, ARM_EXT_HALFWORD, ARM_EXT_THUMB): Delete.
496 (ARM_2UP, ARM_ALL, ARM_3UP, ARM_6UP): Delete.
497 (FPU_CORE, FPU_FPA10, FPA_FPA11, FPU_ALL, FPA_MEMMULTI): Delete.
498 (ARM_EXT_V{1,2,2S,3,3M,4,4T,5T,5ExP}): New defines.
499 (ARM_EXT_V{5,5E}): Synchronize with above.
500 (ARM_ARCH_V*): Define a complete set in terms of above features.
501 (ARM_{1,2,3,250,6,7,8,9,STRONG}): Define in terms of architecture.
502 (FPU_FPA_EXT_V[12]): Define.
503 (FPU_ARCH_FPE, FPU_ARCH_FPA): Define in terms of above.
505 (FPU_DEFAULT): Default to FPA.
506 (CPU_DEFAULT): For XScale, this is now just ARM_ARCH_XSCALE; for
507 Thumb, this is now ARM_ARCH_V5T.
508 (insns): Rework for new feature defines.
510 (opcode_select, do_ldst, md_begin, md_parse_option): Likewise.
514 * configure.in: Handle sh*eb-*-linux* and sh*-*-linux*.
515 * configure: Regenerate.
516 * Makefile.in: Regenerate.
517 * doc/c-sh.texi (SH Options): Added descriptions.
521 * configure.in: Update for MMIX port.
522 * Makefile.am: Ditto. Regenerate dependencies.
523 * Makefile.in: Regenerate.
524 * configure: Regenerate.
525 * config/tc-mmix.h, config/tc-mmix.c: New files.
526 * doc/Makefile.am (CPU_DOCS): Add c-mmix.texi
527 * doc/Makefile.in: Regenerate.
528 * doc/all.texi: @set MMIX.
529 * doc/as.texinfo: Ditto. Add MMIX gas manpage option overview.
531 * doc/c-mmix.texi: New file.
535 * config/tc-mips.c (hilo_interlocks, cop_interlocks): Make
536 these evaluate to true if mips_arch indicates SB-1.
540 * as.c (parse_args): Print the date as well for --version.
544 * config/tc-mips.c (md_show_usage): Print "sb1" for Broadcom
545 SB-1 CPU for consistency.
546 (mips_cpu_info_table): Tweak comment about SB-1.
550 * config/tc-arm.c (do_c_shift): Use ISDIGIT instead of isdigit.
551 (cirrus_parse_offset): Likewise.
555 * config/tc-arm.c: Restore line_comment_chars.
559 * config/tc-arm.c (comment_chars): Remove semi-colon accidentally
564 * write.c (fixup_segment): Handle fixups for SEC_MERGE sections as
565 for undefined symbols.
569 * doc/as.texinfo (MIPS ISA options): Added accidentally
570 omitted "-mips64" option to list of options.
574 * config/tc-ppc.c (md_show_usage): Add missing -maltivec, -m7400,
575 -m7410, -m7450 and -m7455 options.
579 * config/tc-ppc.c (PPC_HA, PPC_HIGHERA, PPC_HIGHESTA): Simplify.
580 (ppc_size): Select PPC_OPCODE_64 if 64 bit.
581 (md_begin): Don't set ppc_size here.
582 (ppc_target_format): Test ppc_size as well as BFD_DEFAULT_TARGET_SIZE.
583 (md_shortopts): Constify.
584 (md_longopts): Likewise.
585 (md_longopts_size): Likewise.
586 (ppc_elf_suffix): Only allow 64-bit relocs when ppc_size specifies
588 (ppc_machine): Explain why this function is a nop.
592 * bit_fix.h: Comment typo fix.
593 * config/tc-mips.c: Likewise.
594 * config/tc-sh.c: Likewise.
595 * config/tc-arc.c (arc_code_symbol): Remove unnecessary test.
597 * configure.in (MIPS_STABS_ELF): AC_DEFINE in only one place so
598 that autoheader doesn't duplicate config.in entries.
599 (DEFAULT_ARCH): Ditto.
600 * configure: Regenerate.
601 * config.in: Regenerate.
606 * config/tc-m68k.c (m68k_ip): Correct absolute jmp opcodes.
611 * config/tc-sh.c (shl): Remove.
613 (little): Remove shl handling. Emit error for endian mismatch.
614 (md_show_usage): Add description of -big.
615 (md_parse_option): Handle OPTION_BIG. Remove shl handling.
617 (md_pseudo_table): Add .big.
618 (md_longopts): Add -big.
619 (md_begin): Don't set target_big_endian here.
620 * config/tc-sh.h (TARGET_BYTES_BIG_ENDIAN): Remove.
621 (LISTING_HEADER, COFF_MAGIC, TARGET_FORMAT): Use target_big_endian.
623 * configure.in (endian): Default is big.
624 (sh-*-pe*): Little endian.
625 (cpu_type): Set sh for target sh*.
626 * configure: Regenerate.
630 * config/tc-mips.c (md_apply_fix): Preliminary handling of NewABI
635 * config/tc-arm.c: Replace CONST with const.
637 * cgen.c: Tidy up formatting.
641 * config/tc-ppc.c (md_parse_option): New -m7410, -m7450 and -m7455
642 flags, equivalent to -m7400. New -maltivec to enable AltiVec
643 instructions. New -mbook64 and -mbooke/-mbooke32 flags to enable
644 64-bit and 32-bit BookE support, respectively. Change -m403 and
645 -m405 to set PPC403 option.
646 (md_show_usage): Adjust for new options.
647 * doc/all.texi: Set PPC.
648 * doc/as.texinfo: Add PPC support and pull in c-ppc.texi.
649 * doc/c-ppc.texi: New file.
650 * doc/Makefile.am (CPU_DOCS): Add c-ppc.texi.
651 * doc/Makefile.in: Regenerate.
655 * read.c (s_bad_endr): New function. Issues a warning message
656 about a bad use of the .rept pseudo op.
657 (po_table[]): Add .endr.
658 * read.h: Prototype s_bad_endr.
660 * CONTRIBUTORS: Update contact email address.
664 * config/tc-alpha.c: Fix comment typos.
665 * config/tc-cris.c: Likewise.
666 * config/tc-hppa.c: Likewise.
667 * config/tc-i370.c: Likewise.
668 * config/tc-mips.c: Likewise.
669 * config/tc-mn10200.c: Likewise.
670 * config/tc-mn10300.c: Likewise.
671 * config/tc-sh.c: Likewise.
672 * config/tc-tahoe.c: Likewise.
673 * config/tc-v850.c: Likewise.
677 * app.c: Fix comment typos.
678 * bit_fix.h: Likewise.
680 * itbl-ops.c: Likewise.
684 * input-file.c: Include safe-ctype.h.
685 (input_file_open): Use ISSPACE instead of isspace.
689 * config/tc-mips.c (mips_cpreturn_offset): New variable.
690 (mips_cpreturn_register): Likewise.
691 (mips_gp_register): Likewise.
692 (s_cpsetup): New function prototype.
693 (s_cplocal): Likewise.
694 (s_cpreturn): Likewise.
695 (s_gpvalue): Likewise.
696 (mips_pseudo_table): Add .cpsetup, .cplocal, .cpreturn, .gpvalue
698 (macro): Don't warn about .cprestore for NewABI.
699 (md_pcrel_from): Code cleanup.
700 (mips_force_relocation): Force output of some NewABI relocations even
701 without a defined symbol.
702 (s_cpload): Ignore .cpload for NewABI.
703 (s_cpsetup): Handle .cpsetup.
704 (s_cplocal): Handle .cplocal.
705 (s_cprestore): Ignore .cprestore for NewABI.
706 (s_cpreturn): Handle .cpreturn.
707 (s_gpvalue): Handle .gpvalue.
708 (s_cpadd): Ignore .cpadd for NewABI.
709 (nopic_need_relax): Take g_switch_value into account as gp
711 (tc_gen_reloc): Don't handle BFD_RELOC_MIPS_{CALL,GOT}* for NewABI.
715 * config/tc-mips.c (my_getSmallParser): New function prototype.
716 (small_ex_type): Named this enum, more return values for
717 my_getSmallExpression.
718 (mips_ip): Allow SPC and HT between arguments. Handle some NewABI
719 triple relocations. Protect some parts with ifdef OBJ_ELF.
720 (percent_op_match): New struct, lookup table for %some_reloc().
721 (my_getSmallParser): New function, parses nested percent_ops also.
722 (my_getSmallExpression): Rewite to support nested percent_ops.
727 (load_address): Support both 32- and 64-bit addresses.
728 (macro): Call load_register correctly. Expand 64-bit loads ans stores.
729 (macro2): Call load_address correctly.
733 * config/tc-z8k.c: Include opcodes/z8k-opc.h after bfd.h
734 since z8k-opc.h now uses the PARAMS macro.
738 * config/tc-arm.c: Change MAVERIK to MAVERICK.
742 * config/obj-elf.c: Make use of elf_group_name and elf_next_in_group
744 (obj_elf_change_section): Rename "group" to "group_name".
745 (obj_elf_section): Likewise.
746 (elf_frob_file): Don't use sec->lineno for SHT_GROUP section to store
747 first member section; Instead use elf_next_in_group.
751 * config/obj-elf.c: (obj_elf_change_section): Add "group" param.
752 Set elf_section_data group from it. Warn if group name changed.
753 (obj_elf_parse_section_letters): Parse 'G' too.
754 (obj_elf_section): Parse group name.
755 (struct group_list): New.
756 (build_group_lists): New function.
757 (elf_frob_file): Create SEC_GROUP section(s).
759 * config/obj-elf.c: (elf_copy_symbol_attributes): Zap trailing
764 * doc/arm/c-arm.texi (ARM Options): Add arm9e documentation.
766 * config/tc-arm.c (ARM_EXT_MAVERIK): New macro.
767 (cirrus_regtype): New enum.
768 (LONGEST_INST): Change to 10.
775 (insns): Add cirrus dsp instructions.
776 (ARM_EXT_MAVERIKSC_REG): New.
777 (cirrus_register): New.
778 (cirrus_mvf_register): New.
779 (cirrus_mvd_register): New.
780 (cirrus_mvfx_register): New.
781 (cirrus_mvdx_register): New.
782 (cirrus_mvax_register): New.
783 (ARM_EXT_MAVERIKsc_register): New.
784 (reg_table): Add cirrus registers.
785 (cirrus_valid_reg): New.
786 (cirrus_reg_required_here): New.
787 (do_c_binops_1): New.
788 (do_c_binops_2): New.
789 (do_c_binops_3): New.
790 (do_c_triple_4): New.
791 (do_c_triple_5): New.
806 (cirrus_parse_offset): New.
808 (md_parse_option): Add arm9e.
809 (md_show_usage): Same.
813 * config/tc-ppc (ppc_bf, ppc_biei) : Set first .bi lineno value to
815 * config/obj-coff.c (coff_frob_symbol) : XCOFF does not use endndx.
819 * config/obj-elf.c: Standardize error/warning messages - don't
820 capitalise, no final period or newline, don't say "ignoring" for
821 as_bad messages. In some cases, change the wording to that used
822 elsewhere for similar messages.
823 (obj_elf_section_name): New function, split out from ..
824 (obj_elf_section): .. here. Correctly mask off SHF_MERGE if
825 entsize not specified.
829 * doc/as.texinfo: Document M and S ELF section flags.
833 * subsegs.c (subseg_text_p): Return 0 for absolute section.
834 * read.c (do_align): If in absolute section, warn about and ignore
835 non-zero fill pattern.
839 * config/tc-mn10300.c (tc_gen_reloc): Don't free
840 reloc->sym_ptr_ptr if it's not allocated.
844 * config/tc-mips.c (prev_insn_reloc_type): Make it an array to hold a
846 (prev_insn_fixp): Likewise.
847 (append_insn): Changed prototype to accept a relocation pointer.
848 (imm_reloc): Make it an array.
849 (offset_reloc): Likewise.
850 (md_assemble): Handle triple relocations.
851 (append_insn): Likewise. Add handling for some NewABI relocations.
852 (mips_no_prev_insn): Handle triple relocations.
853 (macro_build): Likewise. Add handling for some NewABI relocations.
854 Move handling for the 'u' case to append_insn().
855 (mips16_macro_build): Handle triple relocations.
856 (macro_build_lui): Likewise. Don't handle _gp_disp as special symbol
858 (mips_ip): Handle triple relocations.
859 (mips16_ip): Likewise.
860 (mips_force_relocation): Force handling of triple relocations
861 without symbols for NewABI.
862 (md_apply_fix): Add handling for some NewABI relocations.
866 * config/tc-i386.c (parse_register): If not producing code for
867 x86_64, reject x86_64 register name matches.
868 (md_assemble): Remove now redundant check for x86_64 regs.
872 * config/tc-ppc.c (md_assemble <DS relocs>): Test ppc_size as well
873 as BFD_DEFAULT_TARGET_SIZE.
875 (ppc_is_toc_sym): Likewise.
876 (md_apply_fix3): Likewise.
880 * config/tc-mips.c (mips_64): Remove.
881 (mips_target_format): Move downwards in file, use HAVE_64BIT_OBJECTS
883 (mips_abi_level, mips_abi): New enum.
884 (mips_32bit_abi): Remove.
885 (HAVE*PRS): Use mips_abi instead of mips_32bit_abi.
886 (HAVE_NEWABI): New define.
887 (HAVE_64BIT_OBJECTS): New define.
888 (HAVE_32BIT_ADDRESSES): Don't return true for 64bit objects.
889 (HAVE_64BIT_ADDRESSES): New define, inverse of HAVE_32BIT_ADDRESSES.
890 (support_64bit_objects): New prototype.
891 (md_begin): Use mips_abi instead of mips_32bit_abi. Don't write
892 .reginfo section for n32, use .MIPS.options instead.
893 (support_64bit_objects): New function, code from md_parse_option.
894 (md_longopts): Add -n32 option.
895 (md_parse_option): Use mips_abi instead of mips_32bit_abi/mips64.
896 Add -n32 option. Protect with OBJ_ELF.
897 (s_mipsset): Use mips_abi instead of mips_32bit_abi.
898 (mips_elf_final_processing): Likewise. Don't write .reginfo section
899 for n32, use .MIPS.options instead.
903 * po/POTFILES.in: Regenerate.
904 * configure: Regenerate.
908 * config/tc-ppc.c (md_apply_fix3 <BFD_RELOC_CTOR>): Make it 64
909 bits if target is 64 bit.
911 * doc/as.texinfo (listing): Fix typo.
913 * as.c (print_version_id): Use BFD_VERSION_STRING in place of
915 * Makefile.am (Makefile): Depend on bfd/configure.in.
917 * Makefile.in: Regenerate.
921 * config/obj-elf.c (obj_elf_parse_section_letters): Accept "am"
922 and "ams" for compatibility.
926 * config/tc-sh.c (md_pcrel_from_section): Transformed from
927 md_pcrel_from. Handle pc-relativeness against link-time
928 symbol. Handle relativeness to elsewhere than the fixup.
929 * config/tc-sh.h (MD_PCREL_FROM_SECTION): Define.
930 (md_pcrel_from_section): Prototype.
932 * Makefile.am: Update dependencies with "make dep-am".
933 * Makefile.in: Regenerate.
937 * config/tc-m68hc11.c (cmp_opcode): Define prototype.
938 (print_opcode_format, skip_whites): Likewise.
939 (convert_branch, m68hc11_new_insn): Likewise.
940 (build_dbranch_insn, build_indexed_byte): Likewise.
941 (build_reg_mode, find, find_opcode): Likewise.
942 (print_insn_format): Fix call to print_opcode_format.
943 (md_assemble): Fix call to build_dbranch_insn.
947 * config/tc-arm.c: Add missing prototypes.
951 * input-file.c (input_file_open): When reading the
952 first line looking for #NO_APP, prepare for the possibility
953 of finding #APP instead. Also fix algorithm to allow
954 white-space to follow either #NO_APP or #APP directives.
958 * listing.c (buffer_line): Don't write past the end of `line' when
963 * config/tc-mn10300.c (tc_gen_reloc): Don't emit an *ABS*
964 relocation for differences between symbols in a section other
965 than the one in which the difference is to be placed; apply
966 the relocation instead.
970 * config/tc-arc.c: Add missing prototype.
971 (md_atof): Change type to int. Add missing prototype.
972 * config/tc-arc.h: Prototype exported functions.
976 * as.c: Add missing function prototype.
977 * config/obj-aout.c: Fix compile time warning.
981 * config/tc-openrisc.c: Don't include <ctype.h>.
985 * gas/config/tc-s390.c: Add option -mwarn-areg-zero.
989 * config/tc-i370.c: Fix typo in last change.
993 * doc/as.texinfo (Symbol Names): Improve documentation on local
994 labels and add documenation about dollar labels.
998 * as.h: Don't include <ctype.h>.
999 * as.c (main): For gettext, also set the LC_CTYPE locate facet.
1000 * atof-generic.c: Include "safe-ctype.h" instead of <ctype.h>.
1001 (atof_generic): Use ISDIGIT instead of isdigit.
1002 * config/m68k-parse.y: Include "safe-ctype.h".
1003 (yylex): Use ISDIGIT/ISALNUM instead of isdigit/isalnum.
1004 * config/obj-elf.c: Include "safe-ctype.h".
1005 (obj_elf_vtable_inherit): Use ISSPACE instead of isspace.
1006 * config/obj-vms.c: Include "safe-ctype.h".
1007 (Write_VMS_MHD_Records): Use TOUPPER instead of islower/toupper.
1008 (VMS_Case_Hack_Symbol): Use ISUPPER/ISLOWER/TOUPPER/TOLOWER
1009 instead of isupper/islower/toupper/tolower.
1010 * config/tc-a29k.c: Include "safe-ctype.h" instead of <ctype.h>.
1011 (insert_sreg): Use TOUPPER instead of islower/toupper.
1012 (machine_ip): Use ISALPHA/ISALNUM/ISUPPER/TOLOWER instead of
1013 isalpha/isalnum/isupper/tolower.
1014 (a29k_unrecognized_line, md_operand): Use ISDIGIT instead of
1016 * config/tc-alpha.c: Include "safe-ctype.h" instead of <ctype.h>.
1017 (md_undefined_symbol, s_alpha_ent): Use ISDIGIT instead of isdigit.
1018 * config/tc-arc.c: Include "safe-ctype.h" instead of <ctype.h>.
1019 (md_assemble): Use ISSPACE/ISALNUM instead of isspace/isalnum.
1020 (arc_extoper): Use TOLOWER instead of isupper/tolower.
1021 * config/tc-arm.c: Include "safe-ctype.h" instead of <ctype.h>.
1022 (arm_psr_parse): Use ISALPHA instead of isalpha.
1023 (accum0_required_here): Use ISALNUM instead of isalnum.
1024 (do_ldrd): Use ISSPACE instead of isspace.
1025 (decode_shift): Use ISALPHA instead of isalpha.
1026 (insert_reg): Use TOUPPER instead of islower/toupper.
1027 (arm_reg_parse): Use ISALPHA instead of isalpha.
1028 (arm_parse_reloc): Use ISALNUM/ISPUNCT/TOLOWER instead of
1029 isalnum/ispunct/tolower.
1030 * config/tc-avr.c: Include "safe-ctype.h" instead of <ctype.h>.
1031 (md_parse_option): Use TOLOWER instead of tolower.
1032 (avr_operand): Use ISDIGIT/TOLOWER instead of isdigit/tolower.
1033 * config/tc-cris.c: Include "safe-ctype.h" instead of <ctype.h>.
1034 (cris_process_instruction): Use ISLOWER instead of islower.
1035 (get_gen_reg): Use ISALNUM/ISDIGIT instead of isalnum/isdigit.
1036 (get_spec_reg): Use TOLOWER/ISALNUM instead of isupper/tolower/isalnum.
1037 (get_flags): Use ISSPACE instead of isspace.
1038 * config/tc-d10v.c: Include "safe-ctype.h" instead of <ctype.h>.
1039 (do_assemble): Use TOLOWER instead of tolower.
1040 * config/tc-d30v.c: Include "safe-ctype.h" instead of <ctype.h>.
1041 (do_assemble): Use TOLOWER instead of tolower.
1042 (d30v_start_line): Use ISSPACE instead of isspace.
1043 * config/tc-fr30.c: Include "safe-ctype.h" instead of <ctype.h>.
1044 (fr30_is_colon_insn): Use TOLOWER instead of tolower.
1045 * config/tc-h8300.c: Include "safe-ctype.h" instead of <ctype.h>.
1046 (skip_colonthing, get_operand): Use ISDIGIT instead of isdigit.
1047 * config/tc-h8500.c: Include "safe-ctype.h" instead of <ctype.h>.
1048 (start_label): Use ISALPHA instead of isalpha.
1049 * config/tc-hppa.c: Include "safe-ctype.h" instead of <ctype.h>.
1050 (pa_ip): Use ISUPPER/ISLOWER/TOLOWER instead of
1051 isupper/islower/tolower.
1052 (pa_parse_number): Use ISDIGIT instead of isdigit.
1053 (pa_chk_field_selector): Use TOLOWER instead of tolower.
1054 (pa_stringer): Use ISDIGIT instead of isdigit.
1055 * config/tc-i370.c: Include "safe-ctype.h" instead of <ctype.h>.
1056 (register_name): Use ISALPHA instead of isalpha.
1057 (i370_elf_suffix): Use ISALNUM/TOLOWER instead of
1058 isalnum/islower/tolower.
1059 (i370_addr_offset): Use ISDIGIT/ISALPHA instead of
1061 (i370_addr_cons): Use ISALPHA/ISXDIGIT instead of
1063 (md_assemble): Use ISSPACE instead of isspace.
1064 * config/tc-i386.c: Include "safe-ctype.h" instead of <ctype.h>.
1065 (md_begin): Use ISDIGIT/ISLOWER/ISUPPER/TOLOWER/ISALPHA
1066 instead of isdigit/islower/isupper/tolower/isalpha.
1067 (output_invalid): Use ISPRINT instead of isprint.
1068 * config/tc-i860.c: Include "safe-ctype.h" instead of <ctype.h>.
1069 (i860_ip): Use ISLOWER/ISDIGIT instead of islower/isdigit.
1070 * config/tc-i960.c: Include "safe-ctype.h" instead of <ctype.h>.
1071 (get_args): Use ISALNUM instead of isalnum.
1072 * config/tc-ia64.c: Include "safe-ctype.h".
1073 (dot_pred_rel): Use TOUPPER/ISDIGIT instead of toupper/isdigit.
1074 (ia64_unrecognized_line): Use ISDIGIT instead of isdigit.
1075 (ia64_parse_name): Likewise.
1076 * config/tc-m32r.c: Include "safe-ctype.h" instead of <ctype.h>.
1077 (assemble_two_insns): Use ISSPACE/ISALNUM/ISUPPER/TOLOWER
1078 instead of isspace/isalnum/isupper/tolower.
1079 * config/tc-m68hc11.c: Include "safe-ctype.h" instead of <ctype.h>.
1080 (md_assemble): Use TOLOWER/ISALNUM instead of tolower/isalnum.
1081 * config/tc-m68k.c: Include "safe-ctype.h" instead of <ctype.h>.
1082 (mklower_table): Remove variable.
1083 (mklower): Remove macro.
1084 (insert_reg): Use TOUPPER instead of islower/toupper.
1085 (md_begin): Remove initialization of mklower_table.
1086 (s_reg): Use ISALNUM instead of isalnum.
1087 (mri_assemble): Use ISUPPER/TOLOWER instead of isupper/tolower.
1088 (parse_mri_condition): Use TOLOWER instead of isupper/tolower.
1089 (build_mri_control_operand): Use TOLOWER instead of tolower.
1090 (s_mri_else, s_mri_break, s_mri_next, s_mri_for): Likewise.
1091 * config/tc-m88k.c: Include "safe-ctype.h" instead of <ctype.h>.
1092 (md_assemble): Use ISSPACE instead of isspace.
1093 (get_imm16): Use ISALNUM instead of isalnum.
1094 (get_cnd): Use ISDIGIT/ISUPPER/TOLOWER instead of
1095 isdigit/isupper/tolower.
1096 (get_bf_offset_expression): Use ISALPHA/ISUPPER/TOLOWER
1097 instead of isalpha/isupper/tolower.
1098 (hexval): Use ISDIGIT/ISLOWER/ISUPPER instead of
1099 isdigit/islower/isupper.
1100 * config/tc-mcore.c: Include "safe-ctype.h" instead of <ctype.h>.
1101 (mcore_s_section): Use ISSPACE instead of isspace.
1102 (parse_reg): Use ISSPACE/TOLOWER/ISALNUM instead of
1103 isspace/tolower/isalnum.
1104 (parse_creg): Use ISSPACE/TOLOWER instead of isspace/tolower.
1105 (parse_psrmod): Use TOLOWER instead of isascii/tolower.
1106 (parse_exp, parse_mem, md_assemble): Use ISSPACE instead of
1108 * config/tc-mips.c: Include "safe-ctype.h" instead of <ctype.h>.
1109 (mips_ip): Use ISSPACE/ISDIGIT instead of isspace/isdigit.
1110 (mips16_ip): Use ISLOWER/ISDIGIT instead of islower/isdigit.
1111 (my_getSmallExpression): Use ISDIGIT/TOLOWER instead of
1113 (tc_get_register): Likewise.
1114 (get_number): Use ISDIGIT/ISXDIGIT instead of isdigit/isxdigit.
1115 (s_mips_ent): Use ISDIGIT instead of isdigit.
1116 * config/tc-mn10200.c: Include "safe-ctype.h" instead of <ctype.h>.
1117 (md_assemble): Use ISSPACE instead of isspace.
1118 * config/tc-mn10300.c: Include "safe-ctype.h" instead of <ctype.h>.
1119 (md_assemble): Use ISSPACE instead of isspace.
1120 * config/tc-ns32k.c: Don't include <ctype.h>.
1121 * config/tc-pdp11.c: Include "safe-ctype.h".
1122 (mklower): Remove function.
1123 (parse_reg): Use TOLOWER instead of mklower.
1124 * config/tc-pj.c: Include "safe-ctype.h".
1125 (md_assemble): Use ISSPACE instead of isspace.
1126 * config/tc-ppc.c: Include "safe-ctype.h" instead of <ctype.h>.
1127 (register_name): Use ISALPHA instead of isalpha.
1128 (ppc_elf_suffix): Use ISALNUM/TOLOWER instead of
1129 isalnum/islower/tolower.
1130 (md_assemble): Use ISSPACE instead of isspace.
1131 (ppc_canonicalize_symbol_name): Use ISLOWER/TOUPPER instead of
1133 * config/tc-s390.c: Include "safe-ctype.h" instead of <ctype.h>.
1134 (register_name): Use ISALPHA instead of isalpha.
1135 (s390_elf_suffix, s390_lit_suffix): Use ISALNUM instead of isalnum.
1136 (md_gather_operands, md_assemble, s390_insn): Use ISSPACE instead of
1138 * config/tc-sh.c: Include "safe-ctype.h" instead of <ctype.h>.
1139 (sh_elf_suffix): Use ISALNUM/TOLOWER instead of
1140 isalnum/islower/tolower.
1141 (IDENT_CHAR): Use ISALNUM instead of isalnum.
1142 (parse_reg): Use TOLOWER instead of tolower.
1143 (find_cooked_opcode): Use TOLOWER instead of isupper/tolower.
1144 * config/tc-sparc.c: Include "safe-ctype.h" instead of <ctype.h>.
1145 (sparc_ip): Use ISLOWER/ISDIGIT instead of islower/isdigit.
1146 (parse_keyword_arg): Use ISALNUM instead of isalnum.
1147 * config/tc-tahoe.c: Include "safe-ctype.h".
1148 (tahoe_reg_parse): Use ISDIGIT instead of isdigit.
1149 (tip_op): Use TOLOWER instead of isupper/tolower.
1150 * config/tc-tic30.c: Include "safe-ctype.h".
1151 (md_begin): Use ISLOWER/ISDIGIT/ISUPPER/TOLOWER/ISLOWER/
1152 ISALPHA instead of islower/isdigit/isupper/tolower/islower/isalpha.
1153 (tic30_operand): Use TOLOWER instead of tolower.
1154 (tic30_find_parallel_insn): Likewise.
1155 (output_invalid): Use ISPRINT instead of isprint.
1156 * config/tc-tic54x.c: Include "safe-ctype.h".
1157 (tic54x_asg, tic54x_eval): Use ISALPHA instead of isalpha.
1158 (lookup_version): Use TOUPPER instead of toupper.
1159 (tic54x_var): Use ISALPHA instead of isalpha.
1160 (tic54x_mlib): Use ISSPACE instead of isspace.
1161 (subsym_iscons): Use TOUPPER instead of toupper.
1162 (get_operands): Use ISSPACE instead of isspace.
1163 (is_type): Use TOUPPER/ISDIGIT instead of toupper/isdigit.
1164 (encode_indirect, encode_operand): Use TOUPPER instead of toupper.
1165 (next_line_shows_parallel): Use ISSPACE instead of isspace.
1166 (subsym_get_arg, subsym_substitute): Use ISDIGIT instead of isdigit.
1167 (tic54x_start_line_hook, md_assemble, tic54x_start_label): Use
1168 ISSPACE instead of isspace.
1169 * config/tc-tic80.c: Include "safe-ctype.h".
1170 (md_assemble): Use ISSPACE instead of isspace.
1171 * config/tc-v850.c: Include "safe-ctype.h" instead of <ctype.h>.
1172 (system_register_name): Use ISDIGIT instead of isdigit.
1173 (md_assemble): Use ISSPACE instead of isspace.
1174 * config/tc-vax.c: Include "safe-ctype.h" instead of <ctype.h>.
1175 (vax_reg_parse): Use TOLOWER/ISDIGIT instead of
1176 isupper/tolower/isdigit.
1177 (vip_op): Use TOLOWER instead of isupper/tolower.
1178 * config/tc-w65.c: Don't include <ctype.h>.
1179 * config/tc-z8k.c: Include "safe-ctype.h" instead of <ctype.h>.
1180 (tohex): Use ISDIGIT/ISLOWER instead of isdigit/islower.
1181 (whatreg): Use ISDIGIT instead of isdigit.
1182 * ecoff.c: Include "safe-ctype.h" instead of <ctype.h>.
1183 (ecoff_directive_ent, ecoff_stab): Use ISDIGIT instead of isdigit.
1184 * expr.c: Include "safe-ctype.h" instead of <ctype.h>.
1185 (integer_constant): Use ISALNUM/TOUPPER instead of
1186 isalnum/islower/toupper.
1187 (operand): Use TOLOWER instead of isupper/tolower.
1188 * gasp.c: Include "safe-ctype.h" instead of <ctype.h>.
1189 (sb_strtol, level_0, change_base, doinstr): Use ISDIGIT instead
1191 (process_assigns, whatcond): Use TOUPPER instead of toupper.
1192 (chartype_init): Use ISALPHA/ISDIGIT instead of isalpha/isdigit.
1193 (main): For gettext, also set the LC_CTYPE locate facet.
1194 * hash.c: Include "safe-ctype.h".
1195 (main): Use TOLOWER instead of isupper/tolower.
1196 * itbl-lex.l: Don't include <ctype.h>.
1197 * listing.c: Include "safe-ctype.h" instead of <ctype.h>.
1198 (listing_newline): Use ISCNTRL instead of isascii/iscntrl. Don't
1199 omit non-ASCII characters.
1200 (debugging_pseudo): Use ISSPACE instead of isspace.
1201 * macro.c: Include "safe-ctype.h" instead of <ctype.h>.
1202 (buffer_and_nest): Use ISALNUM instead of isalnum.
1203 (get_token): Use ISALPHA/ISALNUM instead of isalpha/isalnum.
1204 (define_macro): Use ISUPPER/TOLOWER instead of isupper/tolower.
1205 (macro_expand_body): Use ISALNUM/ISDIGIT/ISUPPER/ISALPHA
1206 instead of isalnum/isdigit/isupper/isalpha.
1207 (check_macro): Use ISALPHA/ISALNUM/ISUPPER/TOLOWER instead
1208 of isalpha/isalnum/isupper/tolower.
1209 * read.c: Include "safe-ctype.h" instead of <ctype.h>.
1210 (read_a_source_file): Use ISUPPER/TOLOWER/ISDIGIT instead of
1211 isupper/tolower/isdigit.
1212 (s_mri_common): Use ISDIGIT instead of isdigit.
1213 (s_mri_sect): Use ISDIGIT/TOUPPER instead of isdigit/toupper.
1214 (s_float_space): Use ISALPHA instead of isalpha.
1215 (ignore_rest_of_line): Use ISPRINT instead of isprint.
1216 (float_cons): Use ISALPHA instead of isalpha.
1217 (next_char_of_string): Use ISDIGIT/ISXDIGIT instead of
1219 * symbols.c: Include "safe-ctype.h" instead of <ctype.h>.
1220 (save_symbol_name): Use ISLOWER/TOUPPER instead of
1222 (symbol_find_base): Use TOUPPER instead of islower/toupper.
1223 (decode_local_label_name): Use ISDIGIT instead of isdigit.
1227 * tc-s390.c (s390_insn): Add code to cope with 6 byte O_constants
1228 in 64 bit mode and make format "e" work.
1232 * dwarf2dbg.c (dwarf2_directive_file): Avoid signed/unsigned warning.
1234 * write.c (set_symtab): Update bfd_alloc declaration. Use a temp
1235 var to ensure bfd_alloc arg is the right type.
1236 (write_object_file): Cast args of bfd_seek. Replace bfd_write with
1239 * config/obj-coff.c: Replace calls to bfd_write with calls to
1240 bfd_bwrite. Cast args of bfd_seek.
1242 * config/obj-elf.c (obj_elf_change_section): Avoid signed/unsigned
1245 * config/tc-mn10300.c (set_arch_mach): Make param unsigned.
1247 * config/tc-tic54x.c (tic54x_mlib): Replace bfd_read call with
1252 * config/tc-sh.h (sh_force_relocation): Prototype.
1253 (struct fix): Forward declare.
1254 * config/tc-sh.c (type sh_operand_info): Move to top of file.
1255 (cons, s_align_bytes): Remove old-type declarations.
1256 (sh_elf_suffix, parse_reg, dot, parse_exp, parse_at, get_operand,
1257 get_operands, get_specific, insert, build_relax,
1258 insert_loop_bounds, build_Mytes): Prototype.
1259 (little): Make static. Prototype.
1260 (check, tc_Nout_fix_to_chars): Delete unused functions.
1264 * config/tc-mips.c (md_parse_option): Remove setting mips_64 via
1269 * configure.in: Add arm-openbsd target.
1270 * configure: Regenerate.
1274 * configure.in: Change machine triplets from mips-*-linux-gnu* to
1279 * config/tc-mips.c (append_insn): Don't rightshift BFD_RELOC_16_PCREL.
1283 * config/tc-alpha.c (alpha_elf_section_letter): New.
1284 (alpha_elf_section_flags): New.
1285 * config/tc-alpha.h (md_elf_section_letter): New.
1286 (md_elf_section_flags): New.
1287 * config/tc-ia64.c (ia64_elf_section_letter): New.
1288 * config/tc-ia64.h (md_elf_section_letter): New.
1292 * config/obj-elf.c (obj_elf_parse_section_letters): Use 'M' instead
1293 of 'm', 'S' instead of 's'. Update bad_msg.
1294 * config/tc-ppc.c (ppc_section_letter): Update bad_msg.
1295 * config/tc-i370.c (i370_sectioN_letter): Update bad_msg.
1299 * expr.c (expr): Move code setting "retval" to the end of the loop,
1300 and rearrange for efficiency. For "PIC code" subtraction, use
1301 "rightseg" rather than recalculating. For "symbol OP symbol"
1302 subtract, set "retval" to absolute_section if symbols in same
1304 * symbols.c (resolve_symbol_value): Resolve "sym +/- expr" to an
1305 O_symbol. Simplify a +/- b code. Allow equality and non-equality
1306 comparisons on symbols from any section. Allow other comparison
1307 operators as for subtraction.
1308 (symbol_equated_reloc_p): New predicate function.
1309 * symbols.h (symbol_equated_reloc_p): Declare.
1310 * write.c (relax_segment <rs_machine_dependent>): Ensure segment
1311 for expression syms is set correctly.
1312 (adjust_reloc_syms): Use symbol_equated_reloc_p.
1313 (write_relocs): Likewise.
1314 (write_object_file): Likewise.
1315 * config/tc-mips.c (md_estimate_size_before_relax): Likewise.
1316 * config/tc-i386.c (md_assemble <Output jumps>): Don't lose part
1317 of a complex expression when setting up frag_var.
1321 * config/tc-alpha.c (alpha_reloc_op_tag): Replace need_seq with
1322 require_seq and allow_seq. Let !literal omit the sequence number.
1323 (tokenize_arguments): Reject sequence numbers of !allow_seq.
1327 * config/tc-sparc.c (md_apply_fix3): Handle relocs against SEC_MERGE
1328 section symbols the same way as externs.
1332 * config/tc-mips.c (s_mipsset): Reallow unrestricted use of .set mipsX
1337 * config/tc-mips.c (append_insn): Handle BFD_RELOC_16_PCREL.
1338 (macro_build): Use BFD_RELOC_16_PCREL_S2 only for embedded
1339 PIC, BFD_RELOC_16_PCREL for the rest.
1340 (mips_ip): Likewise.
1341 (md_pcrel_from): return the right offset for the differently shifted
1343 (md_apply_fix): Handle BFD_RELOC_16_PCREL.
1347 * config/tc-ia64.c (FUNC_IPLT_RELOC): New.
1348 (pseudo_func): Add @iplt.
1349 (pseudo_opcode): Add data16 and data16.ua.
1350 (md_begin): Set iplt pseudo.
1351 (ia64_cons_fix_new): Handle 16 byte iplt reloc specially.
1355 * config/tc-alpha.c (struct alpha_insn): Make sequence scalar long.
1356 (MACRO_LITERAL, MACRO_BASE, MACRO_BYTOFF, MACRO_JSR): Remove.
1357 (alpha_macros): Remove occurrences of same.
1358 (O_lituse_addr, O_gprel): New.
1359 (DUMMY_RELOC_LITUSE_*): New.
1360 (s_alpha_ucons, s_alpha_arch): Prototype.
1361 (alpha_reloc_op): Construct elements via DEF macro.
1362 (ALPHA_RELOC_SEQUENCE_OK): Remove.
1363 (struct alpha_reloc_tag): Rename from alpha_literal_tag; rename
1364 members to not be literal specific.
1365 (next_sequence_num): New.
1366 (md_apply_fix3): Cope with missing GPDISP_LO16. Adjust for
1367 added/removed BFD relocations.
1368 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
1369 (alpha_adjust_symtab_relocs): Handle GPDISP relocs as well.
1370 (tokenize_arguments): Parse ! relocations properly.
1371 (find_macro_match): Delete unused macro argument types.
1372 (assemble_insn): Add reloc parameter; emit that instead of the
1373 default as appropriate.
1374 (get_alpha_reloc_tag): New. Split from ...
1375 (emit_insn): ... here. Allocate a reloc tag for GPDISP.
1376 (assemble_tokens): Don't search macros if user relocation present.
1377 Copy reloc sequence number to insn struct.
1378 (emit_ldgp): Remove user reloc handling.
1379 (load_expression, emit_lda, emit_ldah, emit_ir_load): Likewise.
1380 (emit_loadstore, emit_ldXu, emit_ldil, emit_stX): Likewise.
1381 (emit_sextX, emit_division, emit_jsrjmp, emit_retjcr): Likewise.
1382 * config/tc-alpha.h (tc_adjust_symtab): Always define.
1383 (struct alpha_fix_tag): Name members less literal specific.
1387 * tc-h8300.c (tc_gen_reloc): Give an error if we try to take the
1388 address of two symbols in different sections.
1393 * config/tc-mips.c (mips_cpu_info): Add support for mipsisa32,
1394 5kc, and 20kc. Clean up old entries.
1399 * config/tc-h8300.c (build_bytes): For OBJ_ELF, make relocation's
1400 offset match H8 ELF spec.
1401 (md_section_align): Alternate implementation for BFD_ASSEMBLER.
1402 (md_apply_fix): Fix argument and return types for BFD_ASSEMBLER.
1404 * tc-h8300.c (relocation mappings): Remove. Moved to tc-h8300.h.
1405 (build_bytes): Mark fixups for PCrel branches as signed. For
1406 OBJ_ELF, make sure the reloc's offset points to the first byte
1408 (md_convert_frag): Update definiton based on BFD_ASSEMBLER.
1409 * tc-h8300.h (relocation mappings): Add.
1413 * tc-h8300.c (tc_crawl_symbol_chain, tc_headers_hook): Don't
1414 define for BFD_ASSEMBLER.
1415 (tc_reloc_mangle): Likewise.
1416 (tc_gen_reloc): New function for BFD_ASSEMBLER.
1420 * config/tc-h8300.c (h8300hmode): Record the machine type
1421 if BFD_ASSEMBLER is defined.
1422 (h8300smode, md_begin): Similarly.
1426 * configure.in (i[3456]86-*-rtems*, m68*-*-rtems*): Change
1427 default from coff to elf.
1428 * configure: Regenerate.
1432 * tc-h8300.c: If OBJ_ELF, include elf/h8.h, and define
1433 assorted coff relocations to the corresponding elf relocations.
1434 * tc-h8300.h (TARGET_ARCH, TARGET_FORMAT): Define appropriately.
1436 * configure.in: Add case for h8300-*-elf.
1437 * configure: Regenerate.
1441 * doc/internals.texi (MD_PCREL_FROM_SECTION): Change
1442 documentation to reflect reality.
1446 * config/tc-mips.c (load_address): Reflect change to MAX_GPREL_OFFSET.
1447 (macro): Reflect change to MAX_GPREL_OFFSET.
1454 * configure.in: Recognise powerpc*le*, not just powerpcle*.
1455 * configure: Regenerate.
1456 * config/tc-ppc.c (PPC_LO, PPC_HI, PPC_HA, PPC_HIGHER,
1457 PPC_HIGHERA, PPC_HIGHEST, PPC_HIGHESTA, SEX16): New macros.
1458 (md_assemble): Use them.
1459 (ppc_machine): Support stub for ELF64 as well as XCOFF.
1460 (md_pseudo_table): Add "llong", "quad".
1461 (md_parse_option): Match default_cpu of powerpc*.
1462 (ppc_arch): Likewise.
1463 (ppc_subseg_align): Only for OBJ_XCOFF.
1464 (ppc_target_format): Return elf64-powerpc strings for 64 bit ELF.
1465 (md_begin): Select PPC_OPCODE_64 for 64 bit.
1466 (ppc_insert_operand): Don't bother testing 'file' before calling
1467 as_bad_where. Use as_bad_where for operand->insert errors.
1468 (mapping): Add ELF64 relocation modifiers.
1469 (ppc_elf_suffix): Replace symbol on BFD_RELOC_PPC64_TOC reloc
1470 expressions with abs_symbol.
1471 (ppc_elf_cons): Correct offset for little endian targets.
1472 (ppc_elf_frob_symbol): New.
1473 (md_assemble): Add support for 64 bit ELF relocs.
1474 (ppc_tc): Ensure 8 byte alignment when 64 bit.
1475 (ppc_is_toc_sym): Only define for OBJ_XCOFF and OBJ_ELF. Match
1476 ".toc" section for 64 bit ELF.
1477 (ppc_fix_adjustable): New. Macro body moved from tc-ppc.h.
1478 (md_apply_fix3): Silence warning with ATTRIBUTE_UNUSED. Only do
1479 the ppc_is_toc_sym check for OBJ_XCOFF and OBJ_ELF. For 64 bit,
1480 use BFD_RELOC_PPC64_TOC16_DS instead of BFD_RELOC_PPC_TOC16.
1481 Expand on comments, error message. Add support for 64 bit relocs,
1482 and use PPC_HI etc. macros.
1483 * config/tc-ppc.h (MAX_MEM_FOR_RS_ALIGN_CODE): Define.
1484 (HANDLE_ALIGN): Define to generate nops in code sections rather
1486 (TC_FORCE_RELOCATION): Force for BFD_RELOC_PPC64_TOC.
1487 (ELF_TC_SPECIAL_SECTIONS): Add 64 bit ELF sections.
1488 (tc_fix_adjustable): Move body of macro to tc-ppc.c.
1489 (ppc_fix_adjustable): Declare.
1490 (tc_frob_symbol): Define.
1491 (ppc_elf_frob_symbol): Declare.
1495 * config/tc-s390.c: Add missing prototypes
1496 (md_gather_operands): Make static.
1500 * config/tc-mips.h (MAX_GPREL_OFFSET): Change it to the maximum
1501 allowed value, not the word beyond maximum.
1502 * config/tc-mips.c (macro_build_lui): Code cleanup.
1503 (macro): Reflect change to MAX_GPREL_OFFSET.
1504 (mips_ip): Check explicitly against S_EX_NONE.
1505 (my_get_SmallExpression): parse for %gp_rel, not %gprel.
1506 (md_apply_fix): Code cleanup.
1510 * config/tc-ppc.h: Fix prototype declaration.
1514 * config/tc-ia64.c (dot_endp): Set function symbol sizes.
1515 * config/obj-elf.c (obj_elf_size): When size is constant, free
1516 and clear symbol_get_obj(sym)->size if any.
1520 * config/tc-mips.c (md_begin): Warn about incompatibility between
1521 -march=FOO and -mipsN option, continue with default ISA.
1525 * config/tc-mips.c (md_parse_option): #ifdef the
1526 traditional/non-traditional names as it is done in mips_target_format.
1530 * config/tc-alpha.h (MD_APPLY_FIX3): Define.
1531 * config/tc-alpha.c (md_apply_fix3): Rename from md_apply_fix;
1532 use seg parameter instead of now_seg.
1536 * config/tc-mips.c (show): Add the missing prototype.
1540 * config/tc-mips.c (S_EX_*): New enum for my_getSmallExpression()
1542 (mips_ip): Use the new return values instead of characters. Add
1543 support for %higher and %highest.
1546 (my_getSmallExpression): Make parsing case insensitive and more
1547 reliable. Add support for %higher and %highest. Further support to
1548 parse %gprel and %neg is implemented but currently deactivated.
1552 * ecoff.c (ecoff_frob_symbol): Remove casts from bfd_get_gp_size.
1553 (ecoff_build_symbols): Likewise.
1554 * read.c (s_lcomm_internal): Fix signed/unsigned warning.
1558 * config/tc-mips.c (mips_elf_final_processing): Set elf header flags
1559 for n32 ABI if appropriate.
1563 * config/tc-mips.c (cons_fix_new_mips): Remove.
1564 * config/tc-mips.h (TC_CONS_FIX_NEW): Remove.
1565 (cons_fix_new_mips): Remove.
1569 * expr.c (operand)[LITERAL_PREFIXDOLLAR_HEX]: Treat $L as
1570 a label, not a hex constant.
1574 * config/tc-v850.h: Add missing prototypes.
1578 * config/tc-s390.c (s390_elf_cons): Correct fixups for PLT
1583 * config/tc-sparc.c (cons_fix_new_sparc): Move
1584 `sparc_no_align_cons = 0;' to...
1585 (s_uacons): ...here so that sparc_no_align_cons will get
1590 * config/tc-mips.c: Fix a comment typo.
1594 * config/tc-mips.c (move_register): New function.
1595 (macro_build): Remove OPCODE_IS_MEMBER's gp32 argument.
1596 (mips_ip): Likewise.
1597 (macro2): Use move_register rather than macro_build for moves.
1598 (mips16_macro): Likewise.
1599 (macro): Likewise. Handle M_MOVE.
1603 * configure.in: Add -Wstrict-prototypes and -Wmissing-prototypes
1605 * configure: Regenerate.
1609 * as.h (alloca): Don't declare if __GNUC__. Remove an old comment.
1610 Comment indentation of #pragma.
1611 * macro.c (alloca): Likewise.
1613 * config/tc-ppc.c: Revert 2001-08-08.
1617 * symbols.c: Add missing prototypes.
1618 * config/e-i386elf.c: Likewise.
1619 * config/e-i386coff.c: Likewise.
1620 * config/e-i386aout.c: Likewise.
1621 * config/obj-coff.c: Likewise.
1622 (def_symbol_in_progress, stack): Move definition.
1623 * config/obj-elf.c: Add missing prototypes.
1624 (obj_elf_change_section): Make static.
1625 (obj_elf_parse_section_letters): Likewise.
1626 (obj_elf_section_word): Likewise.
1627 (obj_elf_section_type): Likewise.
1632 * config/tc-ppc.c (md_apply_fix3): Replace haphazard code for
1633 determining reloc type with code reading operands->reloc field.
1637 * config/obj-elf.c (elf_copy_symbol_attributes): Make it a global
1638 function, and expand old obj-elf.h OBJ_COPY_SYMBOL_ATTRIBUTES here.
1639 * config/obj-elf.h (elf_copy_symbol_attributes): Declare.
1640 (OBJ_COPY_SYMBOL_ATTRIBUTES): Call elf_copy_symbol_attributes.
1642 * output-file.c (output_file_create): Don't try to open a second
1643 time as it's unnecessary. FOPEN_W was undefined too.
1647 * config/tc-mips.c (macro_build): Replace magic constants by defines.
1648 (mips_ip): Likewise. Typo.
1652 * config/tc-mips.c (mips_abi_string): Initialize with NULL instead
1654 (RELAX_ENCODE): Replace bfd_vma by valueT.
1655 (load_address): Formatting.
1656 (macro): Likewise. Code cleanup. Typo.
1657 (macro2): Formatting.
1658 (md_parse_option): Typo.
1659 (s_mipsset): Reorder cases to look nicer.
1660 (md_estimate_size_before_relax): Formatting. Code cleanup.
1661 (mips_elf_final_processing): Check against NULL instead of 0.
1662 (md_obj_end): Remove ugly `' in message string.
1663 (s_mips_ent): Remove ugly `' in message string.
1667 * config/tc-mips.c (md_begin): Take -mcpu value into account even when
1668 -mipsX is specified. Make both -mcpu/-march and -mcpu/-mtune pairs
1669 mutually exclusive (if they are different).
1670 (md_parse_option): Warn if an -march/-mtune/-mcpu/-m<cpu> option is
1675 * config/tc-mips.c (md_apply_fix): Don't subtract the symbol value
1680 * doc/c-mips.tex (-mgp32, -mfp32): Added -mfp32, unified with -mgp32.
1681 * config/tc-mips.c (mips_fp32, mips_32bit_abi): New static variables.
1682 (md_long_opts): Add -mfp32 option.
1683 (md_parse_option): Handle it. Set mips_32bit_abi given -mabi=32.
1684 (md_show_usage): Show usage for -mfp32 and -mgp32.
1685 (HAVE_32BIT_GPRS, HAVE_32BIT_FPRS): New macros.
1686 (HAVE_64BIT_GPRS, HAVE_64BIT_FPRS): New macros, inverse of the above.
1687 (HAVE_32BIT_ADDRESSES): New macro.
1688 (load_register): Use HAVE_32BIT_GPRS to determine the register width.
1689 (load_address): Use HAVE_32BIT_ADDRESSES to determine the address size.
1690 (s_cprestore, s_cpadd): Likewise.
1691 (macro): Use HAVE_32BIT_GPRS to determine the width of registers
1692 used in branch and M_LI_D macros. Use HAVE_64BIT_FPRS to determine
1693 the width registers used in M_LI_DD macros. Use HAVE_32BIT_ADDRESSES
1694 to determine the width of addresses in load, store and jump macros.
1695 (macro2): Use HAVE_32BIT_GPRS to determine the width of registers
1696 used in set instructions; do not check the address size for them.
1697 Use HAVE_32BIT_ADDRESSES to determine the width of addresses in
1698 unaligned load and store macros.
1699 (mips_ip): Use the new macros to check the width of a register when
1700 processing float constants. Force a constant into memory if it is
1701 destined for an FPR and the FPRs are wider than the GPRs. Warn about
1702 odd FPR numbers if HAVE_32BIT_FPRS. Use HAVE_32BIT_GPRS rather
1703 than mips_gp32 to select synthetic instructions.
1704 (macro_build): Use HAVE_32BIT_GPRS rather than mips_gp32 to select
1705 synthetic instructions.
1709 * config/tc-z8k.c (parse_reg): If a string starts with "r", "rl",
1710 "rh", "rr", "rq", or "sp" then don't automatically assume
1711 it's a register name. Check whether there is a number
1716 * config/tc-i386.c (lex_got): Match lower case relocation tokens.
1717 Don't allocate more space than necessary for the input line copy.
1721 * read.c: Standardize error/warning messages - don't capitalise, no
1722 final period or newline, don't say "ignored" or "zero assumed" for
1723 as_bad messages. In some cases, change the wording to that used
1724 elsewhere for similar messages.
1727 * atof-generic.c: Likewise.
1730 * depend.c: Likewise.
1731 * dwarf2dbg.c: Likewise.
1732 * ecoff.c: Likewise.
1734 * frags.c: Likewise.
1735 * input-file.c: Likewise.
1736 * input-scrub.c: Likewise.
1737 * listing.c: Likewise.
1738 * output-file.c: Likewise.
1739 * stabs.c: Likewise.
1740 * subsegs.c: Likewise.
1741 * symbols.c: Likewise.
1742 * write.c: Likewise.
1743 * ecoff.c (ecoff_directive_end): Test for missing name by
1744 comparing input line pointers rather than reading string.
1745 (ecoff_directive_ent): Likewise.
1746 * read.c (s_set): Likewise.
1747 (s_align): Report a warning rather than an error for
1748 alignment too large.
1749 (s_comm): Check for missing symbol name.
1750 (s_lcomm_internal): Likewise.
1752 (s_globl): Use is_end_of_line instead of looking for '\n'.
1753 (s_lcomm_internal): Likewise.
1754 (ignore_rest_of_line): Report a warning rather than an error.
1758 * config/tc-ppc.c (ppc_elf_validate_fix): It's OK to have
1759 any kind of relocation against a not-loaded section.
1763 * config/obj-elf.c (obj_elf_symver): Temporarily modify lex_type
1764 to include '@' in symbol names when parsing versioned symbols
1765 rather than calling get_symbol_end multiple times.
1766 * config/tc-i370.c (register_name): Format fixes. Don't call
1767 get_symbol_end after parsing number.
1768 * config/tc-mn10200.c (data_register_name): Format fixes. Hoist
1769 code out of conditional.
1770 (address_register_name): Likewise.
1771 (other_register_name): Likewise.
1772 * config/tc-mn10300.c (r_register_name): Likewise.
1773 (xr_register_name): Likewise.
1774 (data_register_name): Likewise.
1775 (address_register_name): Likewise.
1776 (other_register_name): Likewise.
1777 * config/tc-ppc.c (register_name): Likewise.
1778 * config/tc-s390.c (register_name): Likewise.
1782 * config/tc-m68hc11.c (md_estimate_size_before_relax): Fix bsr
1783 and bra relax: update fragP->fr_fix after the fixup.
1787 * read.c (s_set): Check for missing symbol name.
1791 * write.c (relax_segment <rs_space>): Account for fr_fix.
1795 * config/tc-mips.c (md_estimate_size_before_relax): Make sure
1796 we treat weak like extern only for ELF.
1797 (mips_fix_adjustable): Make sure we don't adjust extern/weak
1798 symbols only for ELF.
1802 * config/tc-mips.c: Fix spelling error in comment.
1806 * app.c (LEX_IS_PARALLEL_SEPARATOR): New macro.
1807 (IS_PARALLEL_SEPARATOR): New macro.
1808 (do_scrub_begin): Set up characters in tc_parallel_separator_chars
1809 as LEX_IS_PARALLEL_SEPARATOR, if it is defined.
1810 (do_scrub_chars): Handle LEX_PARALLEL_SEPARATOR chars like
1811 LEX_LINE_SEPARATOR except that we go to state 1 (as if the label has
1816 * Makefile.am (DEP_FLAGS): Define, add -DOBJ_MAYBE_ELF.
1817 (DEP1, DEPTC, DEPOBJ, DEP2): Use it.
1818 Update dependencies with "make dep-am".
1819 * Makefile.in: Regenerate.
1820 * doc/Makefile.in: Regenerate.
1824 * config/tc-alpha.h: Fix formatting.
1825 * config/tc-arc.c: Likewise.
1826 * config/tc-d10v.h: Likewise.
1827 * config/tc-hppa.c: Likewise.
1828 * config/tc-i370.c: Likewise.
1829 * config/tc-i386.h: Likewise.
1830 * config/tc-i960.h: Likewise.
1831 * config/tc-ia64.c: Likewise.
1832 * config/tc-ia64.h: Likewise.
1833 * config/tc-m32r.c: Likewise.
1834 * config/tc-m68k.c: Likewise.
1835 * config/tc-m88k.c: Likewise.
1836 * config/tc-ns32k.c: Likewise.
1837 * config/tc-pdp11.c: Likewise.
1838 * config/tc-pj.h: Likewise.
1839 * config/tc-s390.c: Likewise.
1840 * config/tc-sparc.c: Likewise.
1841 * config/tc-sparc.h: Likewise.
1842 * config/tc-tic80.c: Likewise.
1843 * config/tc-w65.h: Likewise.
1847 * symbols.c (S_GET_VALUE): Don't treat O_constant and local
1848 symbols specially. Always resolve, adding fr_address to value.
1849 * write.c (write_object_file): Don't add fr_address to sym values.
1850 (relax_frag): Likewise.
1851 (relax_segment): Likewise.
1852 * config/obj-ieee.c (do_symbols): Likewise.
1853 * config/tc-cris.c (md_convert_frag): Likewise.
1854 * config/tc-fr30.c (md_convert_frag): Likewise.
1855 * config/tc-i386.c (md_convert_frag): Likewise.
1856 * config/tc-m32r.c (md_convert_frag): Likewise.
1857 * config/tc-m68hc11.c (md_convert_frag): Likewise.
1858 * config/tc-mcore.c (md_convert_frag): Likewise.
1859 * config/tc-mips.c (mips16_extended_frag): Likewise.
1860 * config/tc-ns32k.c (md_convert_frag): Likewise.
1861 * config/tc-m68k.c (md_convert_frag_1): Likewise.
1862 (BRANCHBWL, BRABSJUNC, BRABSJCOND, BRANCHBW, FBRANCH, DBCCLBR,
1863 DBCCABSJ, PCREL1632, PCINDEX, ABSTOPCREL): Decrement.
1864 (md_relax_table): Remove first four entries. Format.
1865 (md_estimate_size_before_relax): Remove old_fix. Don't bother
1866 setting fr_var. Simplify byte branch checks.
1870 * config/tc-hppa.c (pa_parse_nonneg_cmpsub_cmpltr): Remove
1871 "isbranch" param as all calls have it set.
1872 (pa_parse_neg_cmpsub_cmpltr): Likewise.
1873 (pa_parse_nonneg_add_cmpltr): Likewise. Remember result of
1874 strcasecmp in "nullify" var.
1875 (pa_parse_neg_add_cmpltr): Likewise.
1876 (pa_ip): Don't "save_s" unnecessarily. Update calls to above
1877 functions. Don't print wrong conditions in error messages.
1881 * config/tc-s390.c (s390_force_relocation): Removed.
1883 * config/tc-s390.h: Remove double declaration of
1884 TC_FORCE_RELOCATION.
1888 * config/tc-mips.c (s_mips_end): Remove unused variables.
1889 (s_mips_frame): Add ATTRIBUTE_UNUSED.
1893 * config/tc-m68k.c (md_pseudo_table) [OBJ_ELF]: Add .file and
1895 (md_assemble) [OBJ_ELF]: Call dwarf2_emit_insn before emitting
1897 * config/tc-m68k.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
1901 * config/tc-i386.c (comment_chars): Don't use '/' as comment start if
1903 (line_comment_chars): Set to '/' if TE_NetBSD.
1907 * configure.in (i386-*-netbsdelf*): New target.
1908 * configure: Regenerate.
1912 * config/obj-coff.c (coff_frob_symbol): Don't merge if the storage
1913 class of the non-debug symbol is C_NULL.
1917 * doc/as.texinfo (Incbin): Fix grammatical errors.
1921 * depend.c: Use FOPEN_.. macros in calls to fopen().
1922 * input-file.c: As above.
1923 * itbl-ops.c: As above.
1924 * listing.c: As above.
1925 * output-file.c: As above.
1930 * read.c (s_incbin): New .incbin function.
1931 * read.c (potable): Add "incbin" pseudo-op.
1932 * read.h: Add s_incbin prototype.
1933 * doc/as.texinfo (incbin): Document .incbin pseudo-op.
1934 * NEWS: Mention new feature.
1938 * ecoff.c (add_file): Only set debug_type to DEBUG_NONE if it is
1943 * cgen.c (gas_cgen_save_fixups): Modified to allow more than one
1944 set of fixups to be stored.
1945 (gas_cgen_restore_fixups): Modified to allow the fixup chain to be
1946 restored to be chosen from any that are saved.
1947 (gas_cgen_swap_fixups): Modified to allow the current set of
1948 fixups to be swapped with any other set that has been saved.
1949 (gas_cgen_initialize_saved_fixups_array): New routine.
1950 * cgen.h: Modifed prototypes for gas_cgen_save_fixups,
1951 gas_cgen_restore_fixups, and gas_cgen_swap_fixups. Added definitions
1952 or MAX_SAVED_FIXUP_CHAINS.
1953 * config/tc-m32r.c (assemble_two_insns): Changed calls to fixup
1954 store, swap and restore fuctions to reflect the new interface.
1958 * config/tc-m68k.c (md_estimate_size_before_relax): Catch and
1959 ignore empty, ineffectual alignment frags when deciding if a
1960 branch can be short.
1964 * config/tc-ia64.c (special_section): Add SPECIAL_SECTION_INIT_ARRAY
1965 and SPECIAL_SECTION_FINI_ARRAY.
1966 (special_section_name): Add .init_array and .fini_array.
1967 (md_pseudo_table): Add init_array and fini_array.
1968 (md): Add pointer_size and pointer_size_shift fields.
1969 (setup_unwind_header): New static function.
1970 (output_unw_records): Modify to use setup_unwind_header.
1971 (generate_unwind_image, dot_endp): Modify to use md.pointer_size and
1972 md.pointer_size_shift.
1973 (md_begin): Initialize md.pointer_size and md.pointer_size_shift.
1977 * README (Supported platforms): m88k no longer suffers bitrot.
1979 * config/tc-m88k.c (md_pseudo_table): Remove redundant "global".
1983 * struc-symbol.h (struct local_symbol): Rename lsy_offset to
1984 lsy_value. Correct typos in comments.
1985 * symbols.c (local_symbol_make): Update for name change.
1986 (local_symbol_convert): Likewise.
1988 (S_GET_VALUE): Likewise.
1989 (S_SET_VALUE): Likewise.
1990 (print_symbol_value_1): Likewise.
1991 (resolve_symbol_value): Likewise. Don't divide local sym values
1992 by OCTETS_PER_BYTE twice.
1996 * config/tc-ia64.c (md_pseudo_table): Add 2byte, 4byte, and 8byte.
1998 * config/tc-ia64.c (is_conditional_branch): Rewrite to exclude mod
2003 * read.c (s_fill): Correct spelling error in comments.
2005 * doc/c-m88k.texi: New file.
2006 * doc/Makefile.am (CPU_DOCS): Add c-m88k.texi.
2007 * doc/Makefile.in: Regenerate.
2008 * doc/all.texi: Set M880X0.
2009 * doc/as.texinfo: Hook in m88k-dependent documentation.
2013 * doc/c-mips.texi (MIPS Opts): Document 12000 as valid cpu.
2015 * config/tc-mips.c (mips_cpu_info_table): Add rm5200, rm5231, rm5261,
2016 and rm5721 as valid cpu strings.
2017 (md_show_usage): Add rm5200, rm5231, rm5261, rm5721.
2021 * config/tc-mips.c (md_apply_fix): Prevent addend from becoming zero
2022 if it's expected to be non-zero.
2026 * config/tc-sh.c (sh_frob_section): Replace symbol_get_frag() with
2029 * write.c (TC_FINALIZE_SYMS_BEFORE_SIZE_SEG): Default to 1.
2030 (write_object_file): Set finalize_syms to
2031 TC_FINALIZE_SYMS_BEFORE_SIZE_SEG just before size_segs is
2034 * doc/internals.texi: Document
2035 TC_FINALIZE_SYMS_BEFORE_SIZE_SEG.
2039 * config/tc-ppc.c: Formatting fixes.
2043 * config/tc-m88k.c: Back out warning fixes from 2001-06-24, as
2044 they cause some subtle breakage. Will fix them again later.
2048 * config/obj-coff.c (obj_coff_ln): Treat a .ln directive
2049 outside of a function as a .appline directive.
2054 * config/tc-mips.c (mips_arch): New. Use mips_arch instead
2055 of mips_cpu for the ISA selection.
2056 (md_longopts): Add OPTION_MARCH and OPTION_MTUNE.
2057 (md_parse_option): Handle OPTION_MARCH and OPTION_MTUNE.
2058 (mips_tune): New. Use mips_tune for scheduling and optimization
2060 (append_insn): Use mips_tune and mips_arch.
2061 (macro_build): Ditto.
2063 (md_begin): Handle mips_arch, mips_tune and mips_cpu. For
2064 backwards compatability mips_cpu generates arch and tune.
2065 (md_show_usage): Document new behavior.
2067 * doc/c-mips.texi (MIPS Opts): Document -march and -mtune.
2074 * write.c (write_object_file): Do not set finalize_syms until
2075 after the segments have been sized. Some backends may still
2076 need to access the syms's frags in order to adjust relaxed
2079 * config/tc-arm.c (do_ldst): Use MVN to build simple inverted
2084 * write.c (fixup_segment <Difference of 2 syms same seg>): Don't
2085 subtract MD_PCREL_FROM_SECTION for 68k.
2086 (fixup_segment <Difference of 2 syms different seg>): Only fudge
2087 by adding MD_PCREL_FROM_SECTION for 68k or if not already pcrel.
2091 * write.c (fixup_segment <Difference of 2 syms same seg>): If
2092 pcrel, subtract MD_PCREL_FROM_SECTION value.
2096 * stabs.c (generate_asm_file): Make local variable `len' a size_t.
2100 * config/tc-alpha.c (tc_gen_reloc): Handle relocs against SEC_MERGE
2101 section symbols the same way as externs.
2105 * config/tc-m88k.c (md_number_to_imm): Remove; unused since 1993.
2106 (emit_relocations): Ditto.
2108 (md_begin): Reformat comments to conform to the GNU standards.
2109 (md_assemble): Ditto.
2111 * config/tc-m88k.c (get_reg): Adjust type of `reg_prefix' to char.
2112 (md_parse_option): Mark parameters as unused.
2113 (md_show_usage): Ditto.
2114 (calcop): Adjust type of `reg_prefix' to char.
2116 (getval): Adjust type of local `c' to char.
2117 (md_create_short_jump): Mark from_addr, to_addr params as unused.
2118 (md_create_long_jump): Ditto.
2119 (md_estimate_size_before_relax): Mark parameters as unused.
2121 * config/tc-m88k.c (md_pseudo_table): Properly terminate the table
2122 to allay a possible compiler warning.
2126 * doc/internals.texi (CPU backend): Document md_atof.
2130 * config/tc-hppa.c (pre_defined_registers): Add %mrp (millicode
2131 return pointer) alias for %r2 or %r31, depending on 32 or 64 bit
2136 * config/tc-arm.c (cplong_flag): Rename to long_flag. All uses
2138 (insns): Remove entry for adrl. Add long_flag for adr.
2139 (do_adr): Handle `l' suffix.
2144 * config/tc-hppa.c (CHECK_FIELD_WHERE): Define.
2145 (md_apply_fix): Use it here. Replace printf with equivalent
2147 (tc_gen_reloc): Use as_bad_where instead of as_bad.
2148 (md_apply_fix): Here too.
2149 * config/tc-i386.c (tc_gen_reloc): Use as_bad_where instead of as_bad.
2150 * config/tc-m68k.c (tc_gen_reloc): Likewise.
2151 (md_convert_frag_1): Likewise.
2155 * config/tc-ppc.c (ppc_comm): Change default alignment to 4 bytes.
2159 * doc/Makefile.am (info): Depend on $(MANS).
2160 (as.1): Remove the prefix `$(srcdir)/'.
2161 * doc/Makefile.in: Regenerated.
2167 * config/tc-arm.c (do_msr): Remove restriction on usage of
2172 * dwarf2dbg.c (dwarf2_finish): Output file info even when no
2177 * config/tc-arm.c (thumb_shift): Improve wording of error message.
2178 (do_t_arit): Likewise.
2182 * config/tc-sh.c (md_pseudo_table): Only intercept the .file and
2183 .loc pseudos if the dfwarf2 functions are available.
2184 (md_assemble): Only call dwarf2_emit_insn if it is available.
2186 * expr.c: Fix typo in comment.
2190 * doc/as.texinfo (Infix Ops): Document that comparison and
2191 combiner operators can be used as infix operators.
2195 * config/tc-m68k.c (parse_mri_control_operand): Fix handling
2197 (swap_mri_condition): Add HS (alias fo CC) and LO (alias for CS).
2198 (reverse_mri_condition): Likewise.
2199 (swap_mri_condition): Issue warning for conditions that can not be
2201 (build_mri_control_operand): Fix order of operands (swapped).
2202 (build_mri_control_operand): Allow upper case extension in structured
2204 (s_mri_else): Likewise.
2205 (s_mri_next): Likewise.
2206 (s_mri_for): Likewise.
2207 (s_mri_if): Fix handling comment ('*') in mri mode.
2208 (s_mri_while): Likewise.
2209 * macro.c (macro_expand): Allow macro invocation with empty
2214 * config/tc-arm.c: Fix test for overlow of literal pool.
2218 * NEWS: Updated for the new -n option for the MIPS assembler.
2220 * config/tc-mips.c (md_show_usage): Add -n.
2222 * doc/as.texinfo: Document the new -n option.
2223 * doc/c-mips.texi: Likewise.
2224 * doc/as.1: Regenerated.
2228 * config/obj-coff.c (obj_coff_section): Formatting fix.
2230 * config/tc-i386.c (md_assemble): Accept branch hints as ",pt" and
2235 * config/tc-mips.c (warn_nops): New variable. Set to 0 to
2236 disable warning about all NOPS that the assembler generates.
2237 (macro): Warn NOPS generated only if warn_nops is not 0.
2238 (md_shortopts): Add `n'.
2239 (md_parse_option): Set warn_nops to 1 for `n'.
2243 * config/tc-mips.c (mips_ip): Properly handle illegal operands.
2247 * config/tc-mips.c (md_apply_fix): Don't adjust common
2248 extern/weak symbols for ELF.
2249 (md_estimate_size_before_relax): Treat weak like extern for
2251 (mips_fix_adjustable): Don't adjust extern/weak symbols for
2256 * Makefile.am: 'som' is not wrongly spelled 'some'.
2257 * Makefile.in: Regenerate.
2259 * config/tc-mips.c (mips16_mark_labels): Reduce number of calls to
2260 S_GET_VALUE by using a temp.
2261 (append_insn): Likewise, and for S_SET_VALUE too.
2262 (mips_emit_delays): Likewise.
2263 (my_getExpression): Likewise.
2264 (md_apply_fix): Likewise. Use "valueT" rather than "long" for "value".
2265 (mips16_extended_frag): Remove code concerned with avoiding
2266 locking in a frag address now that symbols are not finalized until
2267 relaxation is complete. Cater for first relaxation pass having
2268 bogus addresses. Use relax_marker to reliably determine whether a
2269 symbol frag has been reached on the current pass.
2273 * configure.in: Move "mips-*-linux-gnu*" before "mips-*-gnu*".
2274 * configure: Regenerate.
2278 * configure.in: Use MIPS_STABS_ELF for Linux/mips and remove
2280 * configure: Regenerate.
2284 * config/tc-mips.c (mips_pseudo_table): Add "extern" if
2285 MIPS_STABS_ELF is defined.
2289 * as.c (show_usage): Remove L from listing options. It is not a
2294 * config/tc-z8k.c: Removed many warnings by removing unused
2295 variables and tagging unused parameters as such.
2296 (md_begin): Fixed a typo (","instead of ";").
2297 (struct ctrl_table): Add parentheses to initialize array
2299 (struct flag_table): Likewise.
2300 (struct intr_table): Likewise.
2301 (struct table): Likewise.
2302 (check_operand): "#if 0"'ed since it doesn't seem to be used.
2306 * config/tc-m68k.c (md_show_usage): Add all supported ColdFire
2307 options to list (e.g. m5206e, m5307, m5407).
2311 * config/tc-s390.h (TC_FORCE_RELOCATION): Always emit relocations
2312 BFD_RELOC_VTABLE_INHERIT and BFD_RELOC_VTABLE_ENTRY.
2316 * config/tc-i386.c (md_assemble): Handle Pentium4 branch hints.
2317 <JumpByte, JumpDword insn output>: Remove dead code.
2321 * as.c (parse_args): Correct option name "listing-lhs-width2".
2325 * symbols.c (S_SET_EXTERNAL): Do not override a section symbol's
2330 * read.c (emit_leb128_expr): Call md_cons_align.
2334 * config/tc-sparc.c (md_apply_fix3): Handle BFD_RELOC_SPARC_UA16,
2335 BFD_RELOC_SPARC_UA32 and BFD_RELOC_SPARC_UA64.
2336 (tc_gen_reloc): Likewise.
2337 (sparc_cons_align): Don't clear sparc_no_align_cons.
2338 (cons_fix_new_sparc): Substitute BFD_RELOC_SPARC_UA{16|32|64} for
2339 BFD_RELOC_{16|32|64} iff sparc_no_align_cons is set.
2343 * write.c (relax_segment <rs_space>): Check the section of a
2344 symbol to determine whether it is absolute rather than comparing
2345 its frag against zero_address_frag as listings play tricks with
2348 * config/tc-m68k.c (md_assemble): Ensure variable part of frag is
2349 allocated in the same chunk as the fixed part.
2353 * write.c (relax_segment <rs_space>): Calculate growth using
2354 addresses before stretch. Prevent repeated error messages.
2356 Use as_bad_where and as_warn_where to pinpoint errors.
2360 * symbols.c (resolve_symbol_value): Always set segment, even when
2361 not finalizing symbol value.
2363 * config/obj-ieee.c (write_object_file): Set finalize_syms.
2364 * config/obj-coff.c (write_object_file): Likewise.
2365 * (size_section): Remove rs_space assert as fr_symbol is no longer
2367 (fill_section): Likewise.
2369 * configure.in: Replace linuxoldld with linux*oldld.
2370 * configure: Regenerate.
2373 * config/tc-i386.c (md_convert_frag): Don't die on local symbols
2374 that have been finalized.
2376 * symbols.c (resolve_symbol_value): Clear sy_resolving before
2377 taking exit_dont_set_value.
2379 * write.c (relax_segment <rs_space>): Don't zap fr_symbol when
2384 * config/obj-coff.c (add_lineno): xcoff allows negative line
2386 * config/tc-ppc.c (ppc_stabx): fix generated symbol
2390 * write.c (write_object_file): Set finalize_syms = 1 before
2395 * config/tc-mips.c (ISA_HAS_64BIT_REGS): Add ISA_MIPS64 as 64 bit
2396 architecture, remove erraneous ISA_MIPS32.
2397 (md_show_usage): Add MIPS r12k support.
2398 (mips_cpu_info_table): Add MIPS r12k support.
2402 * config/tc-arc.c (md_assemble): Use is_end_of_line instead of
2405 * symbols.c (resolve_symbol_value): Remove "finalize" param,
2406 instead use finalize_syms directly. Don't treat expressions
2407 specially with regard to finalize_syms. Update calls to self.
2408 (resolve_local_symbol): Update call to resolve_symbol_value.
2409 (S_GET_VALUE): Likewise. Return resolve_symbol_value if
2411 * symbols.h (resolve_symbol_value): Update prototype.
2412 * config/obj-aout.c (obj_crawl_symbol_chain): Update call
2413 to resolve_symbol_value.
2414 * config/obj-bout.c (obj_crawl_symbol_chain): Likewise.
2415 * config/obj-coff.c (do_relocs_for): Likewise.
2416 (yank_symbols): Likewise.
2417 (fixup_segment): Likewise.
2418 * config/obj-vms.c (obj_crawl_symbol_chain): Likewise.
2419 * config/tc-mips.c (md_convert_frag): Likewise.
2420 * config/tc-ppc.c (ppc_frob_symbol): Likewise.
2421 (ppc_fix_adjustable): Likewise.
2422 * dwarf2dbg.c (dwarf2dbg_estimate_size_before_relax): Likewise.
2423 (dwarf2dbg_convert_frag): Likewise.
2424 * ehopt.c (eh_frame_estimate_size_before_relax): Likewise.
2425 (eh_frame_convert_frag): Likewise.
2426 * expr.c (make_expr_symbol): Likewise.
2427 * write.c (adjust_reloc_syms): Likewise.
2428 (write_object_file): Likewise.
2429 (relax_segment): Likewise.
2430 (fixup_segment): Likewise.
2431 (finalize_syms): Init to zero, and update comment.
2432 (write_object_file): Set finalize_syms to 1 rather than 2.
2433 * doc/internals.texi (sy_value): Mention finalize_syms.
2434 (S_GET_VALUE): Remove restriction on when S_GET_VALUE can be called.
2436 * config/tc-m68k.c (relaxable_symbol): Only treat external symbols
2437 as relaxable if embedded system, make weak syms non-relaxable.
2439 (tc_m68k_fix_adjustable): ..so it can be used here.
2440 (md_apply_fix_2): Sign extend without conditional.
2442 * config/tc-v850.c: Update copyright.
2446 * cgen.c (gas_cgen_tc_gen_reloc): Changed error message when
2447 howto entry is not found.
2451 * config/tc-mn10300.c (mn10300_force_relocation): Don't
2452 optimize differences between symbols in code sections to
2454 (mn10300_fix_adjustable): Don't adjust to section+offset
2455 relocations pointing at symbols in code sections.
2459 * config/tc-arc.c (md_assemble): Correct dwarf2_emit_insn param
2461 * config/tc-i386.c (md_assemble): Call dwarf2_emit_insn before
2462 opcodes are output rather than after. Delete insn_size.
2463 * config/tc-v850.c (md_assemble): Similarly, but delete
2468 * ehopt.c (eh_frame_convert_frag): Fix missed subtype adjustment
2473 * ehopt.c (get_cie_info): Rename from eh_frame_code_alignment;
2474 also collect whether to expect an FDE augmentation.
2475 (check_eh_frame): Rewrite as a state machine. Track where in
2476 an FDE we are located, skip any augmentation.
2477 (eh_frame_estimate_size_before_relax): Get code alignment from
2478 the fragment subtype.
2479 (eh_frame_relax_frag, eh_frame_convert_frag): Likewise.
2480 * read.c (emit_leb128_expr): Call check_eh_frame.
2484 * config/tc-mn10300.c (md_assemble): Anchor dwarf2 line info
2485 before a relaxable insns.
2489 * config/tc-mn10300.c (tc_gen_reloc): Don't reject differences
2490 between symbols if the base symbol is in the current section;
2491 emit a PC-relative relocation instead.
2495 * config/tc-arc.c: Update copyright and tidy source comments.
2496 (md_pseudo_table): Add directive .cpu back as an alias for
2497 .option. Add .file and .line for dwarf2 support.
2498 (arc_mach_type): Make bfd_mach_arc_6 default.
2499 (md_longopts): Add entry 'pre-v6' representing old command line
2500 option when assembling for 'arc5' core versions.
2501 (md_parse_option): Make OPTION_ARC same as OPTION_ARC6, for new
2503 (arc_code_symbol): Make symbol value for @h30 fixup expression
2504 equal to O_constant.
2505 (md_assemble): Call dwarf2_emit_insn.
2506 Include "dwarf2dbg2.h". Formatting fixes throughout file.
2507 * config/tc-arc.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
2508 * doc/c-arc.texi (ARC_CORE_DEFAULT): Update to new default.
2512 * config/tc-ia64.c (special_linkonce_name): New.
2513 (make_unw_section): Map .gnu.linkonce.t.FOO text section into
2514 .gnu.linkonce.ia64unw{,i}.FOO.
2515 (ia64_elf_section_type): Handle .gnu.linkonce.ia64unw{,i}.FOO.
2516 (dot_endp): Add comment about it.
2520 * config/tc-arm.c (arm_handle_align): When truncating an aligned
2521 block, ensure that the low order bits of the alignment are
2526 * config/obj-vms.c (obj_crawl_symbol_chain): Don't take address of
2529 * config/tc-fr30.c (md_estimate_size_before_relax): Return size of
2530 current variable part of frag.
2531 * config/tc-m32r.c (md_estimate_size_before_relax): Likewise.
2532 * config/tc-openrisc.c (md_estimate_size_before_relax): Likewise.
2533 * config/tc-m68hc11.c (RELAX_STATE): Define.
2534 (RELAX_LENGTH): Define.
2535 (md_estimate_size_before_relax): Handle non-relaxable cases
2536 separately from relaxable cases for clarity, and return correct
2537 size for multi-pass relaxation.
2538 * config/tc-tahoe.c (RELAX_LENGTH): Correct.
2539 (md_estimate_size_before_relax): As for tc-m68hc11.c.
2540 (md_convert_frag): Remove "length_code".
2541 * config/tc-vax.c (RELAX_STATE): Define.
2542 (RELAX_LENGTH): Define.
2543 (md_relax_table): Add missing entry.
2544 (md_estimate_size_before_relax): As for tc-m68hc11.c.
2545 (md_convert_frag): Remove "length_code".
2546 * config/tc-ns32k.c (md_estimate_size_before_relax): Simplify and
2547 don't bother setting fr_var. Return correct size for multi-pass
2549 * config/tc-v850.c (md_estimate_size_before_relax): Rewrite.
2550 (md_convert_frag): Don't bother clearing fr_var.
2551 (md_pseudo_table): Correct initialization.
2552 * config/tc-h8500.c (md_convert_frag): Don't bother clearing fr_var.
2553 (md_estimate_size_before_relax): No need to set fr_var.
2554 * config/tc-mcore.c (md_convert_frag): Don't bother clearing fr_var.
2555 (md_estimate_size_before_relax): No need to set fr_var.
2559 * config/tc-ia64.c (generate_unwind_image): Align the fragment
2560 beginning a function's unwind info block.
2564 * config/tc-mn10300.c (md_apply_fix3): Accept PC-relative relocs.
2568 * config/tc-m68k.c: Instead of replacing -1 by 64 in assignment to
2569 fx_pcrel_adjust explicitly sign extend when reading it.
2573 * config/tc-m68k.c (md_estimate_size_before_relax): Set fr_var
2574 from md_relax_table, and combine some switch cases.
2578 * config/tc-mn10300.c (md_assemble): Subtract operand->shift
2579 from offset in non-pcrel operands too.
2583 * config/tc-arm.h (MAX_MEM_FOR_RS_ALIGN_CODE): Define.
2584 * config/tc-arm.c (arm_handle_align): Do not insert more than
2585 MAX_MEM_FOR_RS_ALIGN_CODE bytes.
2586 (arm_frag_align_code): Use MAX_MEM_FOR_RS_ALIGN_CODE.
2590 * Makefile.am (TARG_ENV_HFILES): Add te-hppa64.h and te-hppalinux64.h.
2592 * Makefile.in: Regenerate.
2593 * configure.in: Remove duplicate mips-*-ecoff* entry.
2594 * configure: Regenerate.
2596 * config/obj-ecoff.c (obj_pseudo_table): Fix terminating entry.
2597 (n_names): Cast away type mismatch.
2598 (ecoff_sec_sym_ok_for_reloc): Add unused attribute.
2599 (obj_ecoff_frob_symbol): Likewise.
2600 * ecoff.c: (add_file): Add unused attribute.
2601 (ecoff_directive_begin): Likewise.
2602 (ecoff_directive_bend): Likewise.
2603 (ecoff_directive_def): Likewise.
2604 (ecoff_directive_dim): Likewise.
2605 (ecoff_directive_scl): Likewise.
2606 (ecoff_directive_size): Likewise.
2607 (ecoff_directive_type): Likewise.
2608 (ecoff_directive_tag): Likewise.
2609 (ecoff_directive_val): Likewise.
2610 (ecoff_directive_endef): Likewise.
2611 (ecoff_directive_end): Likewise.
2612 (ecoff_directive_ent): Likewise.
2613 (ecoff_directive_extern): Likewise.
2614 (ecoff_directive_file): Likewise.
2615 (ecoff_directive_fmask): Likewise.
2616 (ecoff_directive_frame): Likewise.
2617 (ecoff_directive_mask): Likewise.
2618 (ecoff_directive_loc): Likewise.
2619 (mark_stabs): Likewise.
2620 (ecoff_stab): Likewise.
2621 (ecoff_frob_symbol): Cast away type mismatch.
2622 (ecoff_padding_adjust): Likewise.
2623 (ecoff_build_symbols): Likewise.
2624 (ecoff_build_procs): Likewise.
2625 (ecoff_build_aux): Likewise.
2626 (ecoff_build_strings): Likewise.
2627 (ecoff_build_fdr): Likewise.
2628 (ecoff_build_debug): Likewise.
2629 * itbl-ops.c (itbl_assemble): Variable initialization.
2633 * config/tc-i386.c (i386_displacement): Call as_bad for bad GOTOFF
2634 expressions rather than triggering an assert.
2638 * config/tc-openrisc.c: New file.
2639 * config/tc-openrisc.h: Likewise.
2641 * Makefile.am: Add OpenRISC target.
2642 * Makefile.in: Regenerated.
2644 * configure.in (openrisc-*-*): Add target.
2645 * configure: Regenerated.
2649 * config/tc-arm.c (arm_frag_align_code): Change error message to
2650 more explanatory version.
2654 * config/tc-mips.c (md_parse_option): Also accept
2655 elf64-tradbigmips and elf64-tradlittlemips for OPTION_64.
2659 * config/tc-ia64.c (dot_spillmem_p): Fix output_spill_?sprel_p()
2660 argument passing order: predicate goes last, not first.
2664 * configure.in: Add arm-vxworks.
2665 * configure: Regenerate.
2669 * config/tc-arm.c (arm_handle_align): New Function: Generate
2670 no-op filled alignment frags.
2671 (arm_frag_align_code): New Function: Create a code alignment frag.
2672 (arm_init_frag): New Function: Initialse the target dependent
2675 * config/tc-arm.h (TC_FRAG_TYPE): Define.
2676 (TC_FRAG_INIT): Define.
2677 (HANDLE_ALIGN): Define.
2678 (md_do_align): Define.
2682 * config/obj-coff.c (do_linenos_for): Check to see if the filename
2683 symbol has been initialised before extracting its symbol index.
2687 * configure: Regenerated with the right version of autoconf.
2691 * config/tc-z8k.c (build_bytes): 12 and 16 bit displacements now
2692 generate R_CALLR and R_REL16 relocations
2696 * config/tc-d10v.h (tc_frob_label): Update the symbol's frag
2697 since frag_now can change after d10v_cleanup is called.
2701 * configure.in: Add the em type for FreeBSD targets.
2702 * configure: Regenerate.
2706 * struc-symbol.h (struct local_symbol): New TC_LOCAL_SYMFIELD_TYPE.
2707 * symbols.c (local_symbol_convert): Call TC_LOCAL_SYMFIELD_CONVERT.
2709 * config/tc-mn10300.c (md_assemble): Simplify offset adjustment of
2710 pc-relative relocations not placed at the end of the instruction.
2714 * tc-ia64.c (is_conditional_branch): Return true for br, brl, and br.
2719 * config/obj-elf.c (obj_elf_change_section): Add entsize argument,
2720 handle SHF_MERGE and SHF_STRINGS.
2721 (obj_elf_parse_section_letters): Set SHF_MERGE and SHF_STRINGS.
2722 (obj_elf_section): Allow additional argument specifying entity size.
2723 * write.c (adjust_reloc_syms): Keep relocations against local symbols
2724 in SEC_MERGE sections.
2728 * dwarf2dbg.c (process_entries): Don't optimize redundant line notes.
2732 * expr.c (operand): Pass &c to md_parse_name().
2733 * config/tc-ia64.h, config/tc-ppc.h, config/tc-tic54x.h: Adjust.
2737 * config/tc-mips.c: Support ELF64 for traditional MIPS targets.
2739 * Makefile.am: (TARG_ENV_HFILES): Add tc-mips.h.
2740 * Makefile.in: Regenerated.
2742 * configure.in: Use traditional MIPS targets for Linux/MIPS.
2743 * configure: Regenerated.
2747 * config/tc-mn10300.c (xr_registers): Added `pc'.
2751 * configure.in: Add h8500-*-coff and h8500-*-rtems targets.
2752 * configure: Regenerate.
2754 * config/tc-h8500.c (md_estimate_size_before_relax): Add missing
2755 cases, and always return size based on current fr_subtype.
2756 (md_begin): Move initialization of md_relax_table..
2757 (md_relax_table): ..to static initializer. Set rlx_length for
2758 UNDEF_WORD_DISP cases.
2759 * config/tc-w65.c (md_estimate_size_before_relax): Likewise.
2760 (md_begin): Likewise.
2761 (md_relax_table): Likewise.
2762 * config/tc-mcore.c (md_estimate_size_before_relax): Likewise.
2763 (md_relax_table): Set rlx_length for UNDEF_WORD_DISP cases.
2764 Set rlx_backward and rlx_forward to zero for unused states.
2765 * config/tc-sh.c (md_estimate_size_before_relax): Likewise.
2766 (md_relax_table): Set rlx_length for UNDEF_WORD_DISP cases.
2767 (UNCOND12, UNCOND32): Remove duplicate defines.
2771 * config/tc-cris.c (md_estimate_size_before_relax) <case
2772 ENCODE_RELAX (STATE_CONDITIONAL_BRANCH, STATE_UNDF)>: Don't emit
2773 32-bit branch, just set fragP->fr_subtype. Set fragP->fr_var.
2774 <all cases>: Always set fragP->fr_var using md_cris_relax_table.
2775 Add cases to cover all relax states.
2776 (cris_insn_first_word_frag): New.
2777 (md_assemble): Call cris_insn_first_word_frag to get the first
2778 frag in an insn, not frag_more. Don't call dwarf2_emit_insn at
2779 end. Drop variable insn_size.
2780 (gen_bdap): Call cris_insn_first_word_frag, not frag_more.
2781 (cris_sym_leading_underscore): Wrap first as_bad parameter in _().
2782 (cris_sym_no_leading_underscore, s_cris_file, s_cris_loc): Ditto.
2784 * write.c (write_object_file): Reset broken word state before
2789 * listing.c (listing_listing): Enable listing on EDICT_NOLIST_NEXT
2790 for one line if not already enabled.
2791 * cond.c (s_elseif): Correct conditional assembly listing.
2794 * cond.c (s_endif): Correct handling of "if .. elseif .." trees.
2795 Don't abort on NULL current_cframe.
2799 * dwarf2dbg.c (dwarf2_directive_file): Fix warnings.
2800 (dwarf2dbg_convert_frag): Pass `finalize_syms' to resolve_symbol_value.
2801 * config/obj-aout.c (obj_crawl_symbol_chain): Likewise.
2802 * config/obj-bout.c (obj_crawl_symbol_chain): Likewise.
2803 * config/obj-coff.c (do_relocs_for): Likewise.
2804 (yank_symbols): Likewise.
2805 (fixup_segment): Likewise.
2806 * config/tc-ppc.c (ppc_frob_symbol): Likewise.
2807 (ppc_fix_adjustable): Likewise.
2808 * config/tc-mips.c (md_convert_frag): Likewise.
2809 * config/obj-vms.c (obj_crawl_symbol_chain): Likewise.
2810 * ehopt.c (eh_frame_convert_frag): Likewise.
2811 * expr.c (make_expr_symbol): Likewise.
2813 * frags.h (struct frag): Add last_fr_address. Reorder fields for
2815 * symbols.c (resolve_symbol_value): Don't fix expression values
2816 until relaxation is complete.
2817 (resolve_local_symbol): Pass `finalize_syms' to resolve_symbol_value.
2818 (S_GET_VALUE): Likewise, and return unresolved expression value.
2819 * write.c (finalize_syms): New.
2820 (relax_and_size_seg): Split into..
2821 (relax_seg): New function, returns 1 if anything changed..
2822 (size_seg): And the remainder of relax_and_size_seg.
2823 (fixup_segment): Arrange for final resolution of sym values.
2824 (adjust_reloc_syms): Likewise.
2825 (write_object_file): Likewise, and repeatedly call relax_seg until
2826 nothing more changes.
2827 (relax_segment): Return 1 if anything changed. Use correct types
2828 for rs_org `target' and `after'.
2829 * write.h (finalize_syms): Declare.
2830 (relax_segment): Update prototype.
2832 * config/tc-sh.c (md_estimate_size_before_relax): Add extra
2833 do-nothing cases to switch to avoid abort on a second relaxation
2834 pass, and tidy code a little.
2835 * config/tc-cris.c (md_estimate_size_before_relax): Likewise.
2836 * config/tc-h8500.c (md_estimate_size_before_relax): Likewise.
2837 * config/tc-w65.c (md_estimate_size_before_relax): Likewise.
2838 * config/tc-i386.c (UNCOND_JUMP, COND_JUMP, COND_JUMP86): Decrement.
2839 (md_relax_table): Remove first four unused entries. Increment
2840 rlx_length by one throughout table, and update comments to suit.
2841 (md_estimate_size_before_relax): Return size of current variable
2842 part of frag to reflect reality when relaxing more than once.
2843 * config/tc-mcore.c (COND12, UNCD12): Rename to DISP12 throughout.
2844 (COND32, UNCD32): Rename to DISP32 throughout.
2845 (UNDEF_WORD_DISP): Renumber to 3.
2846 (md_estimate_size_before_relax): Add extra do-nothing cases.
2847 * config/tc-mn10200.c (md_estimate_size_before_relax): Rewrite.
2848 * config/tc-mn10300.c (md_estimate_size_before_relax): Rewrite.
2849 * config/tc-ns32k.c (md_estimate_size_before_relax): Add cases to
2850 handle word and dword branches.
2854 * config/tc-cris.h (tc_fix_adjustable): Allow only
2855 BFD_RELOC_CRIS_32_GOTREL of the PIC relocs.
2856 * config/tc-cris.c (cris_get_pic_suffix): Correct reloc type in
2857 example in comment to valid type.
2861 * read.c (equals): Set to local for COFF only if it hasn't been
2866 * config/tc-hppa.c (DEFAULT_LEVEL): Define.
2867 (md_begin): Use it when setting default architecture.
2871 * read.c (equals): (for COFF) default symbols to being local.
2875 * configure.in (cris-*-*): Change default emulation to criself.
2876 (cris-*-*aout*): New rule.
2877 * configure: Regenerate.
2881 * config/tc-s390.c (tc_s390_fix_adjustable): Add test for
2882 BFD_RELOC_390_GOTENT.
2883 * config/tc-s390.h (TC_RELOC_RTSYM_LOC_FIXUP): Add test for
2884 BFD_RELOC_390_GOTENT.
2888 * config/tc-s390.h (TC_FORCE_RELOCATION): Define.
2889 * config/tc-s390.c (s390_force_relocation): New function: Force
2890 relocations for VTINHERIT relocs.
2894 * doc/as.texinfo: Put @c man begin to generate the as man page.
2895 When generating man, define all the variables. Re-organize some
2896 options to obtain better indentation of man page.
2897 * doc/Makefile.am (MANCONF, TEXI2POD, POD2MAN): New variable.
2898 (as.1): Build from as.texinfo.
2899 * doc/Makefile.in: Regenerate.Index: binutils/Makefile.am
2903 * config/tc-i386.c (i386_scale): Accept an absolute expression for
2904 scale factor, and return the end of the expression.
2905 (i386_operand): Modify for above.
2909 * doc/as.texinfo: Document --listing-XXX command line switches.
2910 Explain why listings behave differently when -pipe is used.
2914 * write.c (fix_new_exp): Print an error if passed a register.
2918 * config/tc-cris.c: Tweak attribution. Fix typos. PIC support.
2919 (PIC_SUFFIX_CHAR): New macro.
2920 (cris_get_pic_suffix, cris_get_pic_reloc_size): New functions.
2921 (pic): New variable.
2922 (md_longopts): New option --pic.
2923 (OPTION_PIC): Define.
2924 (md_estimate_size_before_relax): Tweak longish comment.
2925 (md_create_long_jump): Make the long jumps generate ADD.D offset,PC.
2926 (md_assemble): Handle a PIC relocation on prefix and normal
2927 instruction operands.
2928 <output_instruction.insn_type == CRIS_INSN_BRANCH>: Handle
2929 "complex" operand expressions. Call frag_more outside
2930 gen_cond_branch_32 parameter list.
2931 (cris_process_instruction): Handle PIC relocs on parsed
2932 operand expressions. Validize PIC reloc sizes.
2933 (get_autoinc_prefix_or_indir_op): Handle PIC suffix.
2934 (gen_bdap): Handle "complex" operand expressions.
2935 (gen_cond_branch_32): Use as_warn_where, not as_warn. Use
2936 ADD.D offset,PC as jump instruction if generating PIC. Generate
2937 error instead of crashing on complex expressions.
2938 (cris_number_to_imm): Add segT parameter. All callers changed.
2939 Use segT parameter, not now_seg, for segment of fixup. Handle
2940 PCREL relocations but check that they are fully resolved.
2941 (md_parse_option): Handle OPTION_PIC.
2942 (tc_gen_reloc): Handle PIC relocs.
2943 (md_show_usage): Update for --pic.
2944 (md_apply_fix3): Renamed from md_apply_fix.
2945 (md_pcrel_from): Accept emitting PCREL relocs when ELF.
2946 (md_cris_force_relocation): Force relocation for PIC relocs.
2948 * config/tc-cris.h: Tweak attribution.
2949 (MD_APPLY_FIX3): Define.
2950 (IS_CRIS_PIC_RELOC): New macro.
2951 (TC_RELOC_RTSYM_LOC_FIXUP): Define.
2952 (tc_fix_adjustable): Don't adjust a globally visible symbol when
2954 (tc_frob_symbol): Avoid emitting undefined symbols.
2958 * frags.h (struct frag): Add relax_marker.
2959 * write.c (is_dnrange): Delete.
2960 (relax_frag): Use correct types for `aim', `target', `address'.
2961 Delete `offset', `was_address'. Test `relax_marker' instead of
2962 using fragile (and slow) address test.
2963 (relax_segment): Init and flip `relax_marker'.
2967 * config/tc-i386.c (md_assemble <REGISTER_WARNINGS>): Correct
2972 * config/tc-m68hc11.c (md_pseudo_table): Recognize xrefb to comply
2973 with 'Motorola specification for assembly language input standard'.
2977 * dwarf2dbg.c (user_filenum, user_filenum_allocated): Remove.
2978 (dwarf2_directive_loc): Don't use them.
2979 (dwarf2_directive_file): Reject duplicate file definitions.
2980 (get_filenum): Zero allocated memory.
2981 (out_file_list): Complain about missing file definitions.
2985 * read.c (do_org): Handle complex expressions.
2986 * cgen.c (gas_cgen_finish_insn): Likewise.
2990 * config/tc-ia64.c (md): New member keep_pending_output.
2991 (ia64_flush_pending_output): Flush only if md.keep_pending_output
2993 (dot_xdata): Turn on md.keep_pending_output for the duration of
2995 (dot_xfloat_cons): Ditto.
2996 (dot_xstringer): Ditto.
2997 (dot_xdata_ua): Ditto.
2998 (dot_xfloat_cons_ua): Ditto.
3002 * config/tc-ia64.c (ia64_unrecognized_line, case '['): Add local
3007 * config/tc-sh.c (parse_reg): Match capital MACH and MACL.
3011 * stabs.c (s_stab_generic): Don't corrupt the notes obstack by
3012 blindly freeing string if it isn't at the top of the obstack.
3016 * config/tc-i386.c (RELOC_ENUM): Define. Use throughout file.
3017 (NUM_FLAG_CODE): Define.
3018 (lex_got): New function.
3019 (got_reloc): New global var.
3020 (x86_cons_fix_new): New function.
3021 (x86_cons): New function.
3022 (i386_immediate): Use lex_got here, replacing inline code. Change
3023 "ignoring junk.." error message to "junk.."
3024 (i386_displacement): Likewise.
3025 * config/tc-i386.h (TC_PARSE_CONS_EXPRESSION): Define.
3026 (x86_cons): Declare.
3027 (TC_CONS_FIX_NEW): Define.
3028 (x86_cons_fix_new): Declare.
3032 * config/tc-arm.c (md_begin): Always set machine type based on
3037 * config/tc-i386.c (struct _i386_insn): Rename disp_reloc to reloc.
3038 (md_assemble) <smallest displacement>: Use correct field of i.op[]
3040 <JumpInterSegment output>: Use correct i.disp_reloc[].
3041 <immediate output>: Likewise.
3045 * config/tc-arm.c (md_apply_fix3): Clear bit zero of offset in
3050 * config/tc-m68k.c : Add 5407 to archs[] table.
3051 (HAVE_LONG_BRANCH): Add mcf5407.
3052 (select_control_regs): Recognize 5407.
3056 * config/tc-m32r.c (expand_debug_syms): Call frag_align_code rather
3061 * config/atof-ieee.c (TC_LARGEST_EXPONENT_IS_NORMAL): New macro.
3062 (gen_to_words): Print warnings if NaNs are found and the target CPU
3063 does not support them. Allow largest exponent to be used in normal
3064 numbers if TC_LARGEST_EXPONENT_IS_NORMAL evaluates to true.
3068 * config/tc-i386.c (tc_gen_reloc): Remove ugly hack which is not needed
3069 anymore since we use bfd_elf_generic_reloc now.
3070 (md_apply_fix3): Only apply hack for partial_inplace if not using RELA.
3074 * configure.in (BFD_VERSION): New.
3075 (AM_INIT_AUTOMAKE): Use $BFD_VERSION.
3076 * configure: Regenerate.
3077 * Makefile.am: Run "make dep-am"
3078 * Makefile.in: Regenerate.
3079 * doc/Makefile.in: Regenerate.
3083 * config/obj-coff.c [BFD_ASSEMBLER] (obj_coff_section): Set
3084 SEC_NEVER_LOAD when the 'n' flag is used.
3085 Add SEC_NEVER_LOAD to matchflags.
3089 * symbols.c (decode_local_label_name): Initialize message_format
3090 only when an error is reported (perf pb due to I18N).
3094 * dwarf2dbg.c (dwarf2_directive_file): Call s_app_file (0) if
3095 BFD_ASSEMBLER is not defined.
3099 * config/tc-mn10300.c (md_apply_fix3): Don't mark a fixup as
3100 done if it's against a symbol.
3104 * config/tc-ia64.c (ia64_target_format): Return ia64-aix-specific
3105 target formats if applicable.
3106 * config/te-ia64aix.h: New. Configuration for AIX5 on IA-64.
3107 * Makefile.am (TARG_ENV_HFILES): Added config/te-ia64aix.h.
3108 * Makefile.in: Regenerated.
3109 * configure.in: Added configuration for ia64-*-aix*.
3110 * configure: Regenerated.
3114 * config/tc-ia64.c (enum operand_match_result): New type.
3115 (operand_match): Change return type to operand_match_result.
3116 Fix all returns appropriately, adding support for returning the
3117 out-of-range result.
3118 (parse_operands): New locals result, error_pos, out_of_range_pos,
3119 curr_out_of_range_pos. Rewrite operand matching loop to give better
3124 * config/tc-ia64.c (struct unwind): Add member "prologue_count".
3125 (dot_proc): Clear unwind.prologue_count to zero.
3126 (dot_prologue): Increment unwind.prologue_count.
3127 (dot_restore): If second operand is omitted, use
3128 unwind.prologue_count -1 for "ecount" (# of additional regions to
3129 pop). Decrement unwind.prologue_count by number of regions
3134 * doc/as.texinfo (Section): Note that some flags to the COFF
3135 version of .section remove attributes rather than setting them.
3139 * config/tc-pdp11.c: Fix formatting.
3140 * config/tc-pdp11.h: Likewise.
3144 * config/tc-i386.c (tc_i386_fix_adjustable): Fix GOTPCREL GOT
3149 * configure.in (cpu_type, arch): Add a generic FreeBSD specification as
3150 all FreeBSD platforms should look the same at this level.
3151 * configure: Rebuilt.
3152 * config/tc-i386.c: Add support for old FreeBSD a.out hosts.
3156 * Makefile.am: Add PDP-11 target.
3157 * configure.in: Likewise.
3158 * config/tc-pdp11.c: New file.
3159 * config/tc-pdp11.h: New file.
3160 * doc/Makefile.am: Add PDP-11 documentation.
3161 * doc/all.texi: Likewise.
3162 * doc/as.texinfo: Likewise.
3163 * doc/c-pdp11.texi: New file.
3167 * cgen.c (gas_cgen_md_apply_fix3): Support BFD_RELOC_64.
3171 * config/tc-ia64.c (operand_match, case TAG13): Make a BFD_RELOC_UNUSED
3172 reloc instead of a 0 reloc.
3173 (md_apply_fix3): Check for BFD_RELOC_UNUSED instead of 0, and mark it
3175 * config/tc-ia64.h (TC_RELOC_RTSYM_LOC_FIXUP): Likewise.
3179 * write.c (is_dnrange): Stop as soon as the address becomes
3181 (relax_frag): Add segment parameter. Only call symbol_get_frag
3182 once. Only call is_dnrange if the symbol is in the same segment,
3183 and the symbol address is larger.
3184 (relax_segment): Pass segment to md_relax_frag and relax_frag.
3185 * write.h (relax_frag): Update declaration.
3186 * config/tc-fr30.c (fr30_relax_frag): Add segment parameter. Pass
3188 * config/tc-m32r.c (m32r_relax_frag): Likewise.
3189 * config/tc-m32r.h (md_relax_frag): Add segment parameter.
3190 (m32r_relax_frag): Update declaration.
3191 * config/tc-mips.h (md_relax_frag): Add segment parameter.
3192 * config/tc-tic54x.h (md_relax_frag): Likewise.
3193 * doc/internals.texi (CPU backend): Update documentation for
3198 * doc/c-i386.texi (i386-Arch): Add "jumps"/"nojumps" blurb.
3199 Mention effect of < 386 architectures on jump promotion.
3200 (i386-Jumps): xref above. Don't assume long disp is 32 bits.
3202 * config/tc-i386.c (no_cond_jump_promotion): New.
3203 (set_cpu_arch): Parse "jumps" arch modifier.
3204 (insn_size): Modify usage comment.
3205 (ENCODE_RELAX_STATE): Reformat and protect macro arg.
3206 (SIZE_FROM_RELAX_STATE): Rename to DISP_SIZE_FROM_RELAX_STATE.
3207 (TYPE_FROM_RELAX_STATE): New define.
3208 (UNCOND_JUMP, COND_JUMP): Renumber.
3209 (md_relax_table): Reorder to suit.
3210 (COND_JUMP86): New define.
3211 (md_relax_table): Handle COND_JUMP86 cases. Add a few comments.
3212 (md_assemble): Create frag var for jumps of max size, encode relax
3213 state for COND_JUMP86.
3214 (md_estimate_size_before_relax): Handle COND_JUMP86 cases, and
3215 leave conditional jumps small if no_cond_jump_promotion.
3216 (md_convert_frag): Likewise.
3218 * expr.c (operator): Don't bump input_line_pointer for two char
3219 operators. Instead return operator size via new param num_chars.
3220 (expr): Use above to parse multi-char operators correctly.
3224 * config/tc-i386.c (i386_displacement): Fix handling of
3225 BFD_RELOC_X86_64_GOTPCREL.
3226 (i386_validate_fix): Likewise.
3230 * config/tc-arm.c (do_ldst): Improve warnings for unpredictable
3231 ldrt/strt instructions.
3235 * config/tc-mips.c (macro): For M_LA_AB emit a
3236 BFD_RELOC_MIPS_CALL16 relocation or a
3237 BFD_RELOC_MIPS_CALL_HI16/BFD_RELOC_MIPS_CALL_LO16 pair instead of
3238 BFD_RELOC_MIPS_GOT16 and
3239 BFD_RELOC_MIPS_GOT_HI16/BFD_RELOC_MIPS_GOT_LO16, respectively for
3240 loading the jump register when generating SVR4_PIC code.
3244 * configure.in: Make 'mipself' and 'mipsecoff' emulations
3245 map to MIPS-specific files, as they used to do before the
3246 change on 2000-05-21.
3247 * configure: Regerate.
3251 * config/tc-mips.c (md_parse_option): Don't try to compile
3252 ELF-only option code if not ELF.
3256 * config/tc-ia64.h (md_elf_section_type): New macro.
3257 (ELF_TC_SPECIAL_SECTIONS): Drop .IA_64.unwind and .IA_64.unwind_info
3258 (they're now handled via ia64_elf_section_type.
3260 * config/tc-ia64.c (unwind): New members saved_text_seg,
3261 saved_text_subseg, and force_unwind_entry.
3262 (optimize_unw_records): New function to optimize away unnecessary
3264 (ia64_elf_section_type): New function.
3265 (output_unw_records): Generate unwind info only if the size is
3266 non-zero or if it's forced for some other reason (e.g.,
3267 handlerdata or a personality routine).
3268 (generate_unwind_image): Don't switch back to previous
3269 section---stay inside the unwind info section instead so that
3270 handlerdata that may follow goes into the right place.
3271 (dot_handlerdata): Force generation of unwind entry and save the
3272 current active text segment before generating unwind image.
3273 (dot_unwentry): Force generation of unwind entry.
3274 (dot_personality): Ditto.
3275 (dot_endp): Generate unwind table entry only if there is
3276 some unwind info or the unwind entry was forced.
3278 * config/tc-ia64.c (make_unw_section_name): New macro to form
3279 unwind section name.
3280 (generate_unwind_image): Add "text_name" argument. Use it to
3281 form unwind section name.
3282 (dot_handlerdata): Determine current segment (section) name and
3283 pass it to generate_unwind_image().
3284 (dot_endp): Determine current segment (section) name and use
3285 it to determine the appropriate unwind section name.
3286 (ia64_md_do_align): Add missing ATTRIBUTE_UNUSED declarations to
3287 n, fill, and max arguments.
3291 * Makefile.am: Add linux target for S/390.
3292 * configure.in: Likewise.
3293 * config/tc-s390.c: New file.
3294 * config/tc-s390.h: New file.
3298 * config/tc-sh.c (md_pseudo_table): Add uaquad. Use s_uacons for
3299 2byte, 4byte and 8byte.
3303 * config/tc-hppa.c (pa_build_unwind_subspace): Don't call
3304 md_number_to_chars with size > sizeof (valueT).
3308 * config/tc-ia64.h (TC_RELOC_RTSYM_LOC_FIXUP): Do fixup if
3309 there is no relocation.
3313 * config/tc-ia64.h (TC_RELOC_RTSYM_LOC_FIXUP): New. Defined.
3315 * config/tc-ia64.c (md_parse_option): Only accept the valid
3316 ia64 options on "-axxx".
3320 * config/tc-ia64.c (errata_nop_necessary_p): Return 0 instead of
3321 aborting for invalid operands.
3325 * config/tc-hppa.c (fix_new_hppa): Pass in unwind directly rather
3326 than via pointer. Update all callers.
3327 (UNWIND_LOW32): Define.
3328 (UNWIND_HIGH32): Define.
3329 (pa_build_unwind_subspace): Use the above macros instead of dumping
3330 bitfields directly. Call frag_more once rather than multiple times.
3331 (md_assemble): Use UNWIND_LOW32.
3332 (pa_entry): Likewise
3333 (pa_procend): Likewise.
3334 (process_exit): Use UNWIND_HIGH32.
3338 * config/tc-m68hc11.h (LISTING_HEADER): Use m68hc11_listing_header
3339 function to select the header according to the cpu.
3340 (md_after_pass_hook, md_do_align): Remove.
3341 (md_cleanup, m68hc11_cleanup): Remove.
3342 (md_pcrel_from_section): Declare.
3343 * config/tc-m68hc11.c (build_dbranch_insn): Remove insn_size.
3344 (build_jump_insn, build_insn): Likewise.
3345 (m68hc11_listing_header): New function.
3346 (m68hc11_cleanup): Remove.
3350 * config/tc-m68hc11.c (relaxable_symbol): Relax externally visible
3351 symbols because there is no support for shared libraries and these
3352 symbols can't be overridden (unless they are weak).
3356 * dwarf2dbg.c (out_debug_abbrev): Terminate the abbreviations
3357 for the compilation unit with a zero byte.
3361 * config/tc-hppa.c (pa_ip): Support 12 bit branches to absolute
3362 destinations. Correct range check for 17 and 22 bit branches.
3366 * config/tc-m68k.c (tc_gen_reloc): Do not abort if tcbit is
3367 still set. Issue an error message instead.
3368 (md_estimate_size_before_relax): Delete unused variable
3369 'buffer_address'. Fixup parentheses around if statement.
3373 * as.c: Fix formatting.
3374 * ehopt.c: Likewise.
3375 * messages.c: Likewise.
3376 * stabs.c: Likewise.
3377 * symbols.c: Likewise.
3381 * config/tc-m32r.c (m32r_handle_align): Declare type of fragp.
3385 * config/tc-alpha.c: Fix formatting.
3389 * config/tc-alpha.c: Fix formatting.
3393 * config/tc-alpha.c: Fix formatting.
3397 * config/tc-arm.c (ldm_flags): Remove redundant bit from "fa" and
3399 (stm_flags): Remove redundant bit from "ed" and "da" flags.
3403 * configure.in (cpu_type, arch): Match i386 too.
3404 * configure: Rebuilt.
3408 * config/tc-i386.c: Fix formatting.
3412 * config/tc-hppa.c (tc_gen_reloc): Use SEGREL32 instead of DIR32
3413 relocs for .PARISC.unwind section.
3415 * config/tc-hppa.c (pa_build_unwind_subspace): Build unwind
3416 depending on section flags, not just for .text.
3420 * config/tc-ia64.c (ia64_flush_insns): Handle unwind directives
3421 not immediately followed by an instruction.
3425 * config/tc-m68hc11.c: Fix formatting.
3429 * symbols.c (colon): Change 'already defined symbol' from a
3430 fatal error to an ordinary error. There is no reason why this
3431 error should be fatal.
3433 * message.c (as_fatal): Delete output file, if one has been
3438 * config/tc-hppa.h (TARGET_FORMAT): Add hppa-linux variants.
3442 * config/tc-alpha.c: Fix formatting.
3443 * config/tc-arc.c: Likewise.
3444 * config/tc-arc.h: Likewise.
3445 * config/tc-d10v.c: Likewise.
3446 * config/tc-i370.c: Likewise.
3447 * config/tc-i386.c: Likewise.
3448 * config/tc-i960.c: Likewise.
3449 * config/tc-m68k.c: Likewise.
3450 * config/tc-ppc.c: Likewise.
3451 * config/tc-sparc.c: Likewise.
3452 * config/tc-tahoe.c: Likewise.
3453 * config/tc-vax.c: Likewise.
3455 * config/tc-arc.c: Fix formatting.
3457 * config/tc-arc.c: Fix formatting.
3461 * config/tc-hppa.c (pa_build_unwind_subspace): Use SEGREL32 for
3462 both 32 and 64 bit ELF.
3464 * config/tc-hppa.c (pa_ip): Store `a' flag in bit zero of operand
3465 and don't bother storing `m' for "ce" completer. Tidy handling of
3466 'J' and 'K' operands to suit. Handle '<' and '>' operands.
3470 * config/tc-i386.h (TARGET_MACH): New macro.
3471 (i386_mach): Declare.
3472 * config/tc-i386.c (i386_mach): New function.
3476 * doc/as.texinfo: Fix spelling and cross-references.
3478 * doc/c-arm.texi: Fix typos. Say that `;' is a line separator
3479 character for all systems, not just GNU/Linux. Make it explicit
3480 that `-k' doesn't affect code generation, just ELF flags.
3484 * config/tc-i386.c (md_assemble): Check cpu_flags even for nullary
3489 * cgen.c (gas_cgen_finish_insn): Call dwarf2_emit_insn.
3493 * as.c (print_args): Update copyright date to 2001.
3497 * doc/c-arc.texi: New file.
3498 Some sections to be expanded.
3502 * config/tc-i386.c (md_longopts): Recognize "--64" only for ELF.
3503 (md_parse_option): Always accept "--32".
3507 * as.h (TC_ARC): Ensure struc-symbol.h included.
3508 * as.c (dwarf2dbg.h): Include to remove implicit declaration
3510 * struc-symbol.h (SYMBOLS_NEED_BACKPOINTERS): Define.
3511 (TARGET_SYMBOL_FIELDS) added.
3513 * doc/Makefile.am (CPU_DOCS): Added c-arc.texi.
3514 * doc/c-arc.texi: New file.
3515 Some sections to be expanded.
3516 * doc/as.texinfo: Update command-line options.
3517 Removed outdated text for ARC dependant features, instead include
3518 text from above file.
3520 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Define local flag.
3521 (TARGET_SYMBOL_FIELDS): Alias to previous definition.
3522 (targ-cpu.h) header.
3524 * config/tc-arc.c: New updated configuration for
3525 ARC, including selection of core variants, and extensibility of
3526 instructions, registers etc. through directives.
3528 * config/tc-arc.c (arc_extinst): Minor corrections for
3530 (arc_common) Likewise. Make alignment argument optional for local
3531 symbols also, with default of zero.
3535 * config/tc-m68hc11.c (md_estimate_size_before_relax): Fix
3536 STATE_INDEXED_OFFSET when the symbol is undefined (16-bit offset).
3537 (build_indexed_byte): Don't relax indexed byte, use 16-bit offset
3538 and fix_new_exp() instead.
3539 (md_convert_frag): For indexed post byte use the symbol value
3540 rather than the displacement.
3541 (md_relax_table): Fix indexed offset relax.
3545 * config/tc-m68hc11.c (md_estimate_size_before_relax):Don't
3547 (relaxable_symbol): New function.
3551 * config/tc-i386.h (TC_RELOC_GLOBAL_OFFSET_TABLE): Removed, it's
3556 * config/tc-arm.c (arm_fix_adjustable): Define for OBJ_COFF.
3557 * config/tc-arm.h (obj_fix_adjustable): Define for OBJ_COFF
3561 * symbols.c (DOLLAR_LABEL_CHAR): New constant - the magic
3562 character used to dollar local symbols.
3563 (LOCAL_LABEL_CHAR): New constant - the magic character used to
3564 local label symbols.
3565 (dollar_label_name): Use DOLLAR_LABEL_CHAR.
3566 (fb_label_name): Prefix local labels with LOCAL_LABEL_PREFIX,
3568 Use LOCAL_LABEL_CHAR.
3569 (decode_local_label_name): Skip LOCAL_LABEL_PREFIX.
3570 Use DOLLAR_LABEL_CHAR and LOCAL_LABEL_CHAR.
3571 (S_IS_LOCAL): Use DOLLAR_LABEL_CHAR and LOCAL_LABEL_CHAR.
3575 * config/tc-i386.c (i386_immediate, i386_displacement):
3580 * doc/c-i386.texi (i386-Arch): Remove spaces incorrectly inserted
3585 * doc/as.texinfo (Bug Reporting): Update email address for
3591 * configure.in: Define DEFAULT_ARCH for i386.
3592 * config/tc-i386.c (md_assemble): Return after the error message;
3593 move testing for 64bit operands to proper place.
3597 * doc/as.texinfo: Document '#' as comment character for i386 and
3598 x86_64. Add AMD x86-64 into menu of machine dependent information.
3600 * doc/c-i386.texi: Document x86_64 extensions.
3604 * config/tc-i386.c (md_assemble): Handle third byte of the opcode as prefix.
3608 * config/tc-i386.c (cpu_arch): Add Pentium4 and modify sledgehammer entry.
3609 * NEWS: Add note about Pentium4 support.
3613 * config/tc-i387.c (pi, pte, pt): Update.
3614 (type_names): Add new types.
3618 * config/tc-i386.h (CpuK6, CpuAthlon, CpuSledgehammer, CpuMMX, Cpu3dnow,
3619 CpuUnknown): Renumber
3620 (CpuP4, CpuSSE2): New.
3621 (CpuUnknownFlags): Add CpuP4 and CpuSSE2
3625 * config/tc-alpha.c (alpha_force_relocation): Handle vtable
3627 (alpha_fix_adjustable): Likewise.
3628 (md_apply_fix): Likewise.
3632 * listing.c (listing_message): Allocate string only if it is
3635 * configure: Rebuild.
3639 * doc/internals.texi (Relaxing with a table) <after relaxation>:
3640 Point out caveats with generating fixups for the opcode in a frag.
3644 * configure.in: Add support for x86_64 and x86_64-*-linux-gnu*
3649 * listing.c (calc_hex): Print the variable part only if the
3650 fragment type is rs_fill.
3654 * doc/internals.texi (tc_conditional_pseudoop,
3655 TC_LINKRELAX_FIXUP): Fix typos.
3659 * write.c (subsegs_finish): Fix thinko last change -- don't
3660 "optimize" the alignment == 0 case.
3664 * as.h (rs_align_test): New.
3665 * frags.c (NOP_OPCODE): Move default from read.c.
3666 (MAX_MEM_FOR_RS_ALIGN_CODE): New default.
3667 (frag_align_code): New.
3668 * frags.h (frag_align_code): Declare.
3669 * read.c (NOP_OPCODE): Remove.
3670 (do_align): Use frag_align_code.
3671 * write.c (NOP_OPCODE): Remove.
3672 (get_recorded_alignment): New.
3673 (cvt_frag_to_fill): Handle rs_align_test.
3674 (relax_segment): Likewise.
3675 (subsegs_finish): Align last subseg in section to the
3676 section alignment. Use frag_align_code.
3677 * write.h (get_recorded_alignment): Declare.
3678 * config/obj-coff.c (size_section): Handle rs_align_test.
3679 (fill_section, fixup_mdeps): Likewise.
3680 (write_object_file): Use frag_align_code.
3682 * config/tc-alpha.c (alpha_align): Use frag_align_code.
3683 (alpha_handle_align): New.
3684 * config/tc-alpha.h (HANDLE_ALIGN): New.
3685 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
3687 * config/tc-i386.h (md_do_align): Use frag_align_code.
3688 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
3690 * config/tc-ia64.c (ia64_md_do_align): Don't do code alignment.
3691 (ia64_handle_align): New.
3692 * config/tc-ia64.h (HANDLE_ALIGN): New.
3693 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
3695 * config/tc-m32r.c (m32r_do_align): Remove.
3696 (m32r_handle_align): New.
3697 (fill_insn): Use frag_align_code.
3698 * config/tc-m32r.h (md_do_align): Remove.
3699 (HANDLE_ALIGN, MAX_MEM_FOR_RS_ALIGN_CODE): New.
3700 * config/tc-m88k.c, config/tc-m88k.h: Similarly.
3701 * config/tc-mips.c, config/tc-mips.h: Similarly.
3703 * config/tc-sh.c (sh_cons_align): Use rs_align_test.
3704 (sh_handle_align): Likewise. Handle rs_align_code.
3705 (sh_do_align): Remove.
3706 * config/tc-sh.h (md_do_align): Remove.
3707 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
3709 * config/tc-sparc.c (sparc_cons_align): Use rs_align_test.
3710 (sparc_handle_align): Likewise. Handle rs_align_code.
3711 * config/tc-sparc.h (md_do_align): Remove.
3712 (MAX_MEM_FOR_RS_ALIGN_CODE): New.
3716 * config/tc-d10v.c (md_assemble): set prev_seg and prev_subseg
3717 when we assemble the first half of a pair.
3721 * config/tc-i386.c (reloc): Update the macro for non-bfd
3723 (BFD_RELOC_X86_64_GOTPCREL): Set to 0 for non-bfd assembler.
3727 * dwarf2dbg.c (dwarf2_finish): Remove #if BFD_ASSEMBLER.
3731 * config/tc-i386.h (i386_target_format): Define even for ELFs.
3732 (QWORD_MNEM_SUFFIX): New macro.
3733 (CpuK6,CpuAthlon,CpuSledgehammer, Cpu64, CpuNo64, CpuUnknownFlags):
3735 (CpuMMX,CpuSSE,Cpu3dnow, CpuUnknown): Renumber.
3736 (IgnoreSize, DefaultSize, No_?Suf, FWait, IsString, regKludge, IsPrefix,
3738 (Size64, No_qSuf, NoRex64, Rex64): New macros.
3739 (Reg64, Imm32S, Imm64, Disp32S, Disp64): New macros.
3740 (Imm8, Imm8S, Imm16, Imm32, Imm1, BaseIndex, Disp8, Disp16, Disp32,
3741 InOutPortReg,ShiftCount, Control, Debug, Test, FloatReg, FloatAcc,
3742 SReg2, SReg3, Acc, JumpAbsolute, RegMMX, RegXMM, EsSeg, InvMem):
3744 (Reg, WordReg): Add Reg64.
3745 (Imm): Add Imm32S and Imm64.
3747 (Disp): Add Disp64 and Disp32S.
3748 (AnyMem): Add Disp32S.
3749 (RegRex, RegRex64): New macros.
3750 (rex_byte): New type.
3751 * config/tc-i386.c (set_16bit_code_flag): Kill.
3752 (fits_in_unsigned_long, fits_in_signed_long): New functions.
3753 (reloc): New parameter "signed"; support x86_64.
3754 (set_code_flag): New.
3755 (DEFAULT_ARCH): New macro; default to "i386".
3756 (default_arch): New static variable.
3757 (struct _i386_insn): New fields Operand_PCrel; rex.
3758 (flag_16bit_code): Kill; All tests replaced to "flag_code == CODE_64BIT"
3759 (flag_code): New enum and static variable.
3760 (use_rela_relocations): New static variable.
3761 (flag_code_names): New static variable.
3762 (cpu_arch_flags): Default to CpuUnknownFlags|CpuNo64.
3763 (cpu_arch): Add "sledgehammer"; Add CPUAthlon to Athlon and CpuK6 to
3765 (i386_align_code): Return plain "nop" for x86_64.
3766 (mode_from_disp_size): Support Disp32S.
3767 (smallest_imm_type): Support Imm32S and Imm64.
3768 (offset_in_range): Support size of 8.
3769 (set_cpu_arch): Do not clobber to Cpu64/CpuNo64.
3770 (md_pseudo_table): Add "code64"; use set_code_flat.
3771 (md_begin): Emit sane error message on hash failure.
3772 (tc_i386_fix_adjustable): Support x86_64 relocations.
3773 (md_assemble): Support QWORD_MNEM_SUFFIX, REX registers,
3774 instructions supported on particular arch just partially,
3775 output of 64bit immediates, handling of Imm32S and Disp32S type.
3776 (i386_immedaite): Support x86_64 relocations; support 64bit constants.
3777 (i386_displacement): Likewise.
3778 (i386_index_check): Cleanup; support 64bit addresses.
3779 (md_apply_fix3): Support x86_64 relocation and rela.
3780 (md_longopts): Add "32" and "64".
3781 (md_parse_option): Add OPTION_32 and OPTION_64.
3782 (i386_target_format): Call even for ELFs; choose between
3783 elf64-x86-64 and elf32-i386.
3784 (i386_validate_fix): Refuse GOTOFF in 64bit mode.
3785 (tc_gen_reloc): Support rela relocations and x86_64.
3786 (intel_e09_1): Support QWORD.
3790 * config/tc-i386.c (intel_e09_1): Only flag as a memory operand if
3791 it's not an offset expression.
3792 (intel_e10_1): Ditto. Also, if the operand is an offset expression,
3793 keep the braces '[' and ']' in the output string.
3794 (intel_e11): Ditto. Also remove comparison intel_parser.op_modifier
3795 != FLAT. There is no such op_modifier.
3799 * dwarf2dbg.c: If we don't have <limits.h>, try including <sys/param.h>
3804 * as.h: Fix formatting.
3806 * dwarf2dbg.c: Likewise.
3807 * input-scrub.c: Likewise.
3812 * configure.in (i386-*-msdosdjgpp): Set bfd_gas to yes.
3813 configure: Regenerate.
3817 * dwarf2dbg.c: #include <limits.h> only if it exists.
3821 * config/tc-hppa.c (pa_ip): Correct CHECK_FIELD typo.
3822 (md_apply_fix): Here too.
3826 * config/tc-ia64.h (ia64_init): Add prototype.
3830 * dwarf2dbg.c: Enabled only if BFD_ASSEMBLER is defined.
3832 * read.h (outputting_stabs_line_debug): Change it to int.
3833 * stabs.c (outputting_stabs_line_debug): Likewise.
3837 * config/obj-bout.c (obj_crawl_symbol_chain): Don't take
3838 the address of a function result.
3842 * config/tc-ppc.c (md_pseudo_table): Add .file and .loc.
3843 (md_assemble): Call dwarf2_emit_insn.
3844 (shlib): Fix typo SHILB -> SHLIB.
3845 (md_parse_option): Likewise.
3846 (ppc_elf_validate_fix): Likewise:
3847 * config/tc-ppc.h (DWARF2_LINE_MIN_INSN_LENGTH): New.
3851 * cgen.h: Fix formatting.
3852 * input-scrub.c: Fix formatting.
3853 * macro.c: Fix formatting.
3854 * config/tc-mips.c: Fix formatting.
3855 * doc/c-mips.texi: Fix formatting.
3859 * config/tc-i386.c (md_assemble): Refuse 's' and 'l' suffixes in the intel
3860 mode; convert 'd' suffix to 's' or 'l'; remove all DWORD_MNEM_SUFFIX
3862 (intel_e09_1): Convert QWORD to 'l' suffix for FP operations; refuse
3864 * config/tc-i386.h (DWORD_MNEM_SUFFIX): Kill.
3867 * i386.h (*_Suf): Remove No_dSuf.
3868 (d_suf, wld_Suf,sld_Suf, sldx_Suf, bwld_Suf, d_FP, sld_FP, sldx_FP)
3870 (i386_optab): Remove 'd' in the suffixes.
3874 * config/tc-i386.c (T_SHORT): Undefine before defining.
3878 * config/tc-mips.c: Fix formatting.
3882 * config/tc-d10v.c (flag_allow_gstabs_packing): New variable.
3883 (md_longopts): New options --gstabs-packing, --no-gstabs-packing.
3884 (md_show_usage): Ditto.
3885 (md_parse_option): Ditto.
3886 (d10v_cleanup): Writes pending instruction only if
3887 ! outputting_stabs_line_debug || ! flag_allow_gstabs_packing.
3888 Fix compile time warning messages.
3890 * doc/c-d10v.texi: Documents new options.
3894 * stabs.c (outputting_stabs_line_debug): New variable.
3895 (stabs_generate_asm_lineno): Set outputting_stabs_line_debug at
3896 function entry and unset at function exit.
3898 * read.h (outputting_stabs_line_debug): New extern declaration.
3900 * as.c: Include dwarf2dbg.h for definition of dwarf2_finish.
3902 * dwarf2dbg.c: Fix compile time warning messages.
3906 * config/tc-a29k.c: Fix formatting.
3907 * config/tc-alpha.c: Likewise.
3908 * config/tc-arm.c: Likewise.
3909 * config/tc-cris.c: Likewise.
3910 * config/tc-hppa.c: Likewise.
3911 * config/tc-i370.c: Likewise.
3912 * config/tc-i386.c: Likewise.
3913 * config/tc-i860.c: Likewise.
3914 * config/tc-i960.c: Likewise.
3915 * config/tc-ia64.c: Likewise.
3916 * config/tc-m68hc11.c: Likewise.
3917 * config/tc-m68k.c: Likewise.
3918 * config/tc-m88k.c: Likewise.
3919 * config/tc-pj.c: Likewise.
3920 * config/tc-ppc.c: Likewise.
3921 * config/tc-sh.c: Likewise.
3922 * config/tc-sparc.c: Likewise.
3923 * config/tc-tahoe.c: Likewise.
3924 * config/tc-vax.c: Likewise.
3928 * config/tc-mips.c (mips_ip): When calculating offsets,
3929 don't accept as constant the difference between the
3930 addresses of symbols in two different sections.
3932 * config/tc-mips.c (macro_build): Add new 'U' and 'J' operand
3934 (validate_mips_insn): Likewise. Also, update 'B' operand
3935 specifier to use OP_*_CODE20 constants and delete 'm' operand
3937 (mips_ip): Remove 'm' operand specifier, add 'U' and 'J'
3938 operand specifiers. Change warning generated by 'B' operand
3939 specifier to reflect its new multi-purpose usage.
3941 * config/tc-mips.c (mips_set_options): Use ISA_UNKNOWN rather than
3942 -1, and update comment.
3943 (file_mips_isa): Likewise.
3944 (mips_cpu): Use CPU_UNKNOWN rather than -1, and update comment.
3945 (ISA_HAS_COPROC_DELAYS, ISA_HAS_64BIT_REGS, gpr_interlocks): Use
3946 ISA_* constants rather than hard-coded numbers.
3947 (mips_cpu_info): New structure.
3948 (mips_cpu_info_table): New table describing CPU and ISA names
3950 (mips_cpu_info_from_name, mips_cpu_info_from_isa,
3951 mips_cpu_info_from_cpu): New functions.
3952 (mips_isa_to_str): New function to get string for ISA name.
3953 (mips_cpu_to_str): Convert to use mips_cpu_info_from_cpu, and
3954 return const char *.
3955 (md_begin): Redo CPU and ISA selection logic, using
3956 mips_cpu_info_from_*. Convert to use ISA_* constants rather
3957 than hard-coded numbers.
3958 (append_insn, mips_emit_delays, macro, macro2): Convert to use
3959 ISA_* constants rather than hard-coded numbers.
3960 (mips_ip): Convert to use mips_isa_to_str to get ISA name.
3961 (md_longopts): Delete OPTION_NO_MIPS32.
3962 (md_parse_option): Convert to use ISA_* constants rather than
3963 hard-coded numbers. Make OPTIONS_MIPS32 case treat MIPS32
3964 as an ISA. Delete OPTION_NO_MIPS32 case. Convert OPTION_MCPU
3965 to use strcasecmp to recognize "default" and to use
3966 mips_cpu_info_from_name to get CPU numbers from argument.
3967 (md_show_usage): Move -mips32 so it's with the rest of the ISA
3968 flags. Change 4Kc, 4Kp and 4Km CPU entries to just be
3970 (s_mipsset): Accept ISA value 32.
3971 * doc/as.texinfo: Clean up MIPS options summary slightly,
3972 remove -no-mips32. Add note about -mips4 and -mips32
3973 specifying those ISA levels. Delete -mips32 and -no-mips32
3974 cpu flag descriptions.
3975 * doc/c-mips.texi: Add -mips32 to list of ISA switches. Clean
3976 up the supported CPU switch list, and replace 4Kc, 4Km, and
3977 4Kp entries with a single mips32-4k entry. Note that you can
3980 * config/tc-mips.c (ISA_HAS_64BIT_REGS): Add checks for ISA_MIPS5 and
3982 (md_longopts, OPTION_MIPS5, OPTION_MIPS64): Add options for
3984 (md_parse_option): Add cases for OPTION_MIPS5 and
3986 (md_show_usage): Mention -mips5 and -mips64 arguments.
3987 (s_mipsset): Add cases for MIPS5 and MIPS64.
3988 (mips_cpu_info_table): Add entries for MIPS5 and MIPS64 ISAs
3990 * doc/as.texinfo: Mention -mips5 and -mips64 options
3992 * doc/c-mips.texi: Likewise. Also update introduction
3993 and ".set" usage information.
3995 * config/tc-mips.c (md_show_usage): Add "sb1" to the
3997 (mips_cpu_info_table): Add SB-1 entries.
3998 * doc/c-mips.texi: Add "sb1" to the list of CPUs
3999 known to the -mcpu option.
4001 * doc/as.texinfo: Correct description of MIPS -mcpu
4002 option, by copying some of the text from doc/c-mips.texi.
4006 * configure.in (arm-*-rtems*, a29k-*rtems*, h8300-*-rtems*):
4008 (sparc*-*-rtemself*, sparc*-*-rtemsaout*): New targets.
4009 (sparc*-*-rtems*): Switched from a.out to ELF.
4010 * configure: Regenerate.
4011 * config.in: Regenerate.
4012 * Makefile.in: Regenerate.
4013 * aclocal.m4: Regenerate.
4014 * po/gas.pot: Regenerate.
4018 * config/obj-coff.c (obj_coff_weak): Use S_SET_WEAK if it exists,
4019 even in non BFD_ASSEMBLER case.
4023 * config/tc-i386.c (md_assemble): Swap i.disp_relocs when using intel
4028 * dwarf2dbg.c: Rewrite from scratch. Queue all debugging output
4029 until dwarf2_finish; use relaxation to get cross-fragment offsets;
4030 thread multiple subsegments properly; handle multiple code
4031 sections properly; emit proper compilation unit info for assembler
4032 generated debugging.
4034 * as.h (enum _relax_state): Add rs_dwarf2dbg.
4035 * dwarf2dbg.h (struct dwarf2_line_info): Remove filename.
4036 (dwarf2dbg_estimate_size_before_relax): Declare.
4037 (dwarf2dbg_relax_frag, dwarf2dbg_convert_frag): Declare.
4038 * write.c: Include dwarf2dbg.h.
4039 (cvt_frag_to_fill): Handle rs_dwarf2dbg.
4040 (relax_segment): Likewise.
4044 * config/tc-sh.c (md_convert_frag) <undefined symbol, conditional
4045 jump>: Use as_bad_where instead of as_bad. Tweak error message
4046 accordingly. Stabilize frag by updating fix part and resetting
4048 <undefined symbol, unconditional jump>: Ditto.
4049 (sh_elf_cons): Cast *input_line_pointer to unsigned char when
4050 indexing is_end_of_line[].
4051 (md_assemble): Initialize size to 0.
4052 (md_section_align): Mark parameter seg as unused.
4053 (parse_reg): Parse names case-insensitively.
4057 * config/obj-aout.h: Fix formatting.
4058 * config/obj-bout.h: Likewise.
4059 * config/obj-coff.c: Likewise.
4060 * config/obj-coff.h: Likewise.
4061 * config/obj-elf.h: Likewise.
4062 * config/obj-som.h: Likewise.
4063 * config/obj-vms.c: Likewise.
4064 * config/obj-vms.h: Likewise.
4065 * config/tc-h8300.h: Likewise.
4066 * config/tc-ns32k.h: Likewise.
4067 * config/tc-sparc.h: Likewise.
4068 * config/tc-tic54x.h: Likewise.
4069 * config/tc-z8k.h: Likewise.
4073 * doc/as.1 (COPYING): Mention that the GNU Free Documentation
4074 License is present in the sources, but not the output, and
4075 also available from the GNU website.
4076 (GNU Free Documentation License): Comment out this section.
4080 * Makefile.am (CPU_OBJ_VALID): Add case to filter out invalid coff
4081 targets. Remove i860 from valid a.out targets.
4082 * Makefile.in: Regenerate.
4084 * config/tc-cris.c: Include dwarf2dbg.h.
4085 (md_pseudo_table): Add .file and .loc.
4086 (md_assemble): Call dwarf2_emit_insn if generating ELF.
4087 (s_cris_file, s_cris_loc): New.
4088 * config/tc-cris.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
4089 * Makefile.am: Regenerate dependencies.
4090 * Makefile.in: Regenerate.
4094 * expr.c (STANDARD_MUL_PRECEDENCE): Correct value.
4095 (MRI_MUL_PRECEDENCE): Likewise.
4096 (op_rank): Fix a comment typo.
4100 * config/tc-m68hc11.c (build_indexed_byte): Print the offset in
4102 (get_operand): Fix analysis for movw/movb instructions.
4106 * configure.in (xscale-elf): Add target.
4107 (xscale-coff): Add target.
4108 * configure: Regenerate.
4110 * config/tc-arm.c (ARM_EXT_V5E): New ARM architecture
4112 (ARM_EXT_XSCALE): New ARM architecture extension.
4113 (ARM_LONGMUL): Rename to ARM_EXT_LONGMUL.
4114 (ARM_HALFWORD): Rename to ARM_EXT_HALFWORD.
4115 (ARM_THUMB): Rename to ARM_EXT_THUMB.
4116 (ARM_ARCH_V4): Remove processor from architecture.
4117 (ARM_ARCH_3M): New architecutre definition.
4118 (ARM_ARCH_V5TE): New architecutre definition.
4119 (ARM_ARCH_XSCALE): New architecutre definition.
4120 (CPU_DEFAULT): Allow to be defaulted to XScale.
4121 (atpcs): New boolean variable.
4122 (ldr_flags): Support 'd' flag for double word loads.
4123 (str_flags): Support 'd' flag for double word stored.
4124 (do_mia): New function.
4125 (do_mar): New function.
4126 (do_mra): New function.
4127 (do_pld): New function.
4128 (do_ldrd): New function.
4129 (do_blx): New function.
4130 (do_bkpt): New function.
4131 (do_clz): New function.
4132 (do_lstc2): New function.
4133 (do_cdp2): New function.
4134 (do_t_blx): New function.
4135 (do_t_bkpt): New function.
4136 (do_smla): New function.
4137 (do_smlal): New function.
4138 (do_smul): New function.
4139 (do_qadd): New function.
4140 (do_co_reg2c): New function.
4141 (LONGEST_INSN): Redefine to 7.
4143 * doc/c-arm.texi: Document -mxscale, -mmarmv5te and -matpcs
4144 command line switches.
4148 * config/tc-ia64.c (pseudo_func): Add missing initializers.
4149 (struct rsrc): Make line unsigned.
4150 (gr_values): Add missing initializer.
4151 (SLOT_NUM_NOT_SET): Add unsigned cast.
4152 (ia64_elf_section_flags, output_vbyte_mem, count_output, dot_radix,
4153 dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
4154 dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
4155 dot_unwentry, dot_altrp, dot_saveg, dot_savef, dot_saveb, dot_savegf,
4156 dot_spill, dot_spillreg, dot_spillreg_p, dot_label_state,
4157 dot_copy_state, dot_unwabi, dot_personality, dot_proc, dot_body,
4158 dot_prologue, dot_endp, dot_regstk, dot_psr, dot_alias, dot_ln,
4159 dot_reg_val, dot_entry, dot_mem_offset, ia64_init, mark_resource,
4160 md_undefined_symbol, md_apply_fix3, tc_gen_reloc, ia64_md_do_align):
4161 Add ATTRIBUTE_UNUSED to unused parameters.
4162 (convert_expr_to_ab_reg): Add parens.
4163 (convert_expr_to_xy_reg): Add parens. Comment out >= REG_GR test.
4164 (dot_prologue): Initialize grsave when declared.
4165 (md_pseudo_table): Add missing initializers.
4166 (operand_match): Add casts to bfd_vma.
4167 (emit_one_bundle): Delete unused local prev. Make required_template
4169 (specify_resource): Cast i to unsigned.
4170 (note_register_values): Use fprintf_vma.
4171 (print_dependency): Likewise.
4175 * config/tc-ia64.c (generate_unwind_image): Call record_alignment
4176 for unwind info section.
4177 (dot_endp): Likewise for unwind section.
4179 * config/tc-ia64.c (emit_one_bundle): Pass size of 8 not 4 to
4184 * config/tc-sparc.c (md_pseudo_table): Add .file and .loc.
4185 (output_insn): Call dwarf2_emit_insn.
4186 * config/tc-sparc.h (DWARF2_LINE_MIN_INSN_LENGTH): New.
4190 * ehopt.c (eh_frame_code_alignment): New arg `in_seg', update all
4191 callers. Don't switch segments. Expect CIE == -1 in .debug_frame.
4192 (check_eh_frame): Handle .eh_frame and .debug_frame concurrently.
4196 * config/tc-arm.c (md_pseudo_table): Add support for .line and
4201 * config/tc-i386.c (md_pseudo_table): Add .file and .loc.
4205 * dwarf2dbg.c (dwarf2_gen_line_info): Early out for no line number.
4206 * config/obj-elf.h (ECOFF_DEBUGGING) [TC_ALPHA]: Adjust for
4207 tri-state definition of alpha_flag_mdebug.
4208 * config/tc-alpha.c (alpha_flag_mdebug): Init to -1.
4209 (s_alpha_file): Store first .file directive.
4210 (s_alpha_stab): New.
4211 (md_pseudo_table): Add stabs and stabn.
4215 * config/tc-i386.c (md_assemble): Call dwarf2_emit_insn.
4219 * as.c (debug_type): Init to DEBUG_UNSPECIFIED.
4220 (main): Call dwarf2_finish.
4221 * as.h (debug_type): Clarify documentation of the meaning
4223 * dwarf2dbg.c (DWARF2_LINE_MIN_INSN_LENGTH): Default to 1.
4224 (print_stats): Fix parenthesis problem.
4225 (now_subseg_size): New.
4226 (dwarf2_finish): Use it. If DEBUG_DWARF2, emit bits for .debug_info.
4227 (dwarf2_directive_file): Don't set debug_type.
4228 (dwarf2_where): Honor DEBUG_DWARF2 first.
4229 (dwarf2_emit_insn): Renamed from dwarf2_generate_asm_lineno;
4230 do nothing if not emitting dwarf2 debug info, or no work.
4231 * dwarf2dbg.h (dwarf2_emit_insn): Update.
4232 * ecoff.c (add_file): Turn on DEBUG_ECOFF only if DEBUG_UNSPECIFIED.
4233 (ecoff_new_file): Likewise.
4234 * read.c (generate_lineno_debug): Kill ecoff hackery. Update
4235 commentary wrt dwarf2.
4237 * config/tc-alpha.c (alpha_adjust_symtab_relocs): Add
4238 ATTRIBUTE_UNUSED as needed.
4239 (emit_insn): Call dwarf2_emit_insn.
4240 (s_alpha_file): New.
4242 (s_alpha_coff_wrapper): Don't handle them.
4243 (md_pseudo_table): Update for .file and .loc.
4244 * config/tc-alpha.h (DWARF2_LINE_MIN_INSN_LENGTH): New.
4246 * config/tc-arm.c (output_inst): Update for dwarf2_emit_insn;
4247 don't protect with debug_type.
4248 * config/tc-hppa.c (md_assemble): Likewise.
4249 * config/tc-m68hc11.c (m68hc11_new_insn): Likewise.
4250 * config/tc-mn10300.c (md_assemble): Likewise.
4251 * config/tc-sh.c (md_assemble): Likewise.
4252 * config/tc-v850.c (md_assemble): Likewise.
4254 * config/tc-arm.c (arm_end_of_source): Remove.
4255 * config/tc-hppa.c (pa_end_of_source): Remove.
4256 * config/tc-m68hc11.c (m68hc11_end_of_source): Remove.
4257 * config/tc-mn10300.c (mn10300_finalize): Remove.
4258 * config/tc-sh.c (sh_finalize): Remove.
4259 * config/tc-v850.c (sh_finalize): Remove.
4261 * config/tc-arm.h (md_end): Remove.
4262 * config/tc-hppa.h (md_end): Remove.
4263 (DWARF2_LINE_MIN_INSN_LENGTH): New.
4264 * config/tc-m68hc11.h (md_end): Remove.
4265 * config/tc-mn10300.h (md_end): Remove.
4266 * config/tc-sh.h (md_end): Remove.
4267 * config/tc-v850.h (md_end): Remove.
4269 * config/tc-ia64.c (emit_one_bundle): Don't protect
4270 dwarf2 bits with debug_type.
4271 (md_assemble): Likewise.
4272 (ia64_end_of_source): Don't call dwarf2_finish.
4276 * config/tc-ia64.c (errata_nop_necessary_p): Abort if general regno
4277 >= 128 instead of > 128. Abort if predicate regno is >= 64 instead of
4282 * config/obj-elf.c (obj_elf_symver): Don't check the missing
4287 * config/tc-tic30.c: Fix formatting.
4288 * config/tc-tic80.c: Likewise.
4289 * config/tc-v850.c: Likewise.
4290 * config/tc-vax.c: Likewise.
4291 * config/tc-w65.c: Likewise.
4292 * config/tc-z8k.c: Likewise.
4296 * config/tc-v850.c: Support dwarf2.
4297 * config/tc-v850.h: Ditto.
4299 * config/tc-v850.c (cons_fix_new_v850): Don't rely on
4300 parse_cons_expression_v850 to initialize hold_cons_reloc.
4304 * config/tc-ia64.c (struct md): New entries LAST_GROUPS, GROUP_IDX.
4305 (errata_nops_necessary_p): New function.
4306 (emit_one_bundle): Call it. Update the GROUP_IDX field in struct
4311 * config/tc-ia64.c (ia64_target_format): If EF_IA_64_BE not set, then
4312 return little endian bfd formats.
4316 * config/aout_gnu.h: Fix formatting.
4317 * config/atof-vax.c: Likewise.
4318 * config/m68k-parse.h: Likewise.
4319 * config/m88k-opcode.h: Likewise.
4320 * config/obj-elf.c: Likewise.
4321 * config/tc-arm.c: Likewise.
4322 * config/tc-cris.c: Likewise.
4323 * config/tc-i386.c: Likewise.
4324 * config/tc-ia64.c: Likewise.
4325 * config/tc-mn10300.c: Likewise.
4326 * config/te-386bsd.h: Likewise.
4327 * config/te-hppa.h: Likewise.
4328 * config/te-nbsd.h: Likewise.
4329 * config/te-ppcnw.h: Likewise.
4330 * config/te-sparcaout.h: Likewise.
4331 * config/te-tmips.h: Likewise.
4332 * config/vax-inst.h: Likewise.
4333 * config/vms-conf.h: Likewise.
4337 * config/tc-alpha.c (s_alpha_prologue): Preserve visibility bits.
4341 * config/obj-elf.c (elf_frob_symbol): Support
4342 ".symver name,name2@@@nodename".
4343 (elf_frob_file_before_adjust): Likewise.
4345 * doc/as.texinfo: Updated for ".symver name,name2@@@nodename"
4346 and ".symver name,name2@@@nodename".
4351 * config/obj-elf.c (obj_elf_symver): Check missing version
4356 * dwarf2dbg.c (dwarf2_generate_asm_lineno): Use addressT
4357 instead of bfd_vma for non-bfd assemblers.
4361 * itbl-ops.c: Fix comment typos.
4365 * config/tc-ia64.c (struct unw_rec_list): Add slot_frag field.
4366 (struct unwind): Add next_slot_frag field.
4367 (slot_index): New parameters slot_frag and first_frag. Add code
4368 to add in frag sizes when different. Add comments.
4369 (fixup_unw_records): New locals first_frag and last_frag. Pass new
4370 arguments to slot_index.
4371 (emit_one_bundle): Set slot_frag field. Set next_slot_number after
4372 loop end. Set next_slot_frag field.
4376 * doc/as.texinfo (.symver): Updated for versioned symbol
4379 * obj.h (format_ops): Add the frob_file_before_adjust field.
4381 * config/obj-aout.c (aout_format_ops): Set the
4382 frob_file_before_adjust field to 0.
4383 * config/obj-coff.c (coff_format_ops): Likewise.
4384 * config/obj-ecoff.c (ecoff_format_ops): Likewise.
4386 * config/obj-elf.c (obj_elf_symver): Allow duplicated version
4388 (elf_frob_file_before_adjust): New function to remove unneeded
4389 versioned symbols from the symbol table.
4390 (elf_format_ops): Set the frob_file_before_adjust field to
4391 elf_frob_file_before_adjust.
4393 * config/obj-elf.h (obj_frob_file_before_adjust): Defined if
4396 * config/obj-multi.h (obj_frob_file_before_adjust): Defined.
4400 * config/tc-arc.h: Avoid warnings for LITTLE_ENDIAN and
4402 * config/tc-arc.c: Use S_IS_LOCAL to test local symbols.
4403 Fix compile time warning messages.
4407 * stabs.c (generate_asm_file): Increase length of xmalloc'ed
4408 buffer in order to avoid buffer overflows.
4412 * config/tc-ia64.c (md_shortopts, md_parse_option, md_show_usage):
4413 Change M to m for -milp32 or -mlp64 to match gcc.
4414 (dot_endp): Use bytes_per_address instead of 8.
4415 (emit_one_bundle): Use number_to_chars_littleendian instead of
4417 (fix_insn): Likewise.
4418 (ia64_init): New function.
4419 (ia64_target_format): New function.
4420 (md_begin): Set endianness, arch, and machine as appropriate.
4421 * config/tc-ia64.h: (TARGET_BYTES_BIG_ENDIAN, md_number_to_chars):
4422 Make these macros depend on TE_HPUX macro.
4423 (TARGET_FORMAT): Define.
4424 (HOST_SPECIAL_INIT): Define.
4425 * config/te-hpux.h: New file.
4426 * configure.in: Add "ia64-*-hpux*" target to configure.
4427 * configure: Regenerate.
4431 * as.c: Fix formatting.
4432 * dwarf2dbg.c: Likewise.
4433 * input-file.c: Likewise.
4434 * input-file.h: Likewise.
4435 * input-scrub.c: Likewise.
4436 * itbl-ops.c: Likewise.
4437 * listing.c: Likewise.
4438 * macro.h: Likewise.
4439 * messages.c: Likewise.
4441 * subsegs.c: Likewise.
4442 * subsegs.h: Likewise.
4443 * write.c: Likewise.
4447 * doc/as.texinfo: Add GNU Free Documentation License.
4448 * doc/gasp.texi: Add GNU Free Documentation License.
4449 * doc/as.1: Add GNU Free Documentation License.
4453 * config/tc-arm.c: Add include of "dwarf2dbg.h"
4457 * configure.in: Recognise i[3456]86-chaosdev-storm-chaos.
4458 * configure: Regenerate.
4462 * read.c (original_case_string): New global variable.
4463 (read_a_source_file): Copy opcode string into
4464 original_case_string if clobbering the case of the opcode.
4465 * read.h: Export the definition of original_case_string.
4466 * config/tc-arm.c (md_assembler): When parsing a .req
4467 directive use the original opcode string, not the case
4472 * config/tc-mn10300.c (debug_line): Remove this static
4474 (md_assemble): Call dwarf2_generate_asm_lineno instead of
4475 dwarf2_where and dwarf2_gen_line_info.
4479 * config/tc-arm.c (do_msr): Improve error message.
4483 * config/tc-mn10300.c (md_apply_fix3): Use valuep if fully resolved
4484 or pc-relative, else use fx_offset.
4488 * config/tc-ia64.c (struct md): New field tag_fixups.
4489 (ia64_flush_insns): Handle tag_fixups. Error if dangling
4490 qualifying predicate.
4491 (emit_one_bundle): Delete spurious multiplication by one. Handle
4493 (ia64_start_line): Error if dangling qualifying predicate.
4494 (defining_tag): New static variable.
4495 (ia64_unrecognized_line, case '['): Parse tags.
4496 (ia64_frob_label): Create tag_fixups.
4497 (md_assemble): Reset md.qp.X_op after using it.
4501 * config/tc-sh.c (md_apply_fix [BFD_RELOC_SH_PCDISP12BY2]): Allow 4094.
4505 * config/tc-ia64.c (extra_goodness): Only prefer F in slot 1 and B in slot 2.
4509 * expr.c: Fix formatting.
4510 * flonum-copy.c: Likewise.
4511 * flonum.h: Likewise.
4517 * as.h (OPTION_MD_BASE): Bump to 190.
4518 * as.c (parse_args) <std_longopts>: Add comment about the need to
4519 check OPTION_MD_BASE in as.h.
4521 * config/tc-sh.c (md_apply_fix): For ELF, do not "adjust back" VAL
4526 * configure.in (emulations): Add m68hc12.
4527 * configure: Regenerate.
4528 * po/gas.pot: Regenerate.
4532 * config/tc-arm.c (psrs): Remove lowercase versions of spsr* and
4534 (arm_psr_parse): Handle lowercase CPSR and SPSR.
4538 * dwarf2out.c (dwarf2_generate_asm_lineno): New function: Generate
4539 a DWARF2 line number information sequence.
4541 *dwarf2out.h: Add prototype for dwarf2_generate_asm_lineno.
4543 * read.c (generate_lineno_debug): Update comment describing why
4544 DWARF2 line number debug information is not generated
4545 automatically by this function.
4547 * doc/as.texinfo: Note that --gdwarf2 only works on some targets,
4550 * config/tc-arm.h (md_end): Define.
4551 (DWARF2_LINE_MIN_INSN_LENGTH): Define.
4553 * config/tc-arm.c (output_inst): Call dwarf2_generate_asm_lineno
4554 if generating DWARF2 line numbers.
4555 (arm_end_of_source): New function. Call dwarf2_finish if
4558 * config/tc-hppa.c (md_assemble): Use dwarf2_generate_asm_lineno.
4559 * config/tc-m68hc11.c (m68hc11_new_insn): Use dwarf2_generate_asm_lineno.
4560 * config/tc-sh.c (md_assemble): Use dwarf2_generate_asm_lineno.
4564 * config/tc-i386.c: Fix prototype declarations for functions taking no
4569 * config/tc-i386.c (i386_operand_modifier): Remove.
4570 (build_displacement_string): Remove.
4571 (i386_parse_seg): Remove.
4572 (i386_intel_memory_operand): Remove.
4573 (i386_intel_operand): Re-write using recursive descent parser based
4574 on MASM documentation.
4575 (struct intel_parser_s): New structure.
4576 (intel_parser): New static variable.
4577 (struct intel_token): New structure.
4578 (cur_token, prev_token): New static variables.
4591 (intel_match_token): New function.
4592 (intel_get_token): New function.
4593 (intel_putback_token): New function.
4594 (intel_expr): New function.
4595 (intel_e05): New function.
4596 (intel_e05_1): New function.
4597 (intel_e06): New function.
4598 (intel_e06_1): New function.
4599 (intel_e09): New function.
4600 (intel_e09_1): New function.
4601 (intel_e10): New function.
4602 (intel_e10_1): New function.
4603 (intel_e11): New function.
4607 * config/tc-sparc.c (sparc_ip): Fix a bug which caused v9_arg_p
4608 instructions to loose any special insn->architecture mask.
4610 * config/tc-sparc.c (v9a_asr_table): Add v9b ASRs.
4611 (sparc_md_end, sparc_arch_types, sparc_arch,
4612 sparc_elf_final_processing): Handle v8plusb and v9b architectures.
4613 (sparc_ip): Handle siam mode operands. Support v9b ASRs (and
4614 request v9b architecture if they are used).
4618 * config/tc-m68k.c: Fix the previous misapplied patch.
4622 * config/tc-m68k.h (RELAX_RELOC_*): New definitions for both
4623 BFD_ASSEMBLER and !BFD_ASSEMBLER.
4624 * config/tc-m68k.c (md_convert_frag_1): Use them instead of
4629 * debug.c: Fix formatting.
4630 * depend.c: Likewise.
4631 * dwarf2dbg.c: Likewise.
4632 * dwarf2dbg.h: Likewise.
4633 * ecoff.c: Likewise.
4636 * flonum-konst.c: Likewise.
4637 * frags.h: Likewise.
4641 * as.c: New option OPTION_TARGET_HELP. Prints all target specific
4643 * doc/as.texinfo: Added notes about this new option.
4647 * config/tc-sh.c (JREG): Remove.
4648 (md_convert_frag): Remove #if 0:d code using JREG.
4652 * config/tc-i386.c (i386_operand_modifier): Only match
4653 modifiers SHORT and FLAT if they are followed by a space.
4654 (parse_register): When `allow_naked_reg' is set, do not confuse
4655 identifiers that start with a register name with a register.
4659 * app.c: Fix formatting.
4662 * bit_fix.h: Likewise.
4669 * config/obj-elf.c (elf_frob_symbol): Revert 2000-10-07 change.
4673 * config/tc-hppa.c (md_apply_fix): Remove plainly wrong assert.
4674 Re-arrange function a little and improve error message.
4676 * write.c (write_relocs): Fix a comment.
4678 * config/obj-elf.c (elf_frob_symbol): Make section syms global on
4683 * config/tc-ia64.c (resources_match): Handle IA64_RS_PRr.
4687 * config/tc-i386.c: Delete some useless comments, reformat others.
4689 * config/tc-i386.h (TC_FIX_ADJUSTABLE): Add check to cover
4690 non-global syms in linkonce sections.
4694 * config/tc-ia64.c (operand_match): Don't use // style comments.
4695 * config/tc-i370.c: Likewise.
4699 Changes to handle varying register prefix and user symbol prefix.
4700 * config/tc-cris.c (SYNTAX_RELAX_REG_PREFIX,
4701 SYNTAX_ENFORCE_REG_PREFIX, SYNTAX_USER_SYM_LEADING_UNDERSCORE,
4702 SYNTAX_USER_SYM_NO_LEADING_UNDERSCORE, REGISTER_PREFIX_CHAR): New.
4703 (s_syntax, cris_force_reg_prefix, cris_relax_reg_prefix,
4704 cris_sym_leading_underscore, cris_sym_no_leading_underscore): New.
4705 (demand_register_prefix): New variable.
4706 (md_pseudo_table): New pseudo ".syntax".
4707 (md_longopts): New options --no-underscore and --underscore.
4708 (cris_target_format): Return elf32-us-cris or elf32-cris depending
4709 on symbols_have_leading_underscore.
4710 (get_gen_reg): Accept or require REGISTER_PREFIX_CHAR.
4711 (get_spec_reg): Ditto.
4712 (cris_number_to_imm) <case BFD_RELOC_VTABLE_ENTRY>: Remove FIXME.
4714 (md_parse_option) <case 'h' 'H'>: Deprecate; add reference to
4716 <case OPTION_NO_US, case OPTION_US>: New.
4717 (md_show_usage): Be brief and reformat to match continuation of
4719 * po/gas.pot: Regenerate.
4723 * config/tc-hppa.c (hppa_force_relocation): If OBJ_SOM, don't
4724 force relocs for 12 bit branches.
4725 (md_apply_fix): Similarly, adjust logic here.
4729 * config/tc-hppa.c (md_apply_fix): Add fmt assertion. Don't
4730 adjust for external and weak syms as we will use a reloc. Allow
4731 for +8 offset when calculating limits of branches.
4732 (hppa_fix_adjustable): Undo 2000-09-23 change.
4733 (hppa_force_relocation): Likewise. Add fx_addsy assertion.
4734 Correct distance calculation.
4735 (tc_gen_reloc): Print the file name and line number if we can't
4739 * config/tc-hppa.c (nonzero_dibits): Define.
4740 (arg_reloc_stub_needed): Check each arg and return value
4741 separately for zero case.
4742 (pa_align): Declare argument `bytes'.
4746 * config/tc-cris.c: Fix formatting.
4747 * config/tc-d10v.h: Likewise.
4748 * config/tc-d30v.c: Likewise.
4749 * config/tc-d30v.h: Likewise.
4750 * config/tc-fr30.c: Likewise.
4751 * config/tc-fr30.h: Likewise.
4752 * config/tc-m68k.c: Likewise.
4753 * config/tc-m68k.h: Likewise.
4754 * config/tc-pj.h: Likewise.
4755 * config/tc-ppc.c: Likewise.
4756 * config/tc-ppc.h: Likewise.
4757 * config/tc-sh.c: Likewise.
4758 * config/tc-sh.h: Likewise.
4759 * config/tc-sparc.c: Likewise.
4760 * config/tc-v850.h: Likewise.
4761 * config/tc-vax.h: Likewise.
4762 * config/tc-w65.h: Likewise.
4763 * config/tc-z8k.h: Likewise.
4767 * config/tc-hppa.c (hppa_fix_adjustable): Do the external and weak
4768 checks only for ELF.
4769 (hppa_force_relocation): Likewise.
4773 * config/tc-ia64.c (dv_sem): Add "stop".
4774 (specify_resource, case IA64_RS_PR): Only handles regs 1 to 15 now.
4775 (specify_resource, case IA64_RS_PRr): New for regs 16 to 62.
4776 (specify_resource, case IA64_RS_PR63): Reorder (note == 7) test to
4778 (mark_resources): Check IA64_RS_PRr.
4782 * config/tc-m68k.c (md_relax_table, m68k_ip, md_convert_frag_1,
4783 md_estimate_size_before_relax): Redesign and clean up the
4784 relaxation mechanism.
4788 * config/tc-ns32k.c: Fix formatting.
4789 * config/tc-ns32k.h: Likewise.
4793 * config/tc-m32r.c: Fix formatting.
4794 * config/tc-m32r.h: Likewise.
4795 * config/tc-m68851.h: Likewise.
4796 * config/tc-m68hc11.c: Likewise.
4797 * config/tc-m68hc11.h: Likewise.
4798 * config/tc-m88k.c: Likewise.
4799 * config/tc-mcore.c: Likewise.
4800 * config/tc-mcore.h: Likewise.
4801 * config/tc-mips.c: Likewise.
4802 * config/tc-mips.h: Likewise.
4803 * config/tc-mn10200.h: Likewise.
4804 * config/tc-mn10300.h: Likewise.
4805 * config/tc-tahoe.c: Likewise.
4806 * config/tc-tahoe.h: Likewise.
4810 * config/tc-vax.c (synthetic_votstrs): Remove jbssi and jbcci.
4811 Likewise in relaxation description comments.
4815 * config/tc-hppa.h (TC_FORCE_RELOCATION_SECTION): Allow
4816 subtraction of two syms without emitting a relocation.
4819 * config/tc-hppa.c (hppa_force_relocation): Force relocations for
4820 global or weak symbols.
4824 * config/tc-h8300.h: Fix formatting.
4825 * config/tc-h8500.c: Likewise.
4826 * config/tc-h8500.h: Likewise.
4827 * config/tc-hppa.h: Likewise.
4828 * config/tc-i370.h: Likewise.
4829 * config/tc-i386.h: Likewise.
4830 * config/tc-i860.c: Likewise.
4831 * config/tc-i860.h: Likewise.
4832 * config/tc-i960.h: Likewise.
4833 * config/tc-ia64.c: Likewise.
4834 * config/tc-ia64.h: Likewise.
4838 * config/tc-a29k.c: Fix formatting.
4839 * config/tc-alpha.c: Likewise.
4840 * config/tc-arc.c: Likewise.
4841 * config/tc-arc.h: Likewise.
4842 * config/tc-arm.c: Likewise.
4843 * config/tc-arm.h: Likewise.
4844 * config/tc-avr.c: Likewise.
4845 * config/tc-avr.h: Likewise.
4846 * config/tc-tic30.c: Likewise.
4847 * config/tc-tic30.h: Likewise.
4848 * config/tc-tic54x.c: Likewise.
4849 * config/tc-tic54x.h: Likewise.
4850 * config/tc-tic80.c: Likewise.
4851 * config/tc-tic80.h: Likewise.
4855 * config/tc-ia64.c (specify_resource): For PR%/PR63, note types of
4856 parallel comparisons for later use.
4857 (struct rsrc): Add parallel comparison type.
4858 (resources_match): Skip special cases of PR usage (non-conflicting
4863 * config/obj-ecoff.c: Fix formatting.
4864 * config/obj-elf.c: Likewise.
4865 * config/obj-elf.h: Likewise.
4866 * config/obj-evax.h: Likewise.
4867 * config/obj-generic.h: Likewise.
4868 * config/obj-hp300.c: Likewise.
4869 * config/obj-hp300.h: Likewise.
4870 * config/obj-ieee.h: Likewise.
4871 * config/obj-vms.c: Likewise.
4872 * config/obj-vms.h: Likewise.
4876 * config/tc-mips.c (md_begin): Recognize 4Kc, 4Km and 4Kp processors.
4877 (md_parse_option): Ditto.
4878 (md_longopts): Add -mips32 option.
4879 (md_show_usage): Document new options.
4880 (mips_ip): Assemble sdbbp 20 bit 'm' args for MIPS32.
4881 (mips_ip): Assemble mfc0 with a sub-selection code.
4882 (validate_mips_insn): Handle 'H' (OP_*_SEL) and 'm' (OP_*_CODE20).
4883 (mips_cpu_to_str): New function.
4884 (mips_ip): Use mips_cpu_to_str instead of printing numeric cpu value.
4885 Use CPU_* defines instead of hardcoded numbers.
4887 * doc/as.texinfo: Document new options.
4888 * doc/c-mips.texi: Ditto.
4892 * as.h: Fix formatting.
4893 * asintl.h: Likewise.
4894 * bit_fix.h: Likewise.
4895 * config/obj-aout.c: Likewise.
4896 * config/obj-aout.h: Likewise.
4897 * config/obj-bout.c: Likewise.
4898 * config/obj-bout.h: Likewise.
4899 * config/obj-coff.c: Likewise.
4900 * config/obj-coff.h: Likewise.
4901 * dwarf2dbg.h: Likewise.
4903 * flonum.h: Likewise.
4904 * frags.h: Likewise.
4905 * itbl-ops.h: Likewise.
4906 * macro.h: Likewise.
4909 * struc-symbol.h: Likewise.
4910 * subsegs.h: Likewise.
4911 * symbols.h: Likewise.
4913 * write.h: Likewise.
4917 * bignum-copy.c: Fix formatting.
4918 * config/tc-i370.c: Likewise.
4919 * config/tc-i960.c: Likewise.
4920 * config/tc-m68k.c: Likewise.
4921 * ehopt.c: Likewise.
4922 * flonum-copy.c: Likewise.
4923 * flonum-konst.c: Likewise.
4924 * flonum-mult.c: Likewise.
4925 * literal.c: Likewise.
4928 * stabs.c: Likewise.
4929 * subsegs.c: Likewise.
4933 * configure.in (arm*-*-uclinux*): New target.
4934 * configure: Regenerate.
4938 * input-file.c: Fix formatting.
4939 * itbl-ops.c: Likewise.
4940 * messages.c: Likewise.
4944 * config/tc-arm.c (md_apply_fix3): Correct handling of ADRL when
4949 * configure.in (AC_ISC_POSIX): Put after AC_CANONICAL_SYSTEM.
4950 * configure: Rebuild.
4954 * atof-generic.c: Fix formatting.
4955 * config/tc-mips.c: Likewise.
4956 * config/tc-vax.c: Likewise.
4957 * input-scrub.c: Likewise.
4961 * config/tc-sh.h (TARGET_FORMAT): Use sh-linux targets.
4962 * configure.in (sh-*-linux*): Added.
4963 * configure: Rebuilt.
4967 * config/tc-hppa.c: Fix formatting.
4969 * ecoff.c: Fix formatting.
4973 * configure: Rebuilt with new libtool.m4.
4977 * cgen.c: Fix formatting.
4978 * config/tc-ia64.c: Likewise.
4982 * aclocal.m4: Regenerate.
4983 * config.in: Regenerate.
4984 * configure: Regenerate.
4985 * po/gas.pot: Regenerate.
4989 * config/tc-cris.c: Correct comment typos.
4993 * config/tc-mn10300.c: Cleanup.
4994 (md_pcrel_from): Enable.
4998 * expr.c (operand): Fix a comment typo.
4999 * write.c (write_relocs): Fix a signed/unsigned warning.
5001 * config/tc-hppa.c (fudge_reg_expressions): New
5002 (hppa_force_reg_syms_absolute): New.
5003 (pa_equ): Allow reg_section expressions.
5004 * config/tc-hppa.c (md_optimize_expr): Define.
5005 (hppa_force_reg_syms_absolute): Prototype.
5007 * config/tc-hppa.c (pa_11_fp_reg_struct): Delete.
5008 (pa_parse_number): Pass in arg to select fp reg parsing.
5009 Return 1 to indicate format checks pass. If strict, then only
5010 accept a register or register symbol. Return value in...
5011 (pa_number): New static for pa_parse_number.
5012 (FP_REG_BASE): Define.
5013 (FP_REG_RSEL): Define.
5014 (pre_defined_registers): Apply FP_REG_BASE and FP_REG_RSEL as
5015 appropriate. White space changes.
5016 (need_pa11_opcode): Don't bother passing any params, get them from
5018 (pa_ip): Modify all calls to pa_parse_number and need_pa11_opcode.
5019 Remove extraneous check in case 'Q'.
5020 (pa_equ): Modify call to pa_parse_number to do strict parsing. If
5021 reg, set section of resulting symbol to reg_section.
5022 (pa_parse_space_stmt): Modify call to pa_parse_number.
5023 (pa_space): Likewise.
5025 * config/tc-hppa.c: (md_apply_fix): Handle vtable relocs.
5026 (hppa_force_relocation): Handle vtable relocs.
5027 (pa_vtable_entry): New.
5028 (pa_vtable_inherit): New.
5029 (md_pseudo_table): Add entries for vtable pseudos.
5030 (hppa_fix_adjustable): Reject reduction of R_PARISC_GNU_VTINHERIT
5031 and R_PARISC_GNU_VTENTRY relocs. Reject reduction of relocs
5033 (tc_gen_reloc): Remove ELF_ARG_RELOC_INSN code.
5034 (pa_type_args): Don't call symbol_get_bfdsym multiple times.
5035 Set STT_PARISC_MILLICODE for OBJ_ELF when encountering a
5037 * config/obj-elf.c (obj_elf_type): Allow md_elf_symbol_type to
5038 specify a symbol type.
5040 * config/tc-hppa.h: Reorganize file a little, grouping OBJ_ELF
5041 dependent things together.
5042 (md_elf_symbol_type): Define.
5044 * config/tc-hppa.c (fix_new_hppa): Elide "$PIC_pcrel$0" pseudo
5046 * config/tc-hppa.h (tc_frob_symbol): Elide "$PIC_pcrel$0" here too.
5048 * config/obj-elf.h (obj_elf_vtable_inherit): Declare.
5049 (obj_elf_vtable_entry): Declare.
5051 * config/obj-elf.c (obj_elf_vtable_inherit): Return struct fix *
5052 and export function.
5053 (obj_elf_vtable_entry): Similarly.
5054 (elf_pseudo_table): Fix the damage with a cast.
5058 * config/tc-ia64.c (emit_one_bundle): Stop collecting insns
5059 for template selection when a label is needed.
5063 * config/tc-ia64.c: Fix formatting.
5067 * configure.in: Increase version number to 2.10.91.
5068 * configure: Regenerate.
5069 * aclocal.m4: Regenerate.
5070 * config.in: Regenerate.
5071 * po/gas.pot: Regenerate.
5072 * Makefile.in: Regenerate.
5076 * config/tc-sh.h [OBJ_ELF] (TC_FIX_ADJUSTABLE): Define.
5077 * config/tc-sh.c (md_apply_fix): Map 32-bit relocations that
5078 become PC-relative to BFD_RELOC_32_PCREL. Reject 16- or 8-bit
5080 (sh_obj_adjustable): Return 1 for PC-relative offsets used in
5085 * config/tc-sh.h (DIFF_EXPR_OK, GLOBAL_OFFSET_TABLE_NAME,
5086 TC_RELOC_GLOBAL_OFFSET_TABLE, TC_RELOC_RTSYM_LOC_FIXUP): Define.
5087 * config/tc-sh.c (sh_elf_cons, sh_elf_suffix): New functions.
5088 [OBJ_ELF] (md_pseudo_table) <long, int, word, short>: Use them.
5089 (GOT_symbol): New variable.
5090 (md_undefined_symbol): Set it.
5094 * config/tc-ia64.c (match): Don't inline.
5095 (extra_goodness): New.
5096 (md_begin): Prefer nop.f and nop.b for best_template.
5100 * as.c: Fix formatting.
5102 * frags.c: Likewise.
5103 * macro.c: Likewise.
5107 * config/tc-mn10300.c: Cleanup and fix warnings.
5108 (md_pseudo_table): Add initializers.
5109 (md_show_usage): Cleanup.
5110 (md_parse_option): Fix warnings.
5111 (md_undefined_symbol): Fix warnings.
5112 (md_conver_frag): Fix warnings.
5113 (tc_gen_reloc): Fix warnings.
5114 (md_apply_fix3): Fix warnings.
5115 (check_operand): Fix warnings.
5119 * acinclude.m4: Include libtool and gettext macros from the
5121 * aclocal.m4, configure: Rebuilt.
5125 * config/tc-ppc.c (md_parse_option): Recognize -m405.
5129 * listing.c: Fix formatting.
5133 * app.c: Fix a comment typo. Fix formatting.
5137 * config/tc-vax.c (md_convert_frag): Correctly calculate the pc relative
5138 offset of the target destination for jmp instructions.
5139 (md_assemble): Change mode to VAX_ABSOLUTE_MODE as per comments.
5143 * NEWS: Mention support for CRIS.
5147 * config/tc-avr.h (TC_IMPLICIT_LCOMM_ALIGNMENT): New macros.
5148 Sets `.lcomm' alignment to zero.
5152 * config/tc-i386.h (OBJ_MAYBE_ELF, OBJ_MAYBE_COFF,
5153 TC_FIX_ADJUSTABLE): Define.
5157 * config/tc-ia64.c (output_unw_records): Set U & E flags only if
5158 unwind.personality_routine is set.
5162 * write.c (TC_FIX_ADJUSTABLE): Remove the duplicate.
5166 * config/tc-i386.h (TC_FIX_ADJUSTABLE): Do *NOT* define if target
5171 * config.in (STRICTCOFF): New for strict COFF.
5173 * configure.in: Define STRICTCOFF for i386-*-msdosdjgpp*,
5174 i386-*-go32* and i386-go32-rtems*.
5175 * configure: Rebuilt.
5177 * config/obj-coff.c (obj_coff_endef): Follow the historical
5178 behavior if STRICTCOFF is not defined.
5180 * doc/internals.texi: Document STRICTCOFF.
5184 * write.c (TC_FIX_ADJUSTABLE): Define to 1, if not defined.
5185 (fixup_segment) Use it instead of TC_DONT_FIX_NON_ADJUSTABLE.
5186 * config/tc-i386.h (TC_DONT_FIX_NON_ADJUSTABLE): Remove.
5187 <OBJ_ELF, OBJ_COFF, TE_PE> (TC_FIX_ADJUSTABLE): Define.
5188 * config/tc-arm.h (TC_DONT_FIX_NON_ADJUSTABLE): Remove.
5189 <OBJ_ELF> (TC_FIX_ADJUSTABLE): Define.
5190 * config/tc-i960.h, config/tc-m68k.h, config/tc-v850.h:
5195 * config/tc-mn10300.c: (md_apply_fix): New function.
5196 (mn10300_force_relocation): New function.
5197 (mn10300_fix_adjustable): New function.
5199 * config/tc-mn10300.h: (TC_FORCE_RELOCATION): Define.
5200 (TC_HANDLES_FX_DONE): Define.
5201 (obj_fix_adjustable): Define.
5202 (MD_APPLY_FIX3): Define.
5203 (TC_LINKRELAX_FIXUP): Define.
5205 * write.c: (TC_LINKRELAX_FIXUP): Define if not
5207 (fixup_segment): Use TC_LINKRELAX_FIXUP.
5209 * doc/internals.texi: Document TC_LINKRELAX_FIXUP.
5213 * config/tc-i860.c (md_apply_fix3): Do not insert the immediate
5214 if the fixup resulted in a relocation.
5218 * config/tc-arm.c (decode_shift): Replace as_tsktsk with as_warn.
5219 Make reference to first element of shift_names explicit.
5223 * write.c (fixup_segment) [TC_DONT_FIX_NON_ADJUSTABLE]: Use
5224 obj_fix_adjustable() and tc_fix_adjustable() to tell whether to
5225 add a symbol's address. Removed all target-specific #ifdefs that
5226 used to accomplished the same.
5227 * config/tc-v850.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
5228 * config/tc-m68k.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
5229 * config/tc-arm.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
5230 * config/tc-i960.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
5231 * config/tc-i386.h (TC_DONT_FIX_NON_ADJUSTABLE): Define.
5235 * dwarf2dbg.c: Fix formatting.
5239 * config/tc-arm.c (decode_shift): Allow illegal shifts by zero
5240 to be recoded as logical shift lefts by zero.
5244 * config/tc-ia64.c (specify_resource, case IA64_RS_GR): Handle
5245 postincrement modified registers. Handle IA64_OPND_R3_2 addl
5247 (note_register_values): Handle IA64_OPND_R3_2 operands.
5251 * config/tc-i860.c (md_operand): Silly typo fixed.
5255 * config/tc-arm.c (struct asm_shift): Delete.
5257 (enum asm_shift_index): New.
5258 (struct asm_shift_properties): New.
5259 (struct asm_shift_name): New.
5260 (shift_properties[]); New.
5261 (shift_names[]); New.
5263 (decode_shift): Use new structures.
5264 Issue a warning is "ROR #0" is used.
5265 Issue a warning if "ASR #0" or "LSR #0" is used.
5267 (md_begin): Initialise arm_shift_hsh table from new
5268 asm_shift_name array.
5272 * config/tc-sparc.c: Kill all warnings.
5273 (md_parse_option): Set -32/-64 for -xarch=, allow all -A archs
5275 (md_show_usage): Update usage text.
5279 * config/tc-arm.c (do_bx): Warn about "bx px" not being very
5284 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Defined.
5286 * config/tc-sh.c (md_assemble): Changed so debug_type
5287 test performed for ppi_assemble
5288 * config/tc-sh.c: Included dwarf2dbg.h.
5289 (debug_line): Defined.
5290 (md_assemble): Generates dwarf2 line info.
5291 (sh_finalize): New function. Finalize dwarf2 info.
5292 (assemble_ppi): Returns size of code generated.
5293 (build_Mytes): Returns size of code generated.
5294 (md_pseudo_table): Added "file" and "loc" psuedo ops.
5295 * config/tc-sh.h (md_end): Defined.
5296 (sh_finalize): Declared.
5300 * config/tc-sh.c (md_apply_fix) [BFD_RELOC_32, BFD_RELOC_16]: Use
5305 * config/tc-arm.c (do_bx): Allow "bx pc".
5309 * config/tc-ia64.c (md_longopts): Add -mconstant-gp and -mauto-pic.
5310 (md_parse_option): Add OPTION_MCONSTANT_GP and OPTION_MAUTO_PIC.
5311 (md_begin): Change assignment to md.flag to OR in the new bit.
5315 * config/obj-coff.c (obj_coff_endef) [BFD_ASSEMBLER]: Set the debug
5316 flag for storage types C_ARG, C_REGPARM, C_FIELD, C_MOS, C_MOE,
5321 * NEWS: Mention i860 support.
5325 * config/tc-mn10300.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
5329 * doc/c-arm.texi (ARM Directives): Fix warnings from makeinfo.
5333 * doc/c-i860.texi (Opcodes for i860): Remove braces from @item
5338 * expr.c: Fix formatting.
5339 * config/obj-bout.c: Likewise.
5343 * doc/c-i860.texi: Flesh out the i860 section more.
5347 * symbols.c: Fix formatting.
5352 * config/tc-i386.c (md_assemble): Skip suffix check if the opcode
5353 modifier has the IgnoreSize bit set.
5358 * configure.in: Use elf on Unixware 7 (i586-sco-sysv5uw7.1.0)
5359 * configure: Regenerate.
5363 * hash.c: Fix formatting.
5368 * config/tc-i860.h: Rework completely for BFD_ASSEMBLER.
5369 (i860_fix_info): New enum.
5370 (MD_APPLY_FIX3): Define.
5371 (WORKING_DOT_WORD): Define.
5372 (TC_HANDLES_FX_DONE): Define.
5373 (DIFF_EXPR_OK): Define.
5374 (LISTING_HEADER): Define.
5375 (TARGET_FORMAT): Select target format based on endian flag.
5376 (TARGET_BYTES_BIG_ENDIAN): Default to little endian.
5377 (target_big_endian): Add external declaration.
5379 * config/tc-i860.c: All existing code reworked completely. Other
5380 new code shown below.
5381 (SYNTAX_SVR4): Define.
5382 (target_warn_expand): New variable.
5383 (md_shortopts): Declare and define (-Qy, -Qn, and -V options).
5384 (md_longopts): Declare and define with new options (-EL, -EB,
5386 (md_show_usage): New function.
5387 (md_operand): New function.
5388 (obtain_reloc_for_imm16): New function.
5389 (md_apply_fix3): New function.
5390 (tc_gen_reloc): New function.
5394 * config/tc-m68hc11.c (build_jump_insn): Make sure the
5395 2 bytes of the jump address are in the same frag.
5396 (find): Accept 68hc12 register indirect modes.
5398 * NEWS: Mention 68HC11 & 68HC12 support.
5402 * config/tc-ia64.c (unwind): Add prologue_mask member.
5403 (dot_vframe): Elide psp_gr record if it overlaps prologue_gr.
5404 (dot_save): Likewise for pfs_gr, rp_gr, and preds_gr.
5405 (dot_body): Clear unwind.prologue_mask.
5406 (dot_prologue): Set it. Accept a register second argument.
5410 * config/atof-ieee.c: Fix formatting.
5411 * config/atof-tahoe.c: Likewise.
5415 * config/tc-sparc.c (md_begin): Fix typo in recent formatting
5418 * doc/as.texinfo (Pseudo Ops): Update to include descriptions
5419 of .popsection, .previous, .pushsection, .subsection,
5420 .version, .vtable_entry, .vtable_inherit and .weak.
5424 * config/tc-cris.c: Fix formatting.
5425 * config/tc-i386.c: Likewise.
5426 * config/tc-sparc.c (sparc_ip): Simplify the code.
5430 * config/tc-cris.c: Rearrange code for readability.
5431 * config/tc-d10v.c: Fix formatting.
5432 * config/tc-m32r.c: Likewise.
5433 * config/tc-sparc.c: Likewise.
5437 * config/tc-ia64.c (emit_one_bundle): Call ia64_free_opcode
5438 before ia64_find_opcode.
5439 (md_assemble): Likewise.
5443 * config/tc-arm.c (do_mrs): Fix skip of 'cpsr_all' flag.
5444 Undo some formatting fixes.
5448 * config/obj-som.c: Fix formatting.
5449 * config/obj-ieee.c: Likewise.
5450 * config/tc-arm.c: Likewise.
5451 * config/tc-v850.c: Likewise.
5455 * doc/c-m68k.texi (section M680x0 Options): Turn into a table
5456 index by command line option.
5460 * doc/c-m68k.texi (@cindex @samp{--pcrel}): Rewrite option description.
5461 (@node M68K-Branch): Rewrite to match the reality.
5465 * doc/c-i860.texi: New file.
5466 * doc/Makefile.am (CPU_DOCS): Add c-i860.texi.
5467 * doc/Makefile.in: Regenerate.
5468 * doc/all.texi: Add I860 as relevant architecture.
5469 * doc/as.texinfo: Include i860 dependent file c-i860.texi.
5473 * config/tc-d30v.c: Fix formatting.
5477 * config/tc-m68k.c (flag_keep_pcrel, OPTION_PCREL): Add --pcrel option.
5478 (md_convert_frag_1, md_estimate_size_before_relax): When making DBcc
5479 long emit a long branch if available instead of an absolute jump, never
5480 emit absolute jumps for anything with --pcrel.
5482 * doc/c-m68k.texi: Document new command line option.
5486 * config/tc-avr.c: Use PARAMS macro in function declarations.
5487 Don't declare md_pcrel_from_section (already in tc-avr.h).
5488 (avr_operands): Use AVR_UNDEF_P and AVR_SKIP_P macros.
5489 (avr_operand): Don't set (unsigned) op_mask to -1.
5493 * configure.in: Add bits for i860-stardent-{sysv4, elf}*.
5494 * configure: Regenerated.
5495 * config/obj-elf.c (obj_elf_type): Recognize a fifth type
5496 of operand to the .type directive (.e.g, "type").
5500 * as.h (warn_comment, found_comment, found_comment_file): Declare.
5501 * app.c (do_scrub_chars): Record where first comment found.
5502 * read.c (read_a_source_file): Init found_comment on entry, and
5503 notify whether comments found on exit.
5504 * config/tc-hppa.c (md_shortopts): Add "c".
5505 (md_longopts): Add warn-comment.
5506 (md_parse_option): Handle it.
5507 (md_show_usage): Show available options.
5508 * config/tc-hppa.h (WARN_COMMENTS): Define if TE_LINUX
5512 * config/tc-mn10300.c (md_convert_frag): Fix printfs.
5513 (tc_gen_reloc): Add cast when assigning bfd_abs_symbol to
5515 (md_estimate_size_before_relax): Don't fall off end of function.
5519 * config/tc-avr.c: Fix formatting.
5520 * config/tc-ns32k.c: Likewise.
5524 * config/tc-d10v.c (find_opcode): Remove extraneous `='.
5528 * config/tc-d10v.c: Fix formatting.
5529 * config/tc-z8k.c: Likewise.
5530 * config/tc-sparc.c: Likewise.
5534 * cgen.c (queue_fixup): Declare opinfo.
5535 (gas_cgen_parse_operand): Mark unused parameters with ATTRIBUTE_UNUSED.
5536 (gas_cgen_md_operand): Ditto.
5537 (gas_cgen_md_apply_fix3): Ditto.
5541 * config/obj-coff.c (obj_frob_symbol): Don't merge
5542 labels. Don't merge if the symbol isn't constant. Return
5543 immediately if a symbol is merged.
5547 * frags.c (frag_align): Correct absolute section alignment.
5551 * config/obj-coff.c (obj_frob_symbol): revert previous change,
5552 it breaks linking against DLLs.
5556 * configure.in: Add CRIS support.
5557 * configure: Regenerate.
5558 * Makefile.am: (CPU_TYPES): Add cris.
5559 (CPU_OBJ_VALID) [aout]: Add cris.
5560 (MULTI_CPU_TYPES): Add cris.
5561 (MULTI_CPU_OBJ_VALID) [aout]: Add cris.
5562 [coff]: Only i386 and mips are valid.
5563 (TARGET_CPU_CFILES): Add config/tc-cris.c.
5564 (TARGET_CPU_HFILES): Add config/tc-cris.h.
5565 (MULTI_CFILES): Add config/e-crisaout.c and config/e-criself.c.
5566 Regenerate dependencies.
5567 * Makefile.in: Regenerate.
5568 * aclocal.m4: Regenerate.
5569 * as.c: Declare crisaout, criself.
5570 * config/tc-cris.h, config/tc-cris.c: New.
5571 * config/e-criself.c, config/e-crisaout.c: New.
5572 * po/POTFILES.in, po/gas.pot: Regenerate.
5576 * read.c: Fix formatting.
5577 * write.c: Fix formatting.
5581 * sb.c: Include <stdlib.h> if exists for abort ().
5585 * config/tc-hppa.c (hppa_fix_adjustable): Correct LR%/RR% comment.
5586 (arg_reloc_stub_needed): #ifdef OBJ_SOM, not #ifdef SOM.
5587 (pa_type_args): Same here.
5591 * config/obj-coff.c (obj_frob_symbol): Don't merge labels. Don't
5592 merge if the symbol isn't constant. Don't call S_SET_EXTERNAL if
5593 the storage class is already set.
5597 * config/tc-m68hc11.c: Fix formatting.
5598 * config/tc-mn10200.c: Likewise.
5599 * config/tc-mn10300.c: Likewise.
5600 * config/tc-pj.c: Likewise.
5601 * config/tc-tic80.c: Likewise.
5602 * config/tc-w65.c: Likewise.
5606 * expr.c (operand): Permit %bin literals if LITERAL_PREFIXPERCENT_BIN
5611 * doc/c-mips.texi (MIPS Opts): Remove erroneous space after
5616 * config/tc-hppa.c (hppa_fix_adjustable): Use the same checks for
5617 ELF as are used for SOM (except the 32-bit reloc one) to decide
5618 whether a symbol can be reduced to a section symbol. Expand on
5619 the comment for symbols involved in LR% and RR% expressions.
5623 * config/tc-mips.c (mips_disable_float_construction): New
5624 static variable. Set to true if doubles should not be
5625 constructed by loading two single width fp registers with
5626 halves of the value.
5627 (mips_ip): Test mips_disable_float_construction.
5628 (md_longopts): Add command line switches --construct-floats
5629 and --no-construct-floats.
5630 (md_parse_option): Parse new command line options.
5631 (md_show_usage): Describe new command line options.
5633 * doc/c-mips.texi: Document new command line options.
5637 * configure.in: Remove the test /usr/dde for mips-*-sysv4*MP*
5638 * configure: Regenerate.
5642 * configure.in (DEFAULT_EMULATION setting): Revert part of
5643 2000-07-01 change that set te_multi=multi unless set to tmips.
5644 * configure: Regenerate.
5648 * config/obj-coff.c (coff_frob_section): Add padding to the last
5649 section when aligning it increases its size.
5653 * config/tc-tic54x.c: Fix formatting.
5657 * config/tc-h8500.c: Fix formatting.
5658 * config/tc-tic54x.c: Fix formatting.
5662 * config/tc-hppa.c (md_apply_fix): Check fmt 12 and 22 pc-rel
5663 displacements correctly.
5665 * read.h (s_abort): Add ATTRIBUTE_NORETURN.
5669 * hash.c (hash_insert): Add cast to obstruct_alloc to fix
5676 * config/te-hppalinux64.h: Add a new emulation.
5677 * configure.in (emulations): Add configure bits to support the
5678 64-bit Linux/parisc target.
5679 * configure: Regenerate.
5681 * config/tc-hppa.c (md_parse_option): Support `-V' for ELF.
5682 (md_shortopts): Add `V' for ELF.
5684 * config/tc-hppa.c (arg_reloc_stub_needed): Define as zero except
5685 when SOM or ELF_ARG_RELOC are defined.
5686 (pa_type_args): Only set symbol_arg_reloc_info when SOM or
5687 ELF_ARG_RELOC are defined.
5688 (pa_stringer_aux): Don't pa_check_current_space_and_subspace here..
5689 (pa_stringer): ..Do it here instead. Fix comment typos.
5690 (hppa_force_relocation): Cast enums to int before comparing with
5694 * config/tc-hppa.c: Removed unneeded libbfd.h to fix macro
5695 redifinition warning.
5696 (md_apply_fix): Added cast from enum to int for fixP->fx_r_type.
5697 (hppa_force_relocation): ditto
5698 (md_apply_fix): Added cast to buf to fix warnings.
5700 * config/tc-hppa.h (pa_define_label, parse_cons_expression_hppa,
5701 cons_fix_new_hppa, hppa_force_relocation): Prototype.
5703 * config/tc-hppa.c (reloc_type): It's an enum for OBJ_ELF.
5704 (R_N0SEL, R_N1SEL): Define only for OBJ_SOM.
5705 (tc_gen_reloc): Make `code' a reloc_type and `codes' a
5706 reloc_type** to avoid warnings in switch.
5707 (md_apply_fix): Make insn, val signed. Zap buf_wd and read insn a
5708 little earlier instead.
5710 * config/tc-hppa.c (symbol_arg_reloc_info): Define for both som
5712 (pa_type_args): Use symbol_arg_reloc_info.
5713 (struct pa_it): Make arg_reloc unsigned int.
5714 (struct hppa_fix_struct): Likewise for fx_arg_reloc.
5715 (pa_text, pa_data, pa_comm): Don't compile for TE_LINUX.
5716 (pa_code): Delete. pa_text duplicates this function.
5717 (md_pseudo_table): Call obj_elf_text for ".code" if TE_LINUX.
5718 (fix_new_hppa): Argument offset is offsetT, arg_reloc is unsigned
5720 (cons_fix_new_hppa): Actually change selector to e_fsel when
5721 warning about assuming so.
5722 (tc_gen_reloc): More example elf arg reloc code.
5723 (md_apply_fix): Use arg_reloc_stub_needed for elf too.
5724 (hppa_force_relocation): Likewise.
5726 * config/tc-hppa.h: Use TARGET_ARCH_SIZE to select target include
5728 (pa_end_of_source): Prototype.
5729 (hppa_fix_adjustable): Prototype.
5730 (LABELS_WITHOUT_COLONS): Move it..
5731 * config/te-hppa.h: To here.
5733 * config/te-hppa64.h: New file.
5735 * config/tc-hppa.c: Use TARGET_ARCH_SIZE to select target reloc
5737 (md_apply_fix): Pass stdoutput to bfd_hppa_insn2fmt. Handle
5738 format -10, -16, 16 relocs.
5739 (hppa_elf_mark_end_of_function): Test for null
5740 last_call_info->start_symbol
5742 * config/tc-hppa.c (pa_ip): In case 'V', pass `strict' to
5743 CHECK_FIELD, not INSERT_FIELD_AND_CONTINUE. Don't pass opcode to
5744 re_assesmble_* functions. Delete extraneous statements. Fix
5746 (md_apply_fix): Compare against 1048575 in case 21. Don't pass
5747 insn to re_assemble_*.
5751 * doc/internals.texi (Expressions): Fix typo.
5755 * config/tc-sh.c: Fix formatting.
5756 * config/tc-tic54x.c: Fix formatting.
5757 * depend.c: Fix formatting.
5758 * flonum-konst.c: Likewise.
5759 * flonum-mult.c: Likewise.
5763 * config/tc-sh.c: Fix comments.
5764 * config/obj-vms.c: Fix comments.
5765 * config/tc-a29k.c: Likewise.
5766 * config/tc-alpha.c: Likewise.
5767 * config/tc-h8300.c: Likewise.
5768 * config/tc-h8500.c: Likewise.
5769 * config/tc-i370.c: Likewise.
5770 * config/tc-ia64.c: Likewise.
5771 * config/tc-m68hc11.c: Likewise.
5772 * config/tc-m68k.c: Likewise.
5773 * config/tc-mips.c: Likewise.
5774 * config/tc-ns32k.c: Likewise.
5775 * config/tc-ppc.c: Likewise.
5776 * config/tc-z8k.c: Likewise.
5780 * config/obj-coff.c (TC_COFF_SECTION_DEFAULT_ATTRIBUTES): New.
5781 Default to '(SEC_LOAD | SEC_DATA)'.
5782 (obj_coff_section) [BFD_ASSEMBLER]: Use it.
5784 * doc/internals.texi (CPU Backend): Describe
5785 TC_COFF_SECTION_DEFAULT_ATTRIBUTES.
5789 * as.c (parse_args): NULL terminate the long option list.
5793 * config/tc-d30v.h: Include "write.h" for fixS.
5794 (d30v_start_line, md_pcrel_from_section): Add function prototypes.
5798 * config/tc-d30v.c (write_2_short): Further changes to warning
5799 messages produced when combining EITHER_BUT_PREFER_MU attributed
5808 * config/tc-arm.c (psrs): Accept combinations of flags.
5812 * config/tc-avr.c: Change _ () to _() around all strings marked
5813 for translation (exception from the usual coding style).
5814 (avr_opt): New struct variable, how the new switches are set.
5815 (OPTION_MMCU): Define as 'm' and actually use.
5816 (md_longopts): Add -mall-opcodes, -mno-skip-bug, -mno-wrap.
5817 (show_mcu_list): New function, display the list of known MCUs.
5818 (md_show_usage): Document the new switches. Call show_mcu_list.
5819 (avr_set_arch): Change 'm' to OPTION_MMCU.
5820 (md_parse_option): Call show_mcu_list if unknown MCU specified.
5821 Handle the new switches.
5822 (avr_operands): Disable warnings for undefined combinations of
5823 operands if -mall-opcodes. Disable warnings for skipping two-word
5824 instructions if enhanced core or -mno-skip-bug.
5825 (avr_operand): Accept all addressing modes on avr1 if -mall-opcodes.
5826 (md_apply_fix3): Reject 8K wrap if >8K or -mno-wrap.
5827 (md_assemble): Accept opcodes not supported by MCU if -mall-opcodes.
5828 (avr_ldi_expression): Warn about implicit lo8().
5829 * config/tc-avr.h (md_pcrel_from_section): Add prototype.
5833 * configure.in: Add entry for mips-*-sysv4*MP*
5834 * configure: Rebuild
5835 * config/tc-mips.c (mips_target_format): Return elf32-tradbigmips or
5836 elf32-tradlittlemips for traditional mips targets.
5837 * config/tc-mips.c (md_estimate_size_before_relax): Duplicate the
5838 test for Link Once sections as in adjust_reloc_syms.
5839 * config/te-tmips.h: New file for traditional mips targets. Define
5844 * config/obj-coff.c (obj_coff_setcion) [BFD_ASSEMBLER]: If the
5845 flags argument is not present, don't change an existing section's
5846 section's attributes. If the flags argument is present, warn if the
5847 attributes don't match the section's current attributes. When
5848 long section names are supported, set SEC_LINK_ONCE and
5849 SEC_LINK_DUPLICATES_DISCARD for a new .gnu.linkonce section.
5853 * config/obj-aout.c (obj_aout_type): Do not ignore for undefined
5854 symbols; create them.
5858 * write.c (set_segment_vma): New: Set vma and lma for a segment.
5859 (write_object_file) [BFD_ASSEMBLER && OBJ_COFF && TE_GO32]: Use it.
5863 * config/tc-mips.c (mips_ip): handle "(foo-.-4)" type of
5864 expressions. Ignore the problem when handling 16 bit signed
5865 immediates, because the assembler will take care of the relocation
5870 * config/tc-d30v.c (write_2_short): Do not allow opcodes with
5871 the EITHER_BUT_PREFER_MU attribute to be combined into a reverse
5872 sequential order, and emit warning messages if the input source
5873 code contains constructs like that, or parallel constructs
5874 containing such opcodes.
5878 * config/tc-avr.c (mcu_types): Rename avr4 to avr5, add avr4.
5879 Add more MCU types for avr4 and avr5. Replace at94k{10,20,40}
5880 with just at94k. Change AVR_ISA_85xx back to AVR_ISA_2xxx.
5881 (md_show_usage): Update usage message.
5882 (md_parse_option): Allow redefinition of MCU type within the
5883 same avr[1-5] bfd machine type. Show both old and new MCU type
5884 in the error message.
5885 (md_apply_fix3): Support 8K wrap if AVR_ISA_MEGA is not set.
5886 Simplify 8K wrap code.
5890 * config/obj-aout.c: Remove all uses of DEFUN.
5891 * config/obj-ieee.c: Likewise.
5892 * config/tc-sh.c: Fix comment typos.
5893 * config/tc-tahoe.c: Likewise.
5894 * config/tc-vax.c: Likewise.
5895 * config/tc-w65.c: Likewise.
5896 * config/tc-z8k.c: Likewise.
5897 * config/tc-h8300.c (build_bytes): Assemble ldmac correctly.
5901 * config/tc-i386.c (md_estimate_size_before_relax): Revert
5902 more changes from Sept 1999
5903 (tc_i386_fix_adjustable): ditto
5904 (md_apply_fix3): ditto
5908 * cgen.c (expr_jmp_buf_p): New validity flag for expr_jmp_buf.
5909 (gas_cgen_parse_operand): Set it around expression() call.
5910 (gas_cgen_md_operand): Test for it before longjmp().
5914 * config/tc-h8500.c: Remove all uses of DEFUN.
5915 * config/tc-sh.c: Likewise.
5916 * config/tc-w65.c: Likewise.
5917 * config/tc-z8k.c: Likewise.
5919 * config/tc-h8500.c: Fix typos in comments.
5923 * expr.c (operand): Permit $hex literals if LITERAL_PREFIXDOLLAR_HEX
5928 * expr.c (operand): Do not as_bad() if RELAX_PAREN_GROUPING is
5929 defined. Fix error message for `[' grouping.
5933 * config/tc-h8300.c: Fix formatting and comment typos.
5937 * config/tc-ia64.c (note_register_values): Move premature QP
5938 notation clearing into the appropriate place.
5942 * dep-in.sed: Escape literal `.'s on patterns. Trim off `../'
5943 first before anything else. Add bin-bugs.h, emul.h and progress.h
5944 Sort list of files as for $(OBJS) in Makefile.am.
5946 * Makefile.am (DEP): grep for leading `/' in DEPA, and fail if we
5947 find one. Remake dependencies.
5948 ($(OBJS)): Add bin-bugs.h, emul.h, and progress.h Sort the list.
5949 * Makefile.in: Regenerate.
5950 * doc/Makefile.in: Regenerate.
5952 * config/tc-i386.c (i386_displacement): Don't assume a constant
5953 displacement is necessarily 16 bits when in 16 bit code mode.
5954 (md_assemble): Instead size the displacement here after we know
5955 for sure that a .code16gcc operand hasn't automatically added
5956 operand size prefixes.
5960 * Makefile.am: Rebuild dependency.
5961 * Makefile.in: Rebuild.
5965 * config/tc-h8300.c (parse_reg): Make the function static.
5966 (parse_exp): Likewise.
5970 * config/tc-i386.c (tc_i386_fix_adjustable): Revert change from
5971 Sept 1999; RVA relocs need to be treated more like DIR32 relocs
5972 for cygwin import libraries to work properly.
5976 * Makefile.am: Rebuild dependency.
5977 * Makefile.in: Rebuild.
5978 * configure: Likewise.
5979 * doc/Makefile.in: Likewise.
5983 * doc/internals.texi (CPU backend): Add @itemx for
5984 TC_START_LABEL_WITHOUT_COLON.
5985 * doc/c-tic54x.texi: New.
5986 * doc/as.texinfo: Add tic54x features and include primary tic54x
5988 * doc/all.texi: Add C54X.
5989 * doc/Makefile.am (CPU_DOCS): Add c-tic54x.texi.
5990 * doc/Makefile.in: Regenerate.
5991 * configure.in: Add tic54x and define LIBM for tic54x.
5992 * configure: Regenrate.
5993 * config/tc-tic54x.[ch]: New.
5994 * config/obj-coff.h: Add tic54x.
5995 * Makefile.am: (CPU_TYPES): Add tic54x.
5996 (TARGET_CPU_CFILES): Add 'tc-tic54x.c'.
5997 (TARGET_CPU_HFILES): Add 'tc-tic54x.h'.
5998 (as_new_LDADD): Add $(LIBM).
5999 * Makefile.in: Regenerate.
6003 * doc/Makefile.am (CPU_DOCS): Added 68hc11 file.
6004 * doc/c-m68hc11.texi: Document 68HC11 and 68HC12 port.
6005 * doc/as.texinfo: Likewise.
6007 * configure, Makefile.in: Regenerate.
6008 * configure.in (emulations): Recognize m6811 and m6812.
6009 * Makefile.am (CPU_TYPES, TARGET_CPU_CFILES, TARGET_CPU_HFILES):
6010 Added files for 68hc11 and 68hc12 assembler.
6011 * config/tc-m68hc11.c: Assembler for 68hc11 and 68hc12.
6012 * config/tc-m68hc11.h: Header definition for that assembler.
6016 * symbols.c (resolve_symbol_value): Use bfd_octets_per_byte
6017 instead of OCTETS_PER_BYTE.
6019 * config/tc-v850.c: Fix compile time warnings.
6020 * config/tc-ppc.c: Fix compile time warnings.
6024 * configure.in: Don't emulate i386-pc-pe-coff with i386coff.
6025 * configure: Rebuild.
6029 * config/obj-coff.c (obj_coff_weak): Typo fix: Change BFD_ASSEMLER
6034 * config/tc-mips.c (md_parse_option): Accept RM5200,RM5230,
6035 RM5231, RM5261, RM5721 and RM7000 as r5000 cpu variants.
6037 * doc/c-mips.texi: Document newly accepted cpu variants.
6041 * config/tc-mips.h: Remove definition of ONLY_STANDARD_ESCAPES.
6045 * macro.c (getstring): Make it possible to escape the quote
6050 * config/tc-hppa.c (pa_export): Weak symbols can be global.
6054 * configure: Regenerate.
6058 * app.c (do_scrub_begin): Don't default lex[';'] as a line
6060 * doc/internals.texi (line_separator_chars): Semicolon is no
6061 longer a default. Mention null and newline as defaults.
6063 * read.c (is_end_of_line): Remove ifdef TC_HPPA.
6065 * config/tc-i386.h (line_separator_chars): Explicitly mention `;'
6066 * config/tc-i860.h (line_separator_chars): Likewise.
6067 * config/tc-h8300.c (line_separator_chars): Likewise.
6068 * config/tc-i960.c (line_separator_chars): Likewise.
6069 * config/tc-m68k.c (line_separator_chars): Likewise.
6070 * config/tc-mips.c (line_separator_chars): Likewise.
6071 * config/tc-ns32k.c (line_separator_chars): Likewise.
6072 * config/tc-sparc.c (line_separator_chars): Likewise.
6073 * config/tc-vax.c (line_separator_chars): Likewise.
6075 * config/tc-h8300.c (comment_chars): Use string initialiser.
6076 * config/tc-i960.c (line_comment_chars): Likewise.
6077 * config/tc-z8k.c (comment_chars, line_comment_chars,
6078 line_separator_chars): Likewise.
6080 * config/tc-arm.c (line_separator_chars): Always use `;', not just
6085 * config/tc-arm.c (cons_fix_new_arm): Assign correct reloc value
6090 * configure.in (VERSION): Update to show this is the CVS mainline.
6094 * config/tc-alpha.c (md_undefined_symbol): Properly understand that
6095 $at is the integer register $r28, vs. both $r28 and the floating
6096 point register $f28.
6100 * config/tc-ia64.c (generate_unwind_image): Call ia64_flush_insns.
6101 (dot_endp): Don't call ia64_flush_insns.
6102 (emit_one_bundle): Don't delete prologue/body records from
6103 unwind_record list in first loop. Rewrite second loop to account for
6108 * config/tc-ia64.c: Add missing prototypes.
6109 (generate_unwind_image): Cast argument to output_unw_records call.
6113 * config/tc-avr.c (avr_operand): fix the formatting of the comment.
6117 * config/tc-avr.c (AVR_ISA_???): moved to include/opcode/avr.h
6118 (REGISTER_P): likewise.
6119 (avr_opcodes): uses include/opcode/avr.h
6120 (avr_operand): enable ld r,Z or st r,Z for at90s1200.
6124 * read.c (is_end_of_line): No ';' for TC_HPPA. Add missing
6129 * read.c (is_end_of_line): Put back `;'.
6133 * config/tc-i386.c (md_shortopts): Remove 'm', add 'q' to non-elf.
6137 * expr.c (operand): Test is_end_of_line outside switch to catch
6138 line separator chars that are also operators.
6139 (operator): Return O_illegal for line separator chars.
6141 * read.c (is_end_of_line): Use 1 instead of 99. Don't set `;'
6142 entry (or `!' entry for TC_HPPA).
6144 * config/tc-arm.c (my_get_float_expression): Cast to unsigned char
6145 before indexing is_end_of_line. Remove redundant check for '\0'.
6147 * config/tc-h8500.c (md_assemble): Likewise.
6148 * config/tc-mcore.c (md_assemble): Likewise.
6149 * config/tc-tic30.c (tic30_find_parallel_insn): Likewise.
6152 * config/tc-m88k.c (s_bss): Cast to unsigned char before indexing
6154 * config/tc-mcore.c (mcore_cons): Likewise.
6155 (mcore_float_cons): Likewise.
6156 (mcore_stringer): Likewise.
6157 * config/tc-tic30.c (tic30_find_parallel_insn): Likewise.
6161 * config/tc-arm.c (do_mrs): Allow SPSR_BIT to be set correctly.
6165 * config/tc-sh.c: Fix compile time warning messages.
6167 * config/tc-mips.c: Fix compile time warning messages.
6171 * doc/as.texinfo: Update copyright dates.
6172 (Local Labels): Delete misplaced mention of ARM.
6173 * NEWS: Mention ARM ELF support.
6177 * config/tc-mn10300.c (md_assemble): Copy size to real_size before
6178 it is modified, and use the real_size to compute the frag address
6179 for dwarf2 line info.
6183 * Makefile.am (DEP, DEP1, dep, dep-in, dep-am): Use a better sed
6184 line-matching scheme to cope with automake moving variables around.
6185 ($(TARG_CPU_O)): Remove dependency on TARG_CPU_DEP_@target_cpu_type@
6186 * Makefile.in: Regenerate.
6190 * config/tc-sparc.c (sparc_relax): New.
6191 (md_longopts): Add -relax and -no-relax options.
6192 (md_parse_options, md_show_usage): Likewise.
6193 (md_apply_fix3): Optimize tail call into branch always if possible.
6197 * config/tc-d10v.c (write_2_short, parallel_ok, md_assemble,
6198 d10v_cleanup) implement Mitsubishi's newly explained branch-packing
6199 rules, with warning when a GAS statement specifies a packing that
6200 will result in an instruction being squashed.
6201 Added typdef packing_type and enumerals, changed various integer
6202 literals to use the enumerals.
6206 * config/tc-ia64.c (dot_restorereg_p): New function.
6207 (md_pseudo_table): Add restorereg.p.
6208 (output_X3_format): Fix typo: record type should be UNW_X3, not UNW_X1.
6209 (output_X4_format): Fix typo: record type should be UNW_X4, not UNW_X2.
6211 * config/tc-ia64.h (unw_record_type): Add unwabi.
6212 (unw_r_record): Rename member MASK to GRMASK. Add sub-structure
6213 called MASK with members for imask, and the masks produced by
6214 fr_mem, gr_mem, br_mem, and frgr_mem.
6215 (unw_p_record): Add members ABI and CONTEXT.
6216 (unw_x_record): Add member AB.
6217 * config/tc-ia64.c (enum reg_symbol): Add REG_PSP and REG_PRIUNAT
6218 as pseudo-register for use during unwind info generation.
6219 (AR_PFS, AR_LC): New macros.
6220 (enum pseudo_type): Add PSEUDO_FUNC_REG to permit declaring registers
6221 whose names start with an at sign (as in "@priunat").
6222 (pseudo_func): Add "svr4", "hpux", "nt" constants and "priunat"
6224 (unwind_list, unwind_tail, current_unwind_entry, proc_start,
6225 proc_end, unwind_info, personality_routine): Consolidate into
6226 "unwind" structure to reduce offset-table use. Add member
6227 NEXT_SLOT_NUMBER to track the slot number for the next instruction
6229 (output_R1_format, output_R3_format, output_P3_format,
6230 output_P6_format): Initialize R with zero to reduce compiler warnings.
6231 (output_P7_format): Ditto. Add `default' branch to switch
6232 statement to reduce compiler warnings.
6233 (output_P8_format, output_B1_format, output_B4_format): Ditto.
6234 (output_P4_format): Rename 2nd & 3rd arg to IMASK and IMASK_SIZE.
6235 (format_ab_reg): Rename from format_a_b_reg. Merge A and B args
6236 into single argument.
6237 (output_X1_format, output_X3_format): Initialize R with zero to reduce
6238 compiler warnings. Merge A and B args into single argument.
6239 (output_X2_format, output_X4_format): Remove unused variable R. Merge
6240 A and B args into single argument.
6241 (free_record): Removed (wasn't used).
6242 (free_list_records): Also free imasks in prologue records.
6243 (output_prologue, output_prologue_gr): Initialize mask bits to zero.
6244 (output_spill_mask): Remove.
6245 (output_unwabi): New function.
6246 (output_epilogue, output_label_state, output_copy_state): Call
6248 (output_spill_psprel, output_spill_sprel, output_spill_psprel_p,
6249 output_spill_sprel_p, output_spill_reg, output_spill_reg_p): Add AB
6251 (process_one_record): New locals FR_MASK and GR_MASK. Ignore
6252 gr_mem, fr_mem, br_mem, and frgr_mem records and instead emit them
6253 as part of handling the prologue records. Emit region's imask if
6254 we have one. Handle unwabi, epilogue, label_state, copy_state,
6255 spill_psprel, spill_sprel, spill_reg, spill_psprel_p,
6256 spill_sprel_p, and spill_reg_p records.
6257 (set_imask, count_bits, slot_index): New function.
6258 (fixup_unw_records): Fix region size computation. Handle
6259 epilogue, spill_reg, spill_sprel, spill_psprel, spill_reg_p,
6260 spill_sprel_p, and spill_psprel_p records. Merge mask bits of
6261 frgr_mem, fr_mem, gr_mem, br_mem on a per-region basis and
6262 set_imask accordingly. Update imask for gr_gr, and br_gr records.
6263 (convert_expr_to_ab_reg, convert_expr_to_xy_reg): New function.
6264 (dot_save): Use manifest constants for applicaton registers.
6265 Handle REG_PR and REG_PRIUNAT.
6266 (dot_restore): Don't just ignore it.
6267 (dot_restorereg): New function..
6268 (generate_unwind_image): Ensure unwind info is a multiple of eight
6269 bytes, not just four bytes.
6270 (dot_handlerdata, dot_unwentry): Demand empty rest of line.
6271 (dot_altrp): Don't just ignore it.
6272 (dot_savemem): New function. Replaces dot_savesp() and
6273 dot_savepsp(). Use manifest constants for applicaton registers.
6274 Handle REG_PR and REG_PRIUNAT.
6275 (dot_savef): Simplify.
6276 (dot_saveb): Support generation of br_gr.
6277 (dot_spillreg, dot_spillmem, dot_spillreg_p, dot_spillmem_p,
6278 dot_label_state, dot_copy_state): New function.
6279 (dot_unwabi): Don't just ignore it.
6280 (md_pseudo_table): Add restorereg, spillreg, spillsp, spillpsp,
6281 spillreg.p, spillsp.p, spillpsp, label_state, copy_state,
6282 unwabi, vframesp, and vframepsp. Fix typo alprp->altrp.
6283 (emit_one_bundle): Set slot number for prologue/body records
6284 *before* emitting the first insn.
6285 (emit_one_bundle): Set UNWIND.NEXT_SLOT_NUMBER.
6286 (md_begin): Declare "psp" pseudo-register.
6287 (md_operand): Handle PSEUDO_FUNC_REG. Fix printing of error message
6288 so we don't get segfault.
6289 (output_psp_sprel): Output sp/psp relative offsets as 4-byte word
6290 counts as required per SW Conventions manual
6291 (output_rp_psprel, output_rp_sprel, output_pfs_psprel,output_pfs_sprel,
6292 output_preds_psprel, output_preds_sprel, output_spill_base,
6293 output_unat_psprel, output_unat_sprel, output_lc_psprel,
6294 output_lc_sprel, output_fpsr_psprel, output_fpsr_sprel,
6295 output_priunat_psprel, output_priunat_sprel, output_bsp_psprel,
6296 output_bsp_sprel, output_bspstore_psprel, output_bspstore_sprel,
6297 output_rnat_psprel, output_rnat_sprel, output_spill_psprel,
6298 output_spill_sprel, output_spill_psprel_p, output_spill_sprel_p):Ditto.
6299 (dot_vframe): Implement.
6300 (dot_vframesp, dot_vframepsp): New function.
6304 * configure.in (i386-*-freebsd a.out entry): Quote properly.
6305 * configure: Regenerate.
6309 * config/tc-i386.c (md_assemble): Pass jump reloc in fr_var...
6310 (md_estimate_size_before_relax): so we can use it here instead of
6311 old kludges. Localise vars to blocks. Comment.
6313 * frags.c (frag_new): Update fr_var comments.
6314 * frags.h (struct frag): Ditto.
6318 * config/tc-ia64.c (FUNC_PC_RELATIVE): New.
6319 (pseudo_func): Add pcrel.
6320 (operand_match): Handle IA64_OPND_TGT64.
6321 (build_insn): Likewise.
6322 (md_begin): Initialize pseudo_func[FUNC_PC_RELATIVE].
6323 (ia64_gen_real_reloc_type): Handle FUNC_PC_RELATIVE.
6324 (fix_insn): Handle all three 64-bit relocation types.
6328 * obj.h (struct format_ops): New members begin, app_file,
6329 s_set_other, s_set_desc, s_get_type, s_set_type,
6330 separate_stab_sections, init_stab_section.
6332 * config/obj-multi.h: Update GPL notice to v2.
6334 (obj_app_file): New.
6335 (S_SET_SIZE): Test s_set_size for NULL before calling.
6336 (S_SET_ALIGN): Similar for s_set_align.
6341 (SEPARATE_STAB_SECTIONS): New.
6342 (INIT_STAB_SECTION): New.
6343 (EMIT_SECTION_SYMBOLS): New.
6344 (AOUT_STABS) [OBJ_MAYBE_AOUT]: Define.
6346 * config/obj-elf.h: Update GPL notice to v2.
6347 Mention that this file is included from obj-multi.h.
6348 (obj_begin): Wrap definition in ifndef.
6349 (elf_file_symbol): Constify declaration.
6350 (obj_app_file): Ditto.
6351 (SEPARATE_STAB_SECTIONS, INIT_STAB_SECTION, OBJ_PROCESS_STAB):
6352 Wrap in ifndef SEPARATE_STAB_SECTIONS.
6354 * config/obj-elf.c (elf_s_set_other): New.
6355 (elf_file_symbol): Constify argument.
6356 (elf_separate_stab_sections): New.
6357 (elf_init_stab_section): New.
6358 (elf_format_ops): Add new members. Remove comma at end.
6360 * config/obj-ecoff.c (ecoff_separate_stab_sections): New.
6361 (ecoff_format_ops): Add new fields. Remove comma at end.
6362 Mention inconsistency for emit_section_symbols.
6364 * config/obj-coff.h (c_dot_file_symbol): Constify declaration.
6366 * config/obj-coff.c (c_dot_file_symbol): Constify argument.
6367 (coff_separate_stab_sections): New.
6368 (coff_format_ops): Add new members.
6370 * config/obj-aout.c (obj_aout_sec_sym_ok_for_reloc): New.
6371 (obj_aout_s_set_other): New.
6372 (obj_aout_s_set_desc): New.
6373 (obj_aout_s_get_type): New.
6374 (obj_aout_s_set_type): New.
6375 (obj_aout_separate_stab_sections): New.
6376 (aout_format_ops): New members added. Use obj_aout_process_stab,
6377 not 0. Use obj_aout_sec_sym_ok_for_reloc, not 0.
6378 (obj_aout_frob_symbol): Add ATTRIBUTE_UNUSED to args as
6380 (obj_aout_line, obj_aout_weak, obj_aout_type): Ditto.
6384 * config/tc-i386.c (tc_i386_fix_adjustable): Prevent adjustment
6385 for OBJ_MAYBE_ELF too. Use S_IS_EXTERNAL instead of S_IS_EXTERN.
6386 (md_estimate_size_before_relax): Ensure jumps to weak and
6387 externally visible symbols are relocatable.
6391 * stabs.c (aout_process_stab): Make global.
6392 (s_desc): Add ATTRIBUTE_UNUSED to args as appropriate.
6393 * read.h (aout_process_stab): Declare.
6395 * configure.in (EMULATIONS) [i386aout, i386coff, i386elf]:
6396 Generalize to *aout, *coff *elf.
6397 * configure: Regenerated.
6399 * doc/internals.texi (Object format backend): Say
6400 SEPARATE_STAB_SECTIONS needs to be nonzero, not just defined.
6402 * Makefile.am (TARG_ENV_HFILES): Delete te-multi.h.
6403 * Makefile.in: Regenerated.
6407 * cgen.h (GAS_CGEN_MAX_FIXUPS): Check if already defined.
6411 * config/tc-hppa.c (md_apply_fix): Mask out immediate bits of
6412 instruction to reflect change in re_assemble_*.
6416 * configure.in (hppa-*-hpux11*): If the cpu is hppa*64*, then
6417 build PA64 ELF tools.
6418 * configure: Rebuilt.
6422 * Makefile.am: Regenerate dependencies.
6423 * Makefile.in: Regenerate.
6427 * config/tc-arm.c (struct asm_psr): Add boolean field
6428 distinguishing between CSPR and SPSR. Rename 'number' field
6430 (psrs): Rearrange contents to match new asm_psr structure.
6431 (arm_psr_parse): Move next to psr_required_here. Make it
6432 return an asm_psr structure.
6433 (psr_required_here): Use asm_psr structure returned by
6435 (do_msr): Reorganise to allow psr_required_here to be called
6437 (md_undefined_name): Mark 'name' parameter as unused, since
6438 the COFF target does not use it.
6442 * config/te-386bsd.h: Clean up comments to adhere to the GNU coding
6444 * config/te-aux.h: Likewise.
6445 * config/te-dpx2.h: Likewise.
6446 * config/te-go32.h: Likewise.
6447 * config/te-hp300.h: Likewise.
6448 * config/te-hppa.h: Likewise.
6449 * config/te-i386aix.h: Likewise.
6450 * config/te-ic960.h: Likewise.
6451 * config/te-interix.h: Likewise.
6452 * config/te-nbsd532.h: Likewise.
6453 * config/te-pc532mach.h: Likewise.
6454 * config/te-ppcnw.h: Likewise.
6455 * config/te-psos.h: Likewise.
6456 * config/te-sparcaout.h: Likewise.
6457 * config/te-sun3.h: Likewise.
6458 * config/te-sysv32.h: Likewise.
6462 * config/tc-h8300.c (do_a_fix_imm): Don't rely on `short' being 16
6463 bits. Instead explicitly mask and sign extend. Do the 8 bit mask
6464 and sign extend without an if statement.
6465 (build_bytes): Likewise.
6469 * config/tc-h8300.c (do_a_fix_imm): Output a reloc for no
6470 X_add_symbol L_32 case.
6474 * config/te-freebsd.h: New file.
6478 * asintl.h (gettext, dgettext, dcgettext, textdomain,
6479 bindtextdomain): Replace defines with those from intl/libgettext.h
6480 to quieten gcc warnings.
6482 * NEWS: Mention x86 .arch and -q.
6484 * config/tc-i386.c (quiet_warnings): New.
6485 (md_assemble): Use quiet_warnings.
6486 (md_parse_option): Set quiet_warnings from -q.
6487 (md_show_usage): Mention -q, delete -m.
6488 (flag_do_long_jump): Delete.
6489 (md_parse_option): Remove -m.
6490 (md_show_usage): Remove -m.
6491 (md_create_long_jump): Remove useless flag_do_long_jump code.
6493 * as.c (parse_args): In case OPTION_DEFSYM, use a valueT to hold
6494 the symbol value, and use bfd_scan_vma if BFD_ASSEMBLER.
6499 * doc/c-i386.texi (i386-Arch): New section.
6500 (i386-Syntax): Mention .intel_syntax and .att_syntax.
6502 * config/tc-i386.c (cpu_arch_name, cpu_arch_flags): New.
6503 (smallest_imm_type): Use smallest opcode for shift by one if cpu
6504 architecture has been given and is not 486.
6505 (set_cpu_arch): New.
6506 (md_pseudo_table): Add .arch.
6507 (md_assemble): Warn if cpu architecture has been given and an
6508 unsupported instruction.
6510 * config/tc-i386.h (SMALLEST_DISP_TYPE): Delete.
6511 Move operand_types bit defines after relevant template field.
6512 (template): Add cpu_flags.
6518 * config/tc-mn10300.h (md_end): Define.
6519 (mn10300_finalize): Declare.
6520 * config/tc-mn10300.c: Include dwarf2dbg.h.
6521 (debug_line): Define.
6522 (md_assemble): Generate dwarf2 line info.
6523 (mn10300_finalize): New function. Finalize dwarf2 info.
6527 * config/tc-mips.c (md_estimate_size_before_relax): Use the
6528 external version of the relocation for weak symbols.
6532 * config/tc-ia64.c (output_P7_format, case mem_stack_f): Output fixed
6533 frame size in units of 16 bytes, as required per SW Conventions manual.
6534 (output_unw_records): Output info-block header as a dword to get
6539 * as.h: #include "file", not <file> on files from ../include.
6540 (as_abort, as_fatal): Add ATTRIBUTE_NORETURN.
6541 * config/tc-m68k.c (m68k_ip): Fix signed/unsigned warnings.
6542 (md_convert_frag): Add ATTRIBUTE_UNUSED.
6543 (tc_coff_symbol_emit_hook): Ditto.
6544 (OPTCOUNT): Cast to int to avoid compiler warning.
6545 (md_begin): Fix signed/unsigned warnings.
6549 * config/tc-m68k.c (md_convert_frag_1): Abort if we end up in the
6550 ABRANCH LONG case for a conditional branch on a 68000.
6551 (md_estimate_size_before_relax): Likewise. Also handle
6552 flag_short_refs correctly for ABRANCH, BCC68000, and DBCC.
6553 (m68k-ip: case ABSL): Relax absolute references to 16-bit
6554 PC-relative on all CPUs.
6555 (md_estimate_size_before_relax): Likewise.
6559 * as.c (parse_args): Just mention current year in printed
6564 * config/tc-ppc.c (pre_defined_registers): Add entries for vector
6566 (md_parse_option): Recognize -m7400.
6570 * config/atof-ieee.c (gen_to_words): When adding carry back in,
6571 don't permit lp to become less than the words array.
6575 config/tc-mcore.c (md_apply_fix3): BFD_RELOC_MCORE_PCREL_IMM11BY2
6576 Fix little-endian case.
6580 * as.c (parse_args): Update copyright.
6584 * config/tc-i386.h (SUB_SEGMENT_ALIGN): If TE_GO32, return 4
6585 for the .bss section too.
6589 * configure.in: Set em=linux for hppa-*-linux.
6590 * configure: Regenerate.
6591 * doc/Makefile.in: Regenerate with correct automake.
6593 * frags.c (frag_grow): Sanity check chunk_size.
6595 * config/obj-elf.h: #include "bfd.h" not <bfd.h>
6596 * config/obj-som.h: Likewise.
6597 * config/obj-ieee.h: Likewise.
6599 * config/tc-hppa.h: Test BFD_ARCH_SIZE, not BFD64.
6601 * config/tc-hppa.c (log2): Only compile when OBJ_SOM.
6602 (md_pseudo_table): Fully initialise OBJ_ELF cases.
6603 (fix_new_hppa): Add ATTRIBUTE_UNUSED to args as appropriate.
6604 (pa_ip): low_sign_unext now returns via function value. Use
6605 re_assemble_* instead of dis_assemble_* and
6606 INSERT_FIELD_AND_CONTINUE combination. Don't call sign_unext
6608 (md_convert_frag): Add ATTRIBUTE_UNUSED to args as appropriate.
6609 (md_section_align, md_parse_option, md_show_usage,
6610 md_undefined_symbol, pa_align, pa_block, pa_brtab, pa_try,
6611 pa_callinfo, pa_code, pa_comm, pa_end, pa_enter, pa_entry,
6612 pa_exit, pa_export, pa_import, pa_label, pa_leave, pa_level,
6613 pa_origin, pa_param, pa_proc, pa_procend, pa_space, pa_spnum,
6614 pa_version, pa_compiler, pa_copyright, pa_data, pa_fill, pa_lsym,
6616 (md_apply_fix): Change type of new_val to offsetT. Delete w1, w2,
6617 w, resulti. Add insn, val. Move bfd_get_32 and bfd_put_32
6618 outside of switch. Correct mask and shifting errors in case 10
6619 and case -11. In case 21, compare against signed range to suit
6620 hppa_field_adjust changes. In case 12, use re_assemble_12. In
6621 case 17 and case 22, use offsetT variable to properly check range.
6622 Use re_assemble_* here too.
6623 (evaluate_absolute): Change type of value to offsetT. Call
6624 hppa_field_adjust to do the work for us.
6625 (pa_parse_cmpb_64_cmpltr): Delete save_s.
6626 (pa_parse_cmpib_64_cmpltr): Ditto.
6627 (pa_build_unwind_subspace): Delete unused var subseg. Change type
6628 of i to unsigned int.
6629 (pa_type_args): Conditionally declare symbol if OBJ_SOM.
6630 (pa_end_of_source): Return type is void.
6634 * macro.c (macro_expand_body): Don't prepend macro number with zeroes.
6638 * config/tc-avr.c: ATTRIBUTE_UNUSED added to the necessary places.
6639 More comments added.
6640 (md_begin): Removed "construct symbols for each register name".
6641 Because register names conflicts with GCC generated function
6643 (avr_operand): Now constant numbers can be used as a register
6644 identifiers (0 as r0, 31 as r31).
6645 (md_assemble): use skip_space () before parsing instruction
6650 * configure.in: Set bfd_gas=yes on i386-*-pe and i386-*-nt* to
6651 ensure all pe targets use bfd. Remove unnecessary bfd_gas=yes on
6652 arm-*-netbsd* and arm-*-wince as this is set for all arm*.
6653 * configure: Regenerate.
6657 * as.h: Correctly check GCC version.
6661 * doc/as.1: Fix unbalanced brackets.
6663 * config/tc-i386.c (comment_chars): Don't use '/' as comment start if
6665 (line_comment_chars): Set to '/' if TE_FreeBSD.
6669 * configure.in: Configury support for PA64 (currently disabled).
6670 * configure: Rebuilt.
6674 * config/tc-mips.c (s_change_sec): Use record_alignment, not
6675 bfd_set_section_alignment.
6679 * config/tc-i386.c (offset_in_range): Ensure shift counts are less
6684 * doc/c-arm.texi (ARM Directives): Document behaviour of .align 0.
6685 * doc/as.texinfo (Align): Include arm and strongarm in list of
6686 targets that have the second form of the behaviour of the .align
6691 * config/obj-som.c: Terminate obj_pseudo_table.
6695 * as.c (parse_args): Allow md_parse_option to override -a listing
6697 * config/obj-coff.c (add_lineno): Change type of offset parameter
6698 from "int" to "bfd_vma."
6699 * config/tc-ppc.c (md_pseudo_table): Add "llong" and "machine."
6700 (ppc_mach, ppc_subseg_align, ppc_target_format): New.
6701 (ppc_change_csect): Align correctly for XCOFF64.
6702 (ppc_machine): New function, which discards "ppc_machine" line.
6703 (ppc_tc): Cons for 8 when code is 64 bit.
6704 (md_apply_fix3): Don't check operand->insert. Handle 64 bit
6706 (md_parse_option): Handle -a64 and -a32.
6708 * config/tc-ppc.h (TARGET_MACH): Define.
6709 (TARGET_FORMAT): Move to function.
6710 (SUB_SEGMENT_ALIGN): Use ppc_subseg_align.
6714 * config/tc-avr.c: New AVR_ISA_ defined.
6715 (md_assemble): Handle opcodes with optional operands (lpm,elpm).
6716 (avr_operand): Handle 'a', 'v' and 'z' constraint letters needed
6717 for `fmul', `movw' and `lpm R,Z' instructions.
6718 (avr_operands): Warn if current opcode is a two-word instruction
6719 and previous opcode was cpse/sbic/sbis/sbrc/sbrs.
6720 (avr_opcodes): New commands added.
6721 (REGISTER_P): Check 'a' and 'v' constraint letters.
6722 (mcu_types): New MCU added.
6726 * config/tc-ia64.c (pseudo_func[]): Add new "nat" entry equivalent
6728 (operand_match): Conditionally insert default bit values for IMMU9.
6732 * configure.in: Add NetBSD/sparc ELF and NetBSD/sparc64 support.
6733 * configure: Rebuilt.
6738 * config/tc-hppa.c (md_apply_fix): Handle new PA2.0 formats.
6740 * config/tc-hppa.c (CHECK_ALIGN): New macro.
6741 Added handling of new operand types l,y,&,fe,fE,fx.
6749 * Makefile.am (CPU_TYPES): Add ia64.
6750 (TARGET_CPU_CFILES): Add config/tc-ia64.c.
6751 (TARGET_CPU_HFILES): Add config/tc-ia64.h.
6752 * Makefile.in: Rebuild.
6753 * app.c (do_scrub_chars): Handle DOUBLESLASH_COMMENTS.
6754 * configure: Rebuild.
6755 * configure.in: Recognize ia64 as cpu type. Set bfd_gas.
6756 (ia64-*-elf*, ia64-*-linux-gnu*): New targets.
6757 * expr.c (expr): Handle md_optimize_expr.
6758 * read.c (LEX_HASH): Add comment.
6759 * config/tc-ia64.c, config/tc-ia64.h: New files.
6763 * config/tc-d30v.c (write_2_short): Disregard opcode1->ecc when
6764 bundling a non-delayed branch type instruction.
6768 * config/tc-mn10300.c (HAVE_AM30): Define.
6769 (md_assemble): Use it.
6773 * config/obj-elf.c (obj_elf_change_section): Check for changed
6776 * Makefile.am: (CPU_MULTI_VALID): Remove.
6777 (MULTI_CPU_TYPES): Define.
6778 (MULTI_CPU_OBJ_VALID): Define.
6779 (DEPTC): Use the above.
6780 (DEPOBJ): Same here.
6782 Regenerate dependencies.
6783 * Makefile.in: Regenerate.
6787 * Makefile.am (YACC, LEX): Get them from configure.
6791 * config/tc-i386.c (offset_in_range): Use addressT instead of
6792 bfd_vma for non-bfd assemblers.
6796 * config/tc-i386.c (offset_in_range): Sign extend val so BFD64
6797 doesn't give spurious errors.
6801 * as.h (SEEK_SET): Define if undefined.
6805 * config/tc-arm.c (md_apply_fix3): Don't use UL suffix on
6806 constants, and don't assume offsetT is 32 bits.
6810 * config/tc-d10v.h: Include "write.h" to get definition of fixS.
6811 (md_pcrel_from_section): Add prototype.
6812 (d10v_fix_adjustable): Add prototype.
6813 (d10v_force_relocation): Replace 'struct fix' with 'fixS'.
6815 * config/tc-d10v.c (md_apply_fix3): Add paren around &&.
6819 * config/tc-d10v.c (find_opcode): Correctly calculate position of
6820 symbol in frag chain.
6824 * config/tc-i386.c (fits_in_signed_byte): Change arg to offsetT.
6825 (fits_in_unsigned_byte, fits_in_unsigned_word): Ditto.
6826 (fits_in_signed_word, smallest_imm_type): Ditto.
6827 (md_assemble): Use an offsetT var to hold offsetT values, not a
6829 (offset_in_range): New.
6830 (md_assemble): Use it.
6831 (md_convert_frag): Change type of target_address, opcode_address,
6832 and displacement_from_opcode_start to offsetT.
6833 (md_create_short_jump): Change type of offset to offsetT.
6834 (md_create_long_jump): Ditto.
6835 (md_apply_fix3): Use -4, not 0xfffffffc for BFD_RELOC_386_PLT32.
6836 (md_chars_to_number): Remove.
6837 (output_invalid): Remove duplicate prototype.
6841 * Makefile.am (CPU_TYPES): Add 'avr'.
6842 (TARGET_CPU_CFILES): Add 'tc-avr.c'.
6843 (TARGET_CPU_HFILES): Add 'tc-avr.h'.
6845 * Makefile.in: Regenerate.
6847 * doc/as.texinfo: Add M32R documentation.
6851 * configure.in (WARN_CFLAGS): Set to -W -Wall by default. Add
6852 --enable-build-warnings option.
6853 * Makefile.am (AM_CFLAGS, WARN_CFLAGS): Add definitions.
6854 * Makefile.in, configure: Re-generate.
6858 * config/tc-arm.c (md_apply_fix3): Treat BFD_RELOC_ARM_PCREL_BLX
6859 in the same way as BFD_RELOC_ARM_PCREL_BRANCH, and
6860 BFD_RELOC_THUMB_PCREL_BLX lie BFD_RELOC_THUMB_PCREL_BRANCH.
6861 (tc_gen_reloc): Accept BFD_RELOC_{ARM|THUMB}_PCREL_BLX.
6862 (arm_force_relocation): Force relocations for
6863 BFD_RELOC_{ARM|THUMB}_PCREL_BLX as well.
6867 * config/tc-sh.c (get_operands): There's no third operand if the
6868 first operand is an immediate.
6872 * config/tc-sh.c (immediate): Delete.
6873 (sh_operand_info): Add immediate member.
6874 (parse_reg): Use A_PC for pc.
6875 (parse_exp): Add second argument 'op'. All callers changed.
6876 (parse_at): Expect pc to be coded as A_PC.
6877 Use immediate field in *op.
6878 (insert): Add fourth argument 'op'. All callers changed.
6879 (build_relax): Add second argument 'op'. All callers changed.
6880 (insert_loop_bounds): New function.
6881 (build_Mytes): Remove DISP_4.
6882 Split IMM_[48]{,BY[24]} into IMM[01]_[48]{,BY[24]}. Add REPEAT.
6883 (assemble_ppi): Use immediate field in *operand.
6884 (sh_force_relocation): Handle BFD_RELOC_SH_LOOP_{START,END}.
6885 (md_apply_fix): Likewise.
6886 (tc_gen_reloc): Likewise. Check for a pcrel BFD_RELOC_SH_LABEL.
6890 * config/tc-sparc.c (sparc_ip): Avoid string pasting.
6894 * internals.texi (CPU backend): Document
6895 TC_CHECK_ADJUSTED_BROKEN_DOT_WORD.
6899 * po/gas.pot: Regenerate.
6901 * as.c (show_usage): Restore translated part of bug string.
6902 * gasp.c (show_usage): Likewise.
6904 * Makefile.am (MKDEP): Use gcc -MM rather than ../mkdep.
6905 (DEP): Quote when passing vars to sub-make. Use "mv -f" rather
6906 than move-if-change.
6907 (DEP1): Modify for "gcc -MM".
6911 (CLEANFILES): Add DEPTCA, DEPOBJA, DEP2a, DEPA.
6912 Update dependencies.
6913 * Makefile.in: Regenerate.
6917 * config/tc-mn10300.c (md_pseudo_table): Use constant names.
6918 (md_begin): Likewise.
6919 (HAVE_AM33): New macro.
6920 (md_assemble): Use it. Match r_regs and xr_regs only if
6925 * as.h: #include "bin-bugs.h"
6926 * as.c (show_usage): Use REPORT_BUGS_TO.
6927 * gasp.c: #include "bin-bugs.h"
6928 (show_usage): Use REPORT_BUGS_TO.
6930 * config/tc-sparc.c (md_show_usage): Add a trailing newline.
6934 * write.c (write_object_file) [! WORKING_DOT_WORD]: If defined,
6935 call TC_CHECK_ADJUSTED_BROKEN_DOT_WORD for each word after
6940 * config/tc-i386.c (i386_immediate): Don't assume a constant
6941 immediate is necessarily 16 bits when in 16 bit code mode.
6942 (md_assemble): Instead set guess_suffix here after we have checked
6947 * config/tc-d30v.c (check_range): Allow signed or unsigned 32-bit
6948 quantities. Correct right shift sign extension.
6949 (build_insn): Make `number' unsigned long. Mask top 6 bits of
6950 32-bit value when shifting into place.
6954 * app.c: Add ATTRIBUTE_UNUSED as needed.
6955 * config/tc-ppc.c: Likewise.
6956 (ppc_size): Make unsigned long.
6957 (ppc_insert_operand): Add casts to avoid warnings.
6961 * config/tc-d10v.h (md_flush_pending_output): Define.
6965 * config/tc-sh.h (SEG_NAME): New macro: return the name of a
6966 segment. Works for both BFD_ASSEMBLER and others.
6967 (SUB_SEGMENT_ALIGN): Use SEG_NAME.
6971 * config/tc-arm.c (tinsns): Add "bal" instruction pattern.
6975 * listing.c (LISTING_LHS_WIDTH): Default depends on
6977 (LISTING_LHS_WIDTH_SECOND): Default to LISTING_LHS_WIDTH.
6981 * config/tc-sh.c (md_show_usage): Use backslash before newline in
6986 * config/tc-avr.h (TC_HANDLES_FX_DONE): Define.
6988 * config/tc-avr.c (mcu_types): Add missing initialiser.
6989 (md_pcrel_from_section): Add prototype.
6990 (avr_operand): Remove redundant test of unsigned < 0.
6991 (avr_cons_fix_new): Ensure exp_mod_pm zero on function exit.
6995 * config/tc-avr.c: New file for AVR support.
6996 * config/tc-avr.h: Likewise.
6997 * configure.in: Add AVR support.
6998 * configure: Regenerate.
7002 * gasp.c (macro_op): Add new argument to check_macro call.
7003 Macro structure definitions moved to macro.h
7004 * sb.h: Add argument to prototype for input_scrub_include_sb.
7005 * input-scrub.c (input_scrub_include_sb): Allow disabling of sb
7006 nesting checks with an additional flag.
7007 (struct input_save): Add flag to indicate whether current sb
7008 should be checked for proper macro/conditional nesting.
7009 (input_scrub_push/pop): Save/restore nest check flag.
7010 (input_scrub_next_buffer): Ditto. Also call end of macro hook if
7012 * macro.c (check_macro): Allow caller to retrieve parsed macro
7013 information if a pointer is provided. This information may be
7014 used by the new macro hooks.
7015 * macro.h: Update prototype for check_macro. Macro struct
7016 definitions moved here from macro.c/gasp.c.
7017 * read.c (read_a_source_file): Add parameter to check_macro call,
7018 and pass macro info to the macro hook, if defined.
7019 (input_scrub_insert_line): New. Allow insertion of a line of
7020 characters into the input stream.
7021 (input_scrub_insert_file): New. Allow insertion of an arbitrary
7022 file into the input stream.
7023 (s_include): Use input_scrub_insert_file.
7024 * internals.texi: Document new macro hooks.
7025 * as.h: New prototypes added.
7029 * config/tc-i386.c: Don't start any as_bad or as_warn message with
7030 an initial capital letter.
7031 (i386_index_check): Reindent.
7035 * config/tc-arm.c (md_apply_fix3): Fix bug detecting overflow of pc
7040 * config/tc-arm.c (do_t_adr): Flag "adr Rd,label"
7041 instruction operand bad if Rd > 7 when generating
7042 thumb instructions. Prevents for example,
7043 "adr r12,label" from silently failing and generating
7044 the wrong instruction.
7048 * config/tc-arm.c (md_apply_fix3): Handle same-section relocations
7049 that have a destingation >= 0x400000.
7050 Fix compile time warning messages.
7054 * config/tc-sh.c (md_begin): When encountering insn that are
7055 not supported by the current arch, only change the name if
7056 its contents are the same as prev_name.
7057 (get_specific): If the the architecture doesn't match, fail.
7061 * config/tc-sh.c (IDENT_CHAR): Define.
7062 (parse_reg): Use it instead of isalnum. Put r[0..7]_bank operand
7063 matching back where it came from.
7067 * config/tc-sh.c (md_show_usage): Add description of -dsp.
7071 * config/tc-sh.c (parse_reg): Match r[0..7]_bank operands before
7076 * config/tc-h8300.c: Add ATTRIBUTE_UNUSED as appropriate.
7080 * expr.c (operand) [case 'f']: When testing if '0f' can start a
7081 floating-point-number, make sure 'f' is in FLT_CHARS.
7085 * read.c (TC_IMPLICIT_LCOMM_ALIGNMENT): New default-definition.
7086 (s_lcomm_internal): Use it.
7087 * doc/internals.texi (CPU backend): Document it.
7088 * config/obj-evax.h (TC_IMPLICIT_LCOMM_ALIGNMENT): Set to 2**3
7093 * config/tc-mips.c (mips_ip): Don't put stuff in .rodata
7096 * config/tc-mips.c (SWITCH_TABLE): The ELF embedded-pic
7097 implementation doesn't have special handling for switch
7099 (macro_build): Allow for code in sections other than .text.
7101 (mips_ip): Likewise.
7102 (md_apply_fix): Do pc-relative relocation madness for MIPS ELF.
7103 Don't perform relocs if we will be outputting them.
7104 (tc_gen_reloc): For ELF, just use fx_addnumber for pc-relative
7105 relocations. Allow BFD_RELOC_16_PCREL_S2 relocs when
7110 * config/tc-m32r.c (m32r_fix_adjustable): Look up the
7111 relocation type based on the entry in the fixup structure.
7112 Put S_IS_EXTERN processing back in.
7116 * Makefile.am (install-exec-tooldir): Depend on
7117 install-exec-bindir for parallel make.
7118 * Makefile.in: Regenerated.
7122 * config/tc-m32r.c (struct md_longopts): Add -m32r command line
7124 (md_parse_option): Parse -m32r command line switch - disable m32rx
7126 (md_show_usage): Document new option.
7128 * doc/Makefile.am (CPU_DOCS): Add c-m32r.texi.
7129 * doc/Makefile.in: Regenerate.
7130 * doc/c-m32r.texi (M32R-Opts): Document new command line switch.
7134 * config/tc-d30v.c (check_range): Remove code that incorrectly
7135 sign extended values where bits < 32.
7139 * configure.in: Support --enable-targets=all on ia32.
7140 * configure: Regenerated.
7144 * gasp.c (do_align): Remove bogus check of alignment value.
7148 * config/obj-elf.c (elf_frob_symbol): Remove code which when
7149 TC_PPC was defined forced the type of a symbol with no other type
7154 * doc/internals.texi (CPU backend): Mention that
7155 line_separator_chars do not break up comments. Fix typos for
7156 LEX_AT and LEX_NAME descriptions. Document operands for
7157 TC_EQUAL_IN_INSN, md_operand and md_section_align. Correct
7158 description of md_create_short_jump usage. Document argument for
7159 md_undefined_symbol.
7163 * config/tc-sparc.c (OPTION_UNDECLARED_REGS): New option.
7164 (md_parse_option): Handle it.
7165 (md_show_usage): Document it.
7169 * config/tc-alpha.c (md_assemble): Accept `1' and `9' in an
7170 opcode, for the instruction `pal19'. From Andrea Arcangeli
7175 * config/tc-i386.c (i386_immediate): Move constant operand sizing
7177 (md_assemble): To here, before template operands are matched.
7178 Also ensure a constant immediate is sign extended when we know the
7179 size is at most 16 bits. This is to catch cases like "add
7180 $0xffc0,%ax" where we don't know the size, and thus that the
7181 immediate can be represented as Imm8S until after parsing the
7183 (i386_displacement): Similarly sign extend 16 bit constant
7185 (md_assemble): Relax 16-bit jump constant range check to suit sign
7186 extended displacements.
7190 * doc/c-mips.texi (MIPS Opts): Fix typo in last patch.
7194 * config/tc-i386.c (md_assemble): Don't swap intersegment jmp and
7195 call operands when intel_syntax.
7196 (intel_float_operand): Return 2 for "fi...".
7197 (i386_operand_modifier): Change "DWORD PTR" test to suit above.
7198 Return SHORT_MNEM_SUFFIX for "WORD PTR" when "fi...". Revert
7199 earlier "SHORT" change.
7200 (md_assemble): When determining suffix from Regs, exclude
7205 * configure: Add arm-wince, mips-pe and sh-pe targets.
7206 * configure: Regenerate.
7208 * config/obj-coff.h (COFF_WITH_PE): Define for mips-pe and
7210 (TARGET_FORMAT): Set to "pe-shl" for the sh-pe target and to
7211 "pe-mips" for the mips-pe target.
7213 * config/tc-arm.c (insns): Change displacement encoded in BL
7214 and B instructions if the target port is arm-wince.
7215 (do_ldst): Do not bias the relocation offset if the target
7217 (md_pcrel_from): Add in missing relocation offset bias if the
7218 target os arm-wince.
7220 * config/tc-mips.c (mips_target_format): Support COFF flavour.
7221 (md_begin): Disable -G support for mips-pe target.
7222 (md_apply_fix): Treat BFD_RELOC_RVA reloc as BFD_RELOC_32.
7223 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Add support for
7226 * config/tc-sh.c (md_begin): sh-pe target is little endian.
7227 * config/tc-sh.h (SUB_SEGMENT_ALIGN): If using a BFD
7228 assembler, just set the alignment to 4.
7230 * config/te-wince-pe.h: New file for WinCE targets. Define
7235 * config/tc-i386.c (md_assemble): Swap segments too for intel mode
7236 string instructions.
7237 (i386_operand_modifier): Set i.suffix = WORD_MNEM_SUFFIX for SHORT.
7238 (i386_intel_memory_operand): After finding a segment override,
7239 check again for no `[' before looking for a displacement. Bomb if
7240 more than one displacement rather than silently discarding the
7241 second and subsequent ones. Free strings malloc'd by
7242 build_displacement_string.
7246 * config/obj-som.c (obj_pseudo_table): Add "weak".
7247 (obj_som_weak): New routine.
7251 * config/tc-i386.c (union i386_op): New.
7252 (struct _i386_insn): Delete disps[], imms[], regs[]. Add op[].
7253 Throughout file replace occurences of disps[n], imms[n], regs[n]
7254 with equivalent op[n].disps, op[n].imms, op[n].regs. Simplify
7255 intel mode operand swapping. Add assert in regKludge and
7256 fake_zero_displacement code. Test i.types[n] when outputting
7257 displacements and immediates. Combine output of Disp16 with
7259 (md_assemble): Don't try to fix broken UNIXWARE_COMPAT opcodes
7260 when in intel mode by (not) reversing fsub and fdiv operands
7261 before the template search. This fails for single operand
7262 shorthand forms of the instruction, and if UNIXWARE_COMPAT is
7263 undefined. Instead fix the base_opcode after we've found the
7264 template. Move base_opcode xor with found_reverse_match from
7265 opcode output code to before this fix so we test for the correct
7267 (md_assemble): Don't use strcmp when deciding to ignore the suffix
7268 check in intel mode. Instead compare opcodes.
7270 * config/tc-i386.h (TC_RELOC): Delete.
7271 * config/tc-i386.c (TC_RELOC): Delete. Replace usage of TC_RELOC
7272 with equivalent call to reloc.
7274 * as.h (flag_m68k_mri): Move declaration after target include, and
7275 only declare when TC_M68K defined. Define as zero otherwise.
7276 (LABELS_WITHOUT_COLONS, NO_PSEUDO_DOT): If undefined, define as 0.
7277 * app.c (scrub_m68k_mri): Declare only when TC_M68K defined.
7278 Define as zero otherwise.
7279 (do_scrub_begin): Use m68k_mri parameter only when TC_M68K defined.
7280 (struct app_save): Declare scrub_m68k_mri only when TC_M68K.
7281 (app_push, app_pop): Save scrub_m68k_mri only when TC_M68K.
7282 (do_scrub_chars): Use LABELS_WITHOUT_COLONS directly rather than
7283 testing whether defined.
7284 * cond.c (ignore_input): Use NO_PSEUDO_DOT directly.
7285 * expr.c (operand): #ifdef unused case labels when TC_M68K undefined.
7286 * read.c: Use LABELS_WITHOUT_COLONS and NO_PSEUDO_DOT directly
7287 rather than testing whether defined.
7288 (s_mri): Set flag_m68k_mri only when TC_M68K defined.
7289 (parse_mri_cons): Declare and use only when TC_M68K.
7290 * config/tc-hppa.h (LABELS_WITHOUT_COLONS): Define as 1.
7291 * config/tc-m68k.h (NO_PSEUDO_DOT): Define as 1.
7292 * config/tc-m88k.h (NO_PSEUDO_DOT): Define as 1.
7294 * NEWS: Mention IBM 370 support.
7298 * config/tc-i386.c (md_assemble): When swapping operands for
7299 intel_syntax, assume everything that's not Imm or Disp is a
7304 * config/tc-i370.c, config/tc-i370.h: New files.
7305 * Makefile.am: Add support for Linux/IBM 370.
7306 * configure.in: Likewise.
7307 * app.c (do_scrub_begin): Don't lex single quote when TC_I370.
7308 * config/obj-elf.c: Include elf/i370.h
7309 (obj_elf_section): Don't do anything special for flag_mri if TC_I370.
7311 * Makefile.in: Regenerate.
7312 * configure: Regenerate.
7314 * doc/c-i370.texi: New file.
7315 * doc/all.texi: Include it.
7316 * doc/as.texinfo: And here.
7317 * doc/Makefile.am(CPU_DOCS): Add c-i370.texi.
7318 * doc/Makefile.in: Regenerate.
7322 * config/tc-d30v.c (parallel_ok): Use FLAG_NOT_WITH_ADDSUBppp to
7323 determine if an instruction can be used in parallel with an ADDppp
7324 or SUBppp instruction.
7328 * doc/c-mips.texi (MIPS Opts): Document -mgp32 and -mgp64.
7332 * config/tc-mips.c (mips_gp32): New variable.
7333 (macro_build) Use mips_gp32.
7335 (md_longopts): Add "-mgp32" and "-mgp64".
7336 (md_parse_option): Add OPTION_GP32 and OPTION_GP64.
7338 (OPTION_M7900): Change offset
7339 (OPTION_NO_M7900): Ditto.
7343 * config/obj-coff.c (add_lineno): Accept non-positive lineno with
7344 warning, and bump it to 1.
7349 * dwarf2dbg.c (print_stats): Add cast to force printf argument to
7354 * config/tc-mips.c (MF_HILO_INSN): Define.
7355 (mips_7000_hilo_fix): Declare.
7356 (append_insn): Conditionally insert nops after an mfhi/mflo insn.
7357 (md_parse_option): Check for 7000_HILO_FIX options.
7358 (OPTION_M7000_HILO_FIX): Define.
7359 (OPTION_NO_M7000_HILO_FIX): Define.
7360 * doc/c-mips.texi (-mfix7000): Describe.
7364 * listing.c (print_lines): Remove unused variable `end'.
7366 * config/tc-i386.c (md_assemble): Use `reloc()' to select reloc
7367 type for JumpInterSegment output. Use enum bfd_reloc_code_real for
7368 reloc_type when BFD_ASSEMBLER.
7369 (md_estimate_size_before_relax): Use enum bfd_reloc_code_real for
7370 reloc_type when BFD_ASSEMBLER. Move common code out of switch
7371 statement and quell signed vs. unsigned comparison warning.
7375 * config/tc-d10v.c (find_opcode): Add a symbol's value to
7376 the computed frag offset, rather than overwriting it.
7380 * config/tc-sh.c ("elf/sh.h"): Include.
7381 (sh_dsp, valid_arch, reg_x, reg_y, reg_efg): New static variables.
7382 (md.begin): Initialize target_arch.
7383 Only include opcodes in has table that match selected architecture.
7384 (parse_reg): Recognize register names for sh-dsp.
7385 (parse_at): Recognize post-modify addressing.
7386 (get_operands): The leading space is now optional.
7387 (get_specific): Remove FDREG_N support. Add support for sh-dsp
7388 arguments. Update valid_arch.
7389 (build_Mytes): Add support for SDT_REG_N.
7390 (find_cooked_opcode): New function, broken out of md_assemble.
7391 (assemble_ppi, sh_elf_final_processing): New functions.
7392 (md_assemble): Use find_cooked_opcode and assemble_ppi.
7393 (md_longopts, md_parse_option): New option: -dsp.
7394 * config/tc-sh.h (elf_tc_final_processing): Define.
7395 (sh_elf_final_processing): Declare.
7399 * config/tc-hppa.c (pa_build_unwind_subspace): Use subseg_new to create
7400 the unwinder subspace. Save the current seg/subseg before creating
7405 * config/tc-mcore.c (INST_BYTE0): Redefine to handle big and
7406 little endian targets.
7407 (INST_BYTE1): Redefine to handle big and little endian
7409 (cpu_type): New type: Select between M340 and M210.
7410 (parse_psrmod): New function: Parse the PSRCLR and PSRSET
7411 instructions of the M340.
7412 (md_assemble): Add support for the MULSH and OPSR classes of
7414 (md_atof): Add support for little endian targets.
7415 (md_parse_option): Add support for -EL, -EB and -mcpu command
7417 (md_convert_frag): Add support for little endian targets.
7418 (md_apply_fix3): Add support for little endian targets.
7419 (md_number_to_chars): Add support for little endian targets.
7423 * read.c (read_a_source_file): If TC_START_LABEL_WITHOUT_COLON is
7424 defined, use it to verify the symbol just read should be a label.
7428 * app.c (do_scrub_chars): Handle "||" for parallel instructions
7429 when DOUBLEBAR_PARALLEL is defined. Avoid stripping whitespace
7430 around colons when KEEP_WHITE_AROUND_COLON is defined.
7431 * doc/internals.texi (CPU backend): Document DOUBLEBAR_PARALLEL
7432 and KEEP_WHITE_AROUND_COLON.
7436 * read.c (s_rept): Call do_repeat, which abstracts the repeat
7438 (do_repeat): New. Abstract repeat logic so that a "break" can be
7440 (end_repeat): New. Provide support for a "break" out of the
7442 * read.h: Add prototypes for new functions.
7446 * doc/internals.texi: Document NUMBERS_WITH_SUFFIX macro.
7447 * as.h: Provide a default NUMBERS_WITH_SUFFIX definition (zero).
7448 * expr.c: Handle numbers with suffixes if NUMBERS_WITH_SUFFIX is
7453 * read.c: Added elseif to directives table.
7454 * read.h: Added prototype for s_elseif.
7455 * doc/as.texinfo: Added description for elseif.
7456 * cond.c (s_elseif): New function
7460 * listing.c (print_lines): Remove conditionals causing bug in
7465 * as.h: Define OCTETS_PER_BYTE and OCTETS_PER_BYTE_POWER
7467 * frags.c (frag_new): Calculate fr_fix in octets
7468 (frag_now_fix) Return offset as target address offset (bytes).
7469 (frag_now_fix_octets) New - Return offset in octets (8-bit
7471 * frags.h: Added prototype for frag_now_fix_octets().
7472 Distinguish between octets and bytes in field descriptions.
7473 * listing.c (calc_hex): Account for octets vs bytes when
7474 printing addresses/offsets.
7475 (print_lines) Ditto. Also, if LISTING_WORD_SIZE is not 1, and
7476 target is little-endian, print the octets in a word in big-endian
7477 order so that the display looks like a proper hexadecimal number,
7478 instead of having the octets reversed.
7479 * read.c (do_align): When recording alignment, alignment power
7480 should be in terms of target bytes (minimum addressible unit)
7482 (do_org) Convert ORG target address (byte) argument into an
7483 octet offset when generating a variable fragment.
7484 * symbols.c (resolve_symbol_value): Symbol final value
7485 converted to a target address offset (bytes) from its octet offset.
7486 * config/obj-coff.c (coff_frob_symbol): Symbol target address
7487 offset (bytes) is adjusted by the frag offset (octets) converted
7489 (coff_frob_section) Section alignment power is in terms of bytes;
7490 convert it to an octet alignment power when calculating size (and
7491 size mask) in octets. Don't modify the section size in order to
7492 "align" it for TI COFF, since that format has a different method
7493 for storing alignment information.
7497 * stabs.c (generate_asm_file): Escape backslashes in stabs file
7498 entries, matching the way GCC generates them. If not escaped, the
7499 filename is encoded incorrectly.
7503 * config/tc-arm.c (reg_table): Add support for ATPCS register
7508 * config/obj-coff.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Don't define if
7510 * config/tc-ppc.h [OBJ_XCOFF] (OBJ_COPY_SYMBOL_ATTRIBUTES):
7512 * config/tc-ppc.c (ppc_fix_adjustable): Don't look at the frag
7513 of a symbol when we really care about its value.
7517 * config/tc-mcore.c (md_assemble): Give warning message if
7518 operands passes to instruction are more than the spec.
7522 * config/tc-arm.c (armadjust_symtab): If the assembler is in
7523 Thumb mode but the label seen was not declared as '.thumb_func'
7524 then set the ST_INFO type to STT_ARM_16BIT mode. This allows
7525 correct disassembly of Thumb code bounded by non function labels.
7529 * Makefile.am (MULTI_CFILES): Add config/e-i386aout.c
7530 Add dependencies for e-i386aout.o. Fix 2 comment lines.
7532 * Makefile.in: Same here.
7535 * configure.in: Set bfd_gas for i386-aout when primary target
7536 is bfd. Handle i386aout emulation. Don't use te_file=multi, as
7537 we may need the primary te_file. Remove incorrect comment.
7539 * configure: Regenerate.
7541 * config/e-i386aout.c: New file.
7543 * as.c (USE_EMULATIONS): Move to before print_version_id.
7544 (struct emulation): Add i386aout.
7545 (show_usage): Split text strings. Reformat -a text. Add --em
7549 * obj.h (struct format_ops): Add s_get_other and s_get_desc.
7550 (aout_format_ops): New.
7553 * read.c (s_lcomm_internal): Rewrite OBJ_AOUT,OBJ_BOUT
7554 preprocessor conditional and add aout USE_EMULATIONS tests.
7555 (read_a_source_file): Don't pass error strings to printf as
7559 * gasp.c (exp_get_abs): Don't pass error strings to printf as
7561 (do_data): Same here.
7562 (process_file): And here.
7565 * symbols.c (colon): Rewrite "already defined" fatal message
7566 code for aout with USE_EMULATIONS.
7569 * config/obj-aout.c (OBJ_HEADER): Define.
7570 (obj_pseudo_table): Rename to aout_pseudo_table. Init all
7572 (obj_aout_frob_symbol): Expand S_GET_DESC, S_GET_TYPE,
7573 S_GET_OTHER, S_SET_TYPE macros since we don't need obj-multi
7575 (obj_aout_type): Expand S_SET_OTHER here too.
7576 (obj_read_begin_hook): Remove.
7577 (aout_pop_insert): New.
7578 (obj_aout_s_get_other): New.
7579 (obj_aout_s_get_desc): New.
7580 (aout_format_ops): New.
7583 * config/obj-aout.h (obj_pop_insert): Define so non-multi usage
7584 gets aout_pseudo_table.
7585 (aout_pseudo_table): Declare.
7586 (obj_read_begin_hook): Define.
7589 * config/obj-coff.c (obj_pseudo_table): Rename to
7591 (coff_pop_insert): Use coff_pseudo_table.
7592 (coff_sec_sym_ok_for_reloc): Remove.
7593 (coff_format_ops): Add 0 entries for s_get_size, s_set_size,
7594 and comment all zero entries and remove #if 0 code.
7597 * config/obj-coff.h (obj_pop_insert): Define.
7598 (coff_pseudo_table): Declare.
7601 * config/obj-ecoff.c (ecoff_format_ops): Add 0 entries for
7602 s_get_size, s_set_size. Comment all zero entries.
7605 * config/obj-elf.c (elf_s_get_other): New function.
7606 (obj_read_begin_hook): Rename to elf_obj_read_begin_hook.
7607 (obj_symbol_new_hook): Rename to elf_obj_symbol_new_hook.
7608 (elf_format_ops): Add elf_s_get_other, 0 s_get_size entry, and
7610 (obj_elf_parse_section_letters): Don't pass error strings to
7611 printf as format arg.
7614 * config/obj-elf.h (ECOFF_DEBUGGING): Define when
7616 (elf_s_get_other): Declare.
7617 (S_GET_OTHER) Define as elf_s_get_other if not already
7619 (S_SET_OTHER): Only define when not already defined.
7620 (elf_obj_read_begin_hook): Declare.
7621 (obj_read_begin_hook): Define.
7622 (elf_obj_symbol_new_hook): Declare.
7623 (obj_symbol_new_hook): Define.
7626 * config/obj-multi.h: Add copyright header and protect against
7627 multiple inclusion. Add * to all function pointers.
7628 (OBJ_HEADER): If defined, include it rather than other defines
7630 (obj_frob_file_after_relocs): Test for NULL.
7631 (obj_symbol_new_hook): Here too.
7632 (obj_sec_sym_ok_for_reloc): And here.
7633 (S_GET_OTHER): Define.
7634 (S_GET_DESC): Define.
7635 (ECOFF_DEBUGGING): Remove as it's done in obj-elf.h
7636 (OBJ_MAYBE_ELF): Update comment.
7638 * config/tc-i386.c (i386_immediate): Add OBJ_MAYBE_AOUT to
7639 OBJ_AOUT preprocessor conditional and handle emulation by
7640 testing OUTPUT_FLAVOR.
7641 (i386_displacement): Here too.
7642 (md_section_align): Similarly here.
7643 (i386_target_format): Conditionally compile when more than one
7644 of OBJ_MAYBE_{ELF,COFF,AOUT} defined. Add aout case.
7645 (i386_immediate): Fix error message for aout BFD_ASSMBLER.
7646 (i386_displacement): Here too.
7649 * config/tc-i386.h (AOUT_TARGET_FORMAT): Define for each TE_*.
7650 Define TARGET_FORMAT for aout only when not multi.
7653 * config/te-multi.h: Delete file as it's identical to te-generic.h
7657 * config/tc-i386.h (DWORD_MNEM_SUFFIX): Delete.
7658 * config/tc-i386.c (DWORD_MNEM_SUFFIX): Rename all occurrences to
7661 * config/tc-i386.h (INTEL_DWORD_MNEM_SUFFIX): Rename to
7663 * config/tc-i386.c (INTEL_DWORD_MNEM_SUFFIX): Here too. Fix some
7668 * config/tc-mips.c (mips_do_align): New function.
7669 * config/tc-mips.h (md_do_align): Define.
7673 * doc/c-arm.texi (ARM Options): Fix typo.
7674 (ARM-Chars): Correct description of `#'. Mention that `;' is a
7675 line separator for Linux.
7676 * doc/as.texinfo (Comments): Mention the ARM.
7680 * configure.in (arm*-*-conix*): New target.
7681 (arm*-*-linux-gnu*): Match instead of arm-*-linux* and
7683 * configure: Regenerate.
7687 * config/obj-elf.c (elf_pseudo_table): Define visibility pseudos.
7688 (obj_elf_visibility): New function.
7690 * doc/as.texinfo (Visibility): New node: document visibility
7693 For older changes see ChangeLog-9899
7699 version-control: never