gas/ChangeLog

   1 2006-05-19  Thiemo Seufer  <ths@mips.com>
   2             Nigel Stephens  <nigel@mips.com>
   3
   4         * config/tc-mipc.c (ISA_HAS_ODD_SINGLE_FPR): New define.
   5         (mips_oddfpreg_ok): New function.
   6         (mips_ip): Use it.
   7
   8 2006-05-19  Thiemo Seufer  <ths@mips.com>
   9             David Ung  <davidu@mips.com>
  10
  11         * config/tc-mips.h (tc_mips_regname_to_dw2regnum): Declare.
  12         * config/tc-mipc.c (ABI_NEEDS_64BIT_REGS, ISA_HAS_64BIT_REGS,
  13         ISA_HAS_DROR, ISA_HAS_ROR): Reformat.
  14         (regname, RTYPE_MASK, RTYPE_NUM, RTYPE_FPU, RTYPE_FCC, RTYPE_VEC,
  15         RTYPE_GP, RTYPE_CP0, RTYPE_PC, RTYPE_ACC, RTYPE_CCC, RNUM_MASK,
  16         RWARN, GENERIC_REGISTER_NUMBERS, FPU_REGISTER_NAMES,
  17         FPU_CONDITION_CODE_NAMES, COPROC_CONDITION_CODE_NAMES,
  18         N32N64_SYMBOLIC_REGISTER_NAMES, O32_SYMBOLIC_REGISTER_NAMES,
  19         SYMBOLIC_REGISTER_NAMES, MIPS16_SPECIAL_REGISTER_NAMES,
  20         MDMX_VECTOR_REGISTER_NAMES, MIPS_DSP_ACCUMULATOR_NAMES, reg_names,
  21         reg_names_o32, reg_names_n32n64): Define register classes.
  22         (reg_lookup): New function, use register classes.
  23         (md_begin): Reserve register names in the symbol table. Simplify
  24         OBJ_ELF defines.
  25         (mips_ip): Fix comment formatting. Handle symbolic COP0 registers.
  26         Use reg_lookup.
  27         (mips16_ip): Use reg_lookup.
  28         (tc_get_register): Likewise.
  29         (tc_mips_regname_to_dw2regnum): New function.
  30
  31 2006-05-19  Thiemo Seufer  <ths@mips.com>
  32
  33         * config/tc-arm.c, config/tc-arm.h (tc_arm_regname_to_dw2regnum):
  34         Un-constify string argument.
  35         * config/tc-i386.c, config/tc-i386.h (tc_x86_regname_to_dw2regnum):
  36         Likewise.
  37         * config/tc-m68k.c, config/tc-m68k.h (tc_m68k_regname_to_dw2regnum):
  38         Likewise.
  39         * config/tc-ppc.c, config/tc-ppc.h (tc_ppc_regname_to_dw2regnum):
  40         Likewise.
  41         * config/tc-s390.c, config/tc-s390.h (tc_s390_regname_to_dw2regnum):
  42         Likewise.
  43         * config/tc-sh.c, config/tc-sh.h (sh_regname_to_dw2regnum):
  44         Likewise.
  45         * config/tc-sparc.c, config/tc-sparc.h (sparc_regname_to_dw2regnum):
  46         Likewise.
  47
  48 2006-05-19  Nathan Sidwell  <nathan@codesourcery.com>
  49
  50         * gas/config/tc-m68k.c (m68k_init_arch): Move checking of
  51         cfloat/m68881 to correct architecture before using it.
  52
  53 2006-05-16  Bjoern Haase  <bjoern.m.haase@web.de>
  54
  55         * config/tc-avr.h (TC_VALIDATE_FIX): Allow fixups for immediate
  56         constant values.
  57
  58 2006-05-15  Paul Brook  <paul@codesourcery.com>
  59
  60         * config/tc-arm.c (arm_adjust_symtab): Use
  61         bfd_is_arm_special_symbol_name.
  62
  63 2006-05-15  Bob Wilson  <bob.wilson@acm.org>
  64
  65         * config/tc-xtensa.c (is_direct_call_opcode, is_branch_jmp_to_next,
  66         xg_assemble_vliw_tokens, xtensa_mark_narrow_branches,
  67         xtensa_fix_short_loop_frags, is_local_forward_loop, relax_frag_immed):
  68         Handle errors from calls to xtensa_opcode_is_* functions.
  69
  70 2006-05-14  Thiemo Seufer  <ths@mips.com>
  71
  72         * config/tc-mips.c (macro_build): Test for currently active
  73         mips16 option.
  74         (mips16_ip): Reject invalid opcodes.
  75
  76 2006-05-11  Carlos O'Donell  <carlos@codesourcery.com>
  77
  78         * doc/as.texinfo: Rename "Index" to "AS Index",
  79         and "ABORT" to "ABORT (COFF)".
  80
  81 2006-05-11  Paul Brook  <paul@codesourcery.com>
  82
  83         * config/tc-arm.c (parse_half): New function.
  84         (operand_parse_code): Remove OP_Iffff.  Add OP_HALF.
  85         (parse_operands): Ditto.
  86         (do_mov16): Reject invalid relocations.
  87         (do_t_mov16): Ditto.  Use Thumb reloc numbers.
  88         (insns): Replace Iffff with HALF.
  89         (md_apply_fix): Add MOVW and MOVT relocs.
  90         (tc_gen_reloc): Ditto.
  91         * doc/c-arm.texi: Document relocation operators
  92
  93 2006-05-11  Paul Brook  <paul@codesourcery.com>
  94
  95         * config/tc-arm.c (arm_fix_adjustable): Return 0 for function symbols.
  96
  97 2006-05-11  Thiemo Seufer  <ths@mips.com>
  98
  99         * config/tc-mips.c (append_insn): Don't check the range of j or
 100         jal addresses.
 101
 102 2006-05-11  Pedro Alves  <pedro_alves@portugalmail.pt>
 103
 104         * config/tc-arm.c (md_pcrel_from_section): Force a bias for
 105         relocs against external symbols for WinCE targets.
 106         (md_apply_fix): Likewise.
 107
 108 2006-05-09  David Ung  <davidu@mips.com>
 109
 110         * config/tc-mips.c (append_insn): Only warn about an out-of-range
 111         j or jal address.
 112
 113 2006-05-09  Nick Clifton  <nickc@redhat.com>
 114
 115         * config/tc-arm.c (arm_fix_adjustable): For COFF, convert fixups
 116         against symbols which are not going to be placed into the symbol
 117         table.
 118
 119 2006-05-09  Ben Elliston  <bje@au.ibm.com>
 120
 121         * expr.c (operand): Remove `if (0 && ..)' statement and
 122         subsequently unused target_op label.  Collapse `if (1 || ..)'
 123         statement.
 124         * app.c (do_scrub_chars): Remove unused case 0, as it is handled
 125         separately above the switch.
 126
 127 2006-05-08  Nick Clifton  <nickc@redhat.com>
 128
 129         PR gas/2623
 130         * config/tc-msp430.c (line_separator_character): Define as |.
 131
 132 2006-05-08  Thiemo Seufer  <ths@mips.com>
 133             Nigel Stephens  <nigel@mips.com>
 134             David Ung  <davidu@mips.com>
 135
 136         * config/tc-mips.c (mips_set_options): Add ase_smartmips flag.
 137         (mips_opts): Likewise.
 138         (file_ase_smartmips): New variable.
 139         (ISA_HAS_ROR): SmartMIPS implements rotate instructions.
 140         (macro_build): Handle SmartMIPS instructions.
 141         (mips_ip): Likewise.
 142         (md_longopts): Add argument handling for smartmips.
 143         (md_parse_options, mips_after_parse_args): Likewise.
 144         (s_mipsset): Add .set smartmips support.
 145         (md_show_usage): Document -msmartmips/-mno-smartmips.
 146         * doc/as.texinfo: Document -msmartmips/-mno-smartmips and
 147         .set smartmips.
 148         * doc/c-mips.texi: Likewise.
 149
 150 2006-05-08  Alan Modra  <amodra@bigpond.net.au>
 151
 152         * write.c (relax_segment): Add pass count arg.  Don't error on
 153         negative org/space on first two passes.
 154         (relax_seg_info): New struct.
 155         (relax_seg, write_object_file): Adjust.
 156         * write.h (relax_segment): Update prototype.
 157
 158 2006-05-05  Julian Brown  <julian@codesourcery.com>
 159
 160         * config/tc-arm.c (parse_vfp_reg_list): Improve register bounds
 161         checking.
 162         (do_neon_mov): Enable several VMOV variants for VFP. Add suitable
 163         architecture version checks.
 164         (insns): Allow overlapping instructions to be used in VFP mode.
 165
 166 2006-05-05  H.J. Lu  <hongjiu.lu@intel.com>
 167
 168         PR gas/2598
 169         * config/obj-elf.c (obj_elf_change_section): Allow user
 170         specified SHF_ALPHA_GPREL.
 171
 172 2006-05-05  Bjoern Haase  <bjoern.m.haase@web.de>
 173
 174         * gas/config/tc-avr.h (TC_VALIDATE_FIX): Define.  Disable fixups
 175         for PMEM related expressions.
 176
 177 2006-05-05  Nick Clifton  <nickc@redhat.com>
 178
 179         PR gas/2582
 180         * dwarf2dbg.c (INSERT_DIR_SEPARATOR): New macro.  Handles the
 181         insertion of a directory separator character into a string at a
 182         given offset.  Uses heuristics to decide when to use a backslash
 183         character rather than a forward-slash character.
 184         (dwarf2_directive_loc): Use the macro.
 185         (out_debug_info): Likewise.
 186
 187 2006-05-05  Thiemo Seufer  <ths@mips.com>
 188             David Ung  <davidu@mips.com>
 189
 190         * config/tc-mips.c (macro_build): Add case 'k' to handle cache
 191         instruction.
 192         (macro): Add new case M_CACHE_AB.
 193
 194 2006-05-04  Kazu Hirata  <kazu@codesourcery.com>
 195
 196         * config/tc-arm.c (opcode_tag): Add OT_cinfix3_deprecated.
 197         (opcode_lookup): Issue a warning for opcode with
 198         OT_cinfix3_deprecated.  Otherwise treat OT_cinfix3_deprecated
 199         identical to OT_cinfix3.
 200         (TxC3w, TC3w, tC3w): New.
 201         (insns): Use tC3w and TC3w for comparison instructions with
 202         's' suffix.
 203
 204 2006-05-04  Alan Modra  <amodra@bigpond.net.au>
 205
 206         * subsegs.h (struct frchain): Delete frch_seg.
 207         (frchain_root): Delete.
 208         (seg_info): Define as macro.
 209         * subsegs.c (frchain_root): Delete.
 210         (abs_seg_info, und_seg_info, absolute_frchain): Delete.
 211         (subsegs_begin, subseg_change): Adjust for above.
 212         (subseg_set_rest): Likewise.  Add new frchain structs to seginfo
 213         rather than to one big list.
 214         (subseg_get): Don't special case abs, und sections.
 215         (subseg_new, subseg_force_new): Don't set frchainP here.
 216         (seg_info): Delete.
 217         (subsegs_print_statistics): Adjust frag chain control list traversal.
 218         * debug.c (dmp_frags):  Likewise.
 219         * dwarf2dbg.c (first_frag_for_seg): Don't start looking for frag
 220         at frchain_root.  Make use of known frchain ordering.
 221         (last_frag_for_seg): Likewise.
 222         (get_frag_fix): Likewise.  Add seg param.
 223         (process_entries, out_debug_aranges): Adjust get_frag_fix calls.
 224         * write.c (chain_frchains_together_1): Adjust for struct frchain.
 225         (SUB_SEGMENT_ALIGN): Likewise.
 226         (subsegs_finish): Adjust frchain list traversal.
 227         * config/tc-xtensa.c (xtensa_cleanup_align_frags): Likewise.
 228         (xtensa_fix_target_frags, xtensa_mark_narrow_branches): Likewise.
 229         (xtensa_mark_zcl_first_insns, xtensa_fix_a0_b_retw_frags): Likewise.
 230         (xtensa_fix_b_j_loop_end_frags): Likewise.
 231         (xtensa_fix_close_loop_end_frags): Likewise.
 232         (xtensa_fix_short_loop_frags, xtensa_sanity_check): Likewise.
 233         (retrieve_segment_info): Delete frch_seg initialisation.
 234
 235 2006-05-03  Alan Modra  <amodra@bigpond.net.au>
 236
 237         * subsegs.c (subseg_get): Don't call obj_sec_set_private_data.
 238         * config/obj-elf.h (obj_sec_set_private_data): Delete.
 239         * config/tc-hppa.c (tc_gen_reloc): Don't use bfd_abs_symbol.
 240         * config/tc-mn10300.c (tc_gen_reloc): Likewise.
 241
 242 2006-05-02  Joseph Myers  <joseph@codesourcery.com>
 243
 244         * config/tc-arm.c (do_iwmmxt_wldstbh): Don't multiply offset by 4
 245         here.
 246         (md_apply_fix3): Multiply offset by 4 here for
 247         BFD_RELOC_ARM_CP_OFF_IMM_S2 and BFD_RELOC_ARM_T32_CP_OFF_IMM_S2.
 248
 249 2006-05-02  H.J. Lu  <hongjiu.lu@intel.com>
 250             Jan Beulich  <jbeulich@novell.com>
 251
 252         * config/tc-i386.c (output_invalid_buf): Change size for
 253         unsigned char.
 254         * config/tc-tic30.c (output_invalid_buf): Likewise.
 255
 256         * config/tc-i386.c (output_invalid): Cast none-ascii char to
 257         unsigned char.
 258         * config/tc-tic30.c (output_invalid): Likewise.
 259
 260 2006-05-02  Daniel Jacobowitz  <dan@codesourcery.com>
 261
 262         * doc/Makefile.am (AM_MAKEINFOFLAGS): New.
 263         (TEXI2POD): Use AM_MAKEINFOFLAGS.
 264         (asconfig.texi): Don't set top_srcdir.
 265         * doc/as.texinfo: Don't use top_srcdir.
 266         * aclocal.m4, Makefile.in, doc/Makefile.in: Regenerated.
 267
 268 2006-05-02  H.J. Lu  <hongjiu.lu@intel.com>
 269
 270         * config/tc-i386.c (output_invalid_buf): Change size to 16.
 271         * config/tc-tic30.c (output_invalid_buf): Likewise.
 272
 273         * config/tc-i386.c (output_invalid): Use snprintf instead of
 274         sprintf.
 275         * config/tc-ia64.c (declare_register_set): Likewise.
 276         (emit_one_bundle): Likewise.
 277         (check_dependencies): Likewise.
 278         * config/tc-tic30.c (output_invalid): Likewise.
 279
 280 2006-05-02  Paul Brook  <paul@codesourcery.com>
 281
 282         * config/tc-arm.c (arm_optimize_expr): New function.
 283         * config/tc-arm.h (md_optimize_expr): Define
 284         (arm_optimize_expr): Add prototype.
 285         (TC_FORCE_RELOCATION_SUB_SAME): Define.
 286
 287 2006-05-02  Ben Elliston  <bje@au.ibm.com>
 288
 289         * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Make single bit
 290         field unsigned.
 291
 292         * sb.h (sb_list_vector): Move to sb.c.
 293         * sb.c (free_list): Use type of sb_list_vector directly.
 294         (sb_build): Fix off-by-one error in assertion about `size'.
 295
 296 2006-05-01  Ben Elliston  <bje@au.ibm.com>
 297
 298         * listing.c (listing_listing): Remove useless loop.
 299         * macro.c (macro_expand): Remove is_positional local variable.
 300         * read.c (s_comm_internal): Simplify `if' condition 1 || x -> 1
 301         and simplify surrounding expressions, where possible.
 302         (assign_symbol): Likewise.
 303         (s_weakref): Likewise.
 304         * symbols.c (colon): Likewise.
 305
 306 2006-05-01  James Lemke  <jwlemke@wasabisystems.com>
 307
 308         * subsegs.c (subseg_set_rest): Always set seginfp->frchainP if NULL.
 309
 310 2006-04-30  Thiemo Seufer  <ths@mips.com>
 311             David Ung  <davidu@mips.com>
 312
 313         * config/tc-mips.c (validate_mips_insn): Handling of udi cases.
 314         (mips_immed): New table that records various handling of udi
 315         instruction patterns.
 316         (mips_ip): Adds udi handling.
 317
 318 2006-04-28  Alan Modra  <amodra@bigpond.net.au>
 319
 320         * dwarf2dbg.c (get_line_subseg): Attach new struct line_seg to end
 321         of list rather than beginning.
 322
 323 2006-04-26  Julian Brown  <julian@codesourcery.com>
 324
 325         * gas/config/tc-arm.c (neon_is_quarter_float): Move, and rename to...
 326         (is_quarter_float): Rename from above. Simplify slightly.
 327         (parse_qfloat_immediate): Parse a "quarter precision" floating-point
 328         number.
 329         (parse_neon_mov): Parse floating-point constants.
 330         (neon_qfloat_bits): Fix encoding.
 331         (neon_cmode_for_move_imm): Tweak to use floating-point encoding in
 332         preference to integer encoding when using the F32 type.
 333
 334 2006-04-26  Julian Brown  <julian@codesourcery.com>
 335
 336         * config/tc-arm.c (neon_el_type): Make NT_invtype be the zero (so
 337         zero-initialising structures containing it will lead to invalid types).
 338         (arm_it): Add vectype to each operand.
 339         (NTA_HASTYPE, NTA_HASINDEX): Constants used in neon_typed_alias
 340         defined field.
 341         (neon_typed_alias): New structure. Extra information for typed
 342         register aliases.
 343         (reg_entry): Add neon type info field.
 344         (arm_reg_parse): Remove RTYPE argument (revert to previous arguments).
 345         Break out alternative syntax for coprocessor registers, etc. into...
 346         (arm_reg_alt_syntax): New function. Alternate syntax handling broken
 347         out from arm_reg_parse.
 348         (parse_neon_type): Move. Return SUCCESS/FAIL.
 349         (first_error): New function. Call to ensure first error which occurs is
 350         reported.
 351         (parse_neon_operand_type): Parse exactly one type.
 352         (NEON_ALL_LANES, NEON_INTERLEAVE_LANES): Move.
 353         (parse_typed_reg_or_scalar): New function. Handle core of both
 354         arm_typed_reg_parse and parse_scalar.
 355         (arm_typed_reg_parse): Parse a register with an optional type.
 356         (NEON_SCALAR_REG, NEON_SCALAR_INDEX): Extract parts of parse_scalar
 357         result.
 358         (parse_scalar): Parse a Neon scalar with optional type.
 359         (parse_reg_list): Use first_error.
 360         (parse_vfp_reg_list): Use arm_typed_reg_parse instead of arm_reg_parse.
 361         (neon_alias_types_same): New function. Return true if two (alias) types
 362         are the same.
 363         (parse_neon_el_struct_list): Use parse_typed_reg_or_scalar. Return type
 364         of elements.
 365         (insert_reg_alias): Return new reg_entry not void.
 366         (insert_neon_reg_alias): New function. Insert type/index information as
 367         well as register for alias.
 368         (create_neon_reg_alias): New function. Parse .dn/.qn directives and
 369         make typed register aliases accordingly.
 370         (s_dn, s_qn): New functions. Handle incorrectly used .dn/.qn at start
 371         of line.
 372         (s_unreq): Delete type information if present.
 373         (s_arm_unwind_save_mmxwr): Remove arg 3 from arm_reg_parse calls.
 374         (s_arm_unwind_save_mmxwcg): Likewise.
 375         (s_arm_unwind_movsp): Likewise.
 376         (s_arm_unwind_setfp): Likewise.
 377         (parse_shift): Likewise.
 378         (parse_shifter_operand): Likewise.
 379         (parse_address): Likewise.
 380         (parse_tb): Likewise.
 381         (tc_arm_regname_to_dw2regnum): Likewise.
 382         (md_pseudo_table): Add dn, qn.
 383         (parse_neon_mov): Handle typed operands.
 384         (parse_operands): Likewise.
 385         (neon_type_mask): Add N_SIZ.
 386         (N_ALLMODS): New macro.
 387         (neon_check_shape): Fix typo in NS_DDD_QQQ case. Use first_error.
 388         (el_type_of_type_chk): Add some safeguards.
 389         (modify_types_allowed): Fix logic bug.
 390         (neon_check_type): Handle operands with types.
 391         (neon_three_same): Remove redundant optional arg handling.
 392         (do_neon_dyadic_i64_su, do_neon_shl_imm, do_neon_qshl_imm)
 393         (do_neon_logic, do_neon_qdmulh, do_neon_fcmp_absolute)
 394         (do_neon_step): Adjust accordingly.
 395         (neon_cmode_for_logic_imm): Use first_error.
 396         (do_neon_bitfield): Call neon_check_type.
 397         (neon_dyadic): Rename to...
 398         (neon_dyadic_misc): ...this. New name for neon_dyadic. Add bitfield
 399         to allow modification of type of the destination.
 400         (do_neon_dyadic_if_su, do_neon_dyadic_if_i, do_neon_dyadic_if_i_d)
 401         (do_neon_addsub_if_i, do_neon_mul): Adjust accordingly.
 402         (do_neon_compare): Make destination be an untyped bitfield.
 403         (neon_scalar_for_mul): Use NEON_SCALAR_REG, NEON_SCALAR_INDEX.
 404         (neon_mul_mac): Return early in case of errors.
 405         (neon_move_immediate): Use first_error.
 406         (neon_mac_reg_scalar_long): Fix type to include scalar.
 407         (do_neon_dup): Likewise.
 408         (do_neon_mov): Likewise (in several places).
 409         (do_neon_tbl_tbx): Fix type.
 410         (do_neon_ld_st_interleave, neon_alignment_bit, do_neon_ld_st_lane)
 411         (do_neon_ld_dup): Exit early in case of errors and/or use
 412         first_error.
 413         (opcode_lookup): Update for parse_neon_type returning SUCCESS/FAIL.
 414         Handle .dn/.qn directives.
 415         (REGDEF): Add zero for reg_entry neon field.
 416
 417 2006-04-26  Julian Brown  <julian@codesourcery.com>
 418
 419         * config/tc-arm.c (limits.h): Include.
 420         (fpu_arch_vfp_v3, fpu_vfp_ext_v3, fpu_neon_ext_v1)
 421         (fpu_vfp_v3_or_neon_ext): Declare constants.
 422         (neon_el_type): New enumeration of types for Neon vector elements.
 423         (neon_type_el): New struct. Define type and size of a vector element.
 424         (NEON_MAX_TYPE_ELS): Define constant. The maximum number of types per
 425         instruction.
 426         (neon_type): Define struct. The type of an instruction.
 427         (arm_it): Add 'vectype' for the current instruction.
 428         (isscalar, immisalign, regisimm, isquad): New predicates for operands.
 429         (vfp_sp_reg_pos): Rename to...
 430         (vfp_reg_pos): ...this, and add VFP_REG_Dd, VFP_REG_Dm, VFP_REG_Dn
 431         tags.
 432         (arm_reg_type): Add REG_TYPE_NQ (Neon Q register) and REG_TYPE_NDQ
 433         (Neon D or Q register).
 434         (reg_expected_msgs): Sync with above. Allow VFD to mean VFP or Neon D
 435         register.
 436         (GE_OPT_PREFIX_BIG): Define constant, for use in...
 437         (my_get_expression): Allow above constant as argument to accept
 438         64-bit constants with optional prefix.
 439         (arm_reg_parse): Add extra argument to return the specific type of
 440         register in when either a D or Q register (REG_TYPE_NDQ) is
 441         requested. Can be NULL.
 442         (parse_scalar): New function. Parse Neon scalar (vector reg and index).
 443         (parse_reg_list): Update for new arm_reg_parse args.
 444         (parse_vfp_reg_list): Allow parsing of Neon D/Q register lists.
 445         (parse_neon_el_struct_list): New function. Parse element/structure
 446         register lists for VLD<n>/VST<n> instructions.
 447         (s_arm_unwind_save_vfp): Update for new parse_vfp_reg_list args.
 448         (s_arm_unwind_save_mmxwr): Likewise.
 449         (s_arm_unwind_save_mmxwcg): Likewise.
 450         (s_arm_unwind_movsp): Likewise.
 451         (s_arm_unwind_setfp): Likewise.
 452         (parse_big_immediate): New function. Parse an immediate, which may be
 453         64 bits wide. Put results in inst.operands[i].
 454         (parse_shift): Update for new arm_reg_parse args.
 455         (parse_address): Likewise. Add parsing of alignment specifiers.
 456         (parse_neon_mov): Parse the operands of a VMOV instruction.
 457         (operand_parse_code): Add OP_RND, OP_RNQ, OP_RNDQ, OP_RNSC, OP_NRDLST,
 458         OP_NSTRLST, OP_NILO, OP_RNDQ_I0, OP_RR_RNSC, OP_RNDQ_RNSC, OP_RND_RNSC,
 459         OP_VMOV, OP_RNDQ_IMVNb, OP_RNDQ_I63b, OP_I0, OP_I16z, OP_I32z, OP_I64,
 460         OP_I64z, OP_oI32b, OP_oRND, OP_oRNQ, OP_oRNDQ.
 461         (parse_operands): Handle new codes above.
 462         (encode_arm_vfp_sp_reg): Rename to...
 463         (encode_arm_vfp_reg): ...this. Handle D regs (0-31) too. Complain if
 464         selected VFP version only supports D0-D15.
 465         (do_vfp_sp_monadic, do_vfp_sp_dyadic, do_vfp_sp_compare_z)
 466         (do_vfp_dp_sp_cvt, do_vfp_reg_from_sp, do_vfp_reg2_from_sp2)
 467         (do_vfp_sp_from_reg, do_vfp_sp2_from_reg2, do_vfp_sp_ldst)
 468         (do_vfp_dp_ldst, vfp_sp_ldstm, vfp_dp_ldstm): Update for new
 469         encode_arm_vfp_reg name, and allow 32 D regs.
 470         (do_vfp_dp_rd_rm, do_vfp_dp_rn_rd, do_vfp_dp_rd_rn, do_vfp_dp_rd_rn_rm)
 471         (do_vfp_rm_rd_rn): New functions to encode VFP insns allowing 32 D
 472         regs.
 473         (do_vfp_sp_const, do_vfp_dp_const, vfp_conv, do_vfp_sp_conv_16)
 474         (do_vfp_dp_conv_16, do_vfp_sp_conv_32, do_vfp_dp_conv_32): Handle
 475         constant-load and conversion insns introduced with VFPv3.
 476         (neon_tab_entry): New struct.
 477         (NEON_ENC_TAB): Bit patterns for overloaded Neon instructions, and
 478         those which are the targets of pseudo-instructions.
 479         (neon_opc): Enumerate opcodes, use as indices into...
 480         (neon_enc_tab): ...this. Hold data from NEON_ENC_TAB.
 481         (NEON_ENC_INTEGER, NEON_ENC_ARMREG, NEON_ENC_POLY, NEON_ENC_FLOAT)
 482         (NEON_ENC_SCALAR, NEON_ENC_IMMED, NEON_ENC_INTERLV, NEON_ENC_LANE)
 483         (NEON_ENC_DUP): Define meaningful helper macros to look up values in
 484         neon_enc_tab.
 485         (neon_shape): Enumerate shapes (permitted register widths, etc.) for
 486         Neon instructions.
 487         (neon_type_mask): New. Compact type representation for type checking.
 488         (N_SU_ALL, N_SU_32, N_SU_16_64, N_SUF_32, N_I_ALL, N_IF_32): Common
 489         permitted type combinations.
 490         (N_IGNORE_TYPE): New macro.
 491         (neon_check_shape): New function. Check an instruction shape for
 492         multiple alternatives. Return the specific shape for the current
 493         instruction.
 494         (neon_modify_type_size): New function. Modify a vector type and size,
 495         depending on the bit mask in argument 1.
 496         (neon_type_promote): New function. Convert a given "key" type (of an
 497         operand) into the correct type for a different operand, based on a bit
 498         mask.
 499         (type_chk_of_el_type): New function. Convert a type and size into the
 500         compact representation used for type checking.
 501         (el_type_of_type_ckh): New function. Reverse of above (only when a
 502         single bit is set in the bit mask).
 503         (modify_types_allowed): New function. Alter a mask of allowed types
 504         based on a bit mask of modifications.
 505         (neon_check_type): New function. Check the type of the current
 506         instruction against the variable argument list. The "key" type of the
 507         instruction is returned.
 508         (neon_dp_fixup): New function. Fill in and modify instruction bits for
 509         a Neon data-processing instruction depending on whether we're in ARM
 510         mode or Thumb-2 mode.
 511         (neon_logbits): New function.
 512         (neon_three_same, neon_two_same, do_neon_dyadic_i_su)
 513         (do_neon_dyadic_i64_su, neon_imm_shift, do_neon_shl_imm)
 514         (do_neon_qshl_imm, neon_cmode_for_logic_imm, neon_bits_same_in_bytes)
 515         (neon_squash_bits, neon_is_quarter_float, neon_qfloat_bits)
 516         (neon_cmode_for_move_imm, neon_write_immbits, neon_invert_size)
 517         (do_neon_logic, do_neon_bitfield, neon_dyadic, do_neon_dyadic_if_su)
 518         (do_neon_dyadic_if_su_d, do_neon_dyadic_if_i, do_neon_dyadic_if_i_d)
 519         (do_neon_addsub_if_i, neon_exchange_operands, neon_compare)
 520         (do_neon_cmp, do_neon_cmp_inv, do_neon_ceq, neon_scalar_for_mul)
 521         (neon_mul_mac, do_neon_mac_maybe_scalar, do_neon_tst, do_neon_mul)
 522         (do_neon_qdmulh, do_neon_fcmp_absolute, do_neon_fcmp_absolute_inv)
 523         (do_neon_step, do_neon_abs_neg, do_neon_sli, do_neon_sri)
 524         (do_neon_qshlu_imm, do_neon_qmovn, do_neon_qmovun)
 525         (do_neon_rshift_sat_narrow, do_neon_rshift_sat_narrow_u, do_neon_movn)
 526         (do_neon_rshift_narrow, do_neon_shll, neon_cvt_flavour, do_neon_cvt)
 527         (neon_move_immediate, do_neon_mvn, neon_mixed_length)
 528         (do_neon_dyadic_long, do_neon_abal, neon_mac_reg_scalar_long)
 529         (do_neon_mac_maybe_scalar_long, do_neon_dyadic_wide, do_neon_vmull)
 530         (do_neon_ext, do_neon_rev, do_neon_dup, do_neon_mov)
 531         (do_neon_rshift_round_imm, do_neon_movl, do_neon_trn, do_neon_zip_uzp)
 532         (do_neon_sat_abs_neg, do_neon_pair_long, do_neon_recip_est)
 533         (do_neon_cls, do_neon_clz, do_neon_cnt, do_neon_swp, do_neon_tbl_tbx)
 534         (do_neon_ldm_stm, do_neon_ldr_str, do_neon_ld_st_interleave)
 535         (neon_alignment_bit, do_neon_ld_st_lane, do_neon_ld_dup)
 536         (do_neon_ldx_stx): New functions. Neon bit encoding and encoding
 537         helpers.
 538         (parse_neon_type): New function. Parse Neon type specifier.
 539         (opcode_lookup): Allow parsing of Neon type specifiers.
 540         (REGNUM2, REGSETH, REGSET2): New macros.
 541         (reg_names): Add new VFPv3 and Neon registers.
 542         (NUF, nUF, NCE, nCE): New macros for opcode table.
 543         (insns): More VFP registers allowed in fcpyd, fmdhr, fmdlr, fmrdh,
 544         fmrdl, fabsd, fnegd, fsqrtd, faddd, fsubd, fmuld, fdivd, fmacd, fmscd,
 545         fnmuld, fnmacd, fnmscd, fcmpd, fcmpzd, fcmped, fcmpezd, fmdrr, fmrrd.
 546         Add Neon instructions vaba, vhadd, vrhadd, vhsub, vqadd, vqsub, vrshl,
 547         vqrshl, vshl, vqshl{u}, vand, vbic, vorr, vorn, veor, vbsl, vbit, vbif,
 548         vabd, vmax, vmin, vcge, vcgt, vclt, vcle, vceq, vpmax, vpmin, vmla,
 549         vmls, vpadd, vadd, vsub, vtst, vmul, vqdmulh, vqrdmulh, vacge, vacgt,
 550         vaclt, vacle, vrecps, vrsqrts, vabs, vneg, v{r}shr,  v{r}sra, vsli,
 551         vsri, vqshrn, vq{r}shr{u}n, v{r}shrn, vshll, vcvt, vmov, vmvn, vabal,
 552         vabdl, vaddl, vsubl, vmlal, vmlsl, vaddw, vsubw, v{r}addhn, v{r}subhn,
 553         vqdmlal, vqdmlsl, vqdmull, vmull, vext, vrev64, vrev32, vrev16, vdup,
 554         vmovl, v{q}movn, vzip, vuzp, vqabs, vqneg, vpadal, vpaddl, vrecpe,
 555         vrsqrte, vcls, vclz, vcnt, vswp, vtrn, vtbl, vtbx, vldm, vstm, vldr,
 556         vstr, vld[1234], vst[1234], fconst[sd], f[us][lh]to[sd],
 557         fto[us][lh][sd].
 558         (tc_arm_regname_to_dw2regnum): Update for arm_reg_parse args.
 559         (arm_cpu_option_table): Add Neon and VFPv3 to Cortex-A8.
 560         (arm_option_cpu_value): Add vfp3 and neon.
 561         (aeabi_set_public_attributes): Support VFPv3 and NEON attributes. Fix
 562         VFPv1 attribute.
 563
 564 2006-04-25  Bob Wilson  <bob.wilson@acm.org>
 565
 566         * config/xtensa-relax.c (widen_spec_list): Use new "WIDE.<opcode>"
 567         syntax instead of hardcoded opcodes with ".w18" suffixes.
 568         (wide_branch_opcode): New.
 569         (build_transition): Use it to check for wide branch opcodes with
 570         either ".w18" or ".w15" suffixes.
 571
 572 2006-04-25  Bob Wilson  <bob.wilson@acm.org>
 573
 574         * config/tc-xtensa.c (xtensa_create_literal_symbol,
 575         xg_assemble_literal, xg_assemble_literal_space): Do not set the
 576         frag's is_literal flag.
 577
 578 2006-04-25  Bob Wilson  <bob.wilson@acm.org>
 579
 580         * config/xtensa-relax.c (XCHAL_HAVE_WIDE_BRANCHES): Provide default.
 581
 582 2006-04-23  Kazu Hirata  <kazu@codesourcery.com>
 583
 584         * config/obj-coff.c, config/tc-arm.c, config/tc-bfin.c,
 585         config/tc-cris.c, config/tc-crx.c, config/tc-i386.c,
 586         config/tc-ia64.c, config/tc-maxq.c, config/tc-maxq.h,
 587         config/tc-mips.c, config/tc-msp430.c, config/tc-sh.c,
 588         config/tc-tic4x.c, config/tc-xtensa.c: Fix comment typos.
 589
 590 2005-04-20  Paul Brook  <paul@codesourcery.com>
 591
 592         * config/tc-arm.c (s_arm_arch, s_arm_cpu, s_arm_fpu): Enable for
 593         all targets.
 594         (md_pseudo_table): Enable .arch, .cpu and .fpu for all targets.
 595
 596 2006-04-19  Alan Modra  <amodra@bigpond.net.au>
 597
 598         * Makefile.am (CPU_TYPES): Add maxq and mt.  Sort.
 599         (CPU_OBJ_VALID): Change sense of COFF test to default to invalid.
 600         Make some cpus unsupported on ELF.  Run "make dep-am".
 601         * Makefile.in: Regenerate.
 602
 603 2006-04-19  Alan Modra  <amodra@bigpond.net.au>
 604
 605         * configure.in (--enable-targets): Indent help message.
 606         * configure: Regenerate.
 607
 608 2006-04-18  H.J. Lu  <hongjiu.lu@intel.com>
 609
 610         PR gas/2533
 611         * config/tc-i386.c (i386_immediate): Check illegal immediate
 612         register operand.
 613
 614 2006-04-18  Alan Modra  <amodra@bigpond.net.au>
 615
 616         * config/tc-i386.c: Formatting.
 617         (output_disp, output_imm): ISO C90 params.
 618
 619         * frags.c (frag_offset_fixed_p): Constify args.
 620         * frags.h (frag_offset_fixed_p): Ditto.
 621
 622         * config/tc-dlx.h (tc_coff_symbol_emit_hook): Delete.
 623         (COFF_MAGIC): Delete.
 624
 625         * config/tc-xc16x.h (TC_LINKRELAX_FIXUP): Delete.
 626
 627 2006-04-16  Daniel Jacobowitz  <dan@codesourcery.com>
 628
 629         * po/POTFILES.in: Regenerated.
 630
 631 2006-04-16  Mark Mitchell  <mark@codesourcery.com>
 632
 633         * doc/as.texinfo: Mention that some .type syntaxes are not
 634         supported on all architectures.
 635
 636 2006-04-14  Sterling Augustine  <sterling@tensilica.com>
 637
 638         * config/tc-xtensa.c (emit_single_op): Do not relax MOVI
 639         instructions when such transformations have been disabled.
 640
 641 2006-04-10  Sterling Augustine  <sterling@tensilica.com>
 642
 643         * config/tc-xtensa.c (xg_assemble_vliw_tokens): Record loop target
 644         symbols in RELAX[_CHECK]_ALIGN_NEXT_OPCODE frags.
 645         (xtensa_fix_close_loop_end_frags): Use the recorded values instead of
 646         decoding the loop instructions.  Remove current_offset variable.
 647         (xtensa_fix_short_loop_frags): Likewise.
 648         (min_bytes_to_other_loop_end): Remove current_offset argument.
 649
 650 2006-04-09  Arnold Metselaar  <arnold.metselaar@planet.nl>
 651
 652         * config/tc-z80.c (z80_optimize_expr): Removed.
 653         * config/tc-z80.h (z80_optimize_expr, md_optimize_expr): Removed.
 654
 655 2006-04-07  Joerg Wunsch <j.gnu@uriah.heep.sax.de>
 656
 657         * gas/config/tc-avr.c (mcu_types): Add support for attiny261,
 658         attiny461, attiny861, attiny25, attiny45, attiny85,attiny24,
 659         attiny44, attiny84, at90pwm2, at90pwm3, atmega164, atmega324,
 660         atmega644, atmega329, atmega3290, atmega649, atmega6490,
 661         atmega406, atmega640, atmega1280, atmega1281, at90can32,
 662         at90can64, at90usb646, at90usb647, at90usb1286 and
 663         at90usb1287.
 664         Move atmega48 and atmega88 from AVR_ISA_M8 to AVR_ISA_PWMx.
 665
 666 2006-04-07  Paul Brook  <paul@codesourcery.com>
 667
 668         * config/tc-arm.c (parse_operands): Set default error message.
 669
 670 2006-04-07  Paul Brook  <paul@codesourcery.com>
 671
 672         * config/tc-arm.c (parse_tb): Set inst.error before returning FAIL.
 673
 674 2006-04-07  Paul Brook  <paul@codesourcery.com>
 675
 676         * config/tc-arm.c (md_apply_fix): Set H bit on blx instruction.
 677
 678 2006-04-07  Paul Brook  <paul@codesourcery.com>
 679
 680         * config/tc-arm.c (THUMB2_LOAD_BIT): Define.
 681         (move_or_literal_pool): Handle Thumb-2 instructions.
 682         (do_t_ldst): Call move_or_literal_pool for =N addressing modes.
 683
 684 2006-04-07  Alan Modra  <amodra@bigpond.net.au>
 685
 686         PR 2512.
 687         * config/tc-i386.c (match_template): Move 64-bit operand tests
 688         inside loop.
 689
 690 2006-04-06  Carlos O'Donell  <carlos@codesourcery.com>
 691
 692         * po/Make-in: Add install-html target.
 693         * Makefile.am: Add install-html and install-html-recursive targets.
 694         * Makefile.in: Regenerate.
 695         * configure.in: AC_SUBST datarootdir, docdir, htmldir.
 696         * configure: Regenerate.
 697         * doc/Makefile.am: Add install-html and install-html-am targets.
 698         * doc/Makefile.in: Regenerate.
 699
 700 2006-04-06  Alan Modra  <amodra@bigpond.net.au>
 701
 702         * frags.c (frag_offset_fixed_p): Reinitialise offset before
 703         second scan.
 704
 705 2006-04-05  Richard Sandiford  <richard@codesourcery.com>
 706             Daniel Jacobowitz  <dan@codesourcery.com>
 707
 708         * config/tc-sparc.c (sparc_target_format): Handle TE_VXWORKS.
 709         (GOTT_BASE, GOTT_INDEX): New.
 710         (tc_gen_reloc): Don't alter relocations against GOTT_BASE and
 711         GOTT_INDEX when generating VxWorks PIC.
 712         * configure.tgt (sparc*-*-vxworks*): Remove this special case;
 713         use the generic *-*-vxworks* stanza instead.
 714
 715 2006-04-04  Alan Modra  <amodra@bigpond.net.au>
 716
 717         PR 997
 718         * frags.c (frag_offset_fixed_p): New function.
 719         * frags.h (frag_offset_fixed_p): Declare.
 720         * expr.c (expr): Use frag_offset_fixed_p when simplifying subtraction.
 721         (resolve_expression): Likewise.
 722
 723 2006-04-03  Sterling Augustine  <sterling@tensilica.com>
 724
 725         * config/tc-xtensa.c (init_op_placement_info_table): Check for formats
 726         of the same length but different numbers of slots.
 727
 728 2006-03-30  Andreas Schwab  <schwab@suse.de>
 729
 730         * configure.in: Fix help string for --enable-targets option.
 731         * configure: Regenerate.
 732
 733 2006-03-28  Nathan Sidwell  <nathan@codesourcery.com>
 734
 735         * gas/config/tc-m68k.c (find_cf_chip): Merge into ...
 736         (m68k_ip): ... here.  Use for all chips.  Protect against buffer
 737         overrun and avoid excessive copying.
 738
 739         * config/tc-m68k.c (m68000_control_regs, m68010_control_regs,
 740         m68020_control_regs, m68040_control_regs, m68060_control_regs,
 741         mcf_control_regs, mcf5208_control_regs, mcf5213_control_regs,
 742         mcf5329_control_regs, mcf5249_control_regs, mcf528x_control_regs,
 743         mcfv4e_control_regs, m68010_control_regs): Rename and reorder to ...
 744         (m68000_ctrl, m68010_ctrl, m68020_ctrl, m68040_ctrl, m68060_ctrl,
 745         mcf_ctrl, mcf5208_ctrl, mcf5213_ctrl, mcf5235_ctrl, mcf5249_ctrl,
 746         mcf5216_ctrl, mcf5250_ctrl, mcf5271_ctrl, mcf5272_ctrl,
 747         mcf5282_ctrl, mcfv4e_ctrl): ... these.
 748         (mcf5275_ctrl, mcf5329_ctrl, mcf5373_ctrl): New.
 749         (struct m68k_cpu): Change chip field to control_regs.
 750         (current_chip): Remove.
 751         (control_regs): New.
 752         (m68k_archs, m68k_extensions): Adjust.
 753         (m68k_cpus): Reorder to be in cpu number order.  Adjust.
 754         (CPU_ALLOW_MC, CPU_ALLOW_NEGATION): Remove.
 755         (find_cf_chip): Reimplement for new organization of cpu table.
 756         (select_control_regs): Remove.
 757         (mri_chip): Adjust.
 758         (struct save_opts): Save control regs, not chip.
 759         (s_save, s_restore): Adjust.
 760         (m68k_lookup_cpu): Give deprecated warning when necessary.
 761         (m68k_init_arch): Adjust.
 762         (md_show_usage): Adjust for new cpu table organization.
 763
 764 2006-03-25  Bernd Schmidt  <bernd.schmidt@analog.com>
 765
 766         * config/bfin-defs.h (Expr_Node_Type enum): Add Expr_Node_GOT_Reloc.
 767         * config/bfin-lex.l: Recognize GOT17M4 and FUNCDESC_GOT17M4.
 768         * config/bfin-parse.y: Include "libbfd.h", "elf/common.h" and
 769         "elf/bfin.h".
 770         (GOT17M4, FUNCDESC_GOT17M4): New tokens of type <value>.
 771         (any_gotrel): New rule.
 772         (got): Use it, and create Expr_Node_GOT_Reloc nodes.
 773         * config/tc-bfin.c: Include "libbfd.h", "elf/common.h" and
 774         "elf/bfin.h".
 775         (DEFAULT_FLAGS, bfin_flags, bfin_pic_flag): New.
 776         (bfin_pic_ptr): New function.
 777         (md_pseudo_table): Add it for ".picptr".
 778         (OPTION_FDPIC): New macro.
 779         (md_longopts): Add -mfdpic.
 780         (md_parse_option): Handle it.
 781         (md_begin): Set BFD flags.
 782         (md_apply_fix3, bfin_fix_adjustable): Handle new relocs.
 783         (bfin_gen_ldstidxi): Adjust to match the trees that the parser gives
 784         us for GOT relocs.
 785         * Makefile.am (bfin-parse.o): Update dependencies.
 786         (DEPTC_bfin_elf): Likewise.
 787         * Makefile.in: Regenerate.
 788
 789 2006-03-25  Richard Sandiford  <richard@codesourcery.com>
 790
 791         * config/tc-m68k.c (m68k_cpus): Change cpu_cf5208 entries to use
 792         mcfemac instead of mcfmac.
 793
 794 2006-03-23  Michael Matz  <matz@suse.de>
 795
 796         * config/tc-i386.c (type_names): Correct placement of 'static'.
 797         (reloc): Map some more relocs to their 64 bit counterpart when
 798         size is 8.
 799         (output_insn): Work around breakage if DEBUG386 is defined.
 800         (output_disp): A BFD_RELOC_64 with GOT_symbol as operand also
 801         needs to be mapped to BFD_RELOC_X86_64_GOTPC64 or
 802         BFD_RELOC_X86_64_GOTPC32.  Also x86-64 handles pcrel addressing
 803         different from i386.
 804         (output_imm): Ditto.
 805         (lex_got): Recognize @PLTOFF and @GOTPLT.  Make @GOT accept also
 806         Imm64.
 807         (md_convert_frag): Jumps can now be larger than 2GB away, error
 808         out in that case.
 809         (tc_gen_reloc): New relocs are passed through.  BFD_RELOC_64
 810         and BFD_RELOC_64_PCREL are mapped to BFD_RELOC_X86_64_GOTPC64.
 811
 812 2006-03-22  Richard Sandiford  <richard@codesourcery.com>
 813             Daniel Jacobowitz  <dan@codesourcery.com>
 814             Phil Edwards  <phil@codesourcery.com>
 815             Zack Weinberg  <zack@codesourcery.com>
 816             Mark Mitchell  <mark@codesourcery.com>
 817             Nathan Sidwell  <nathan@codesourcery.com>
 818
 819         * config/tc-mips.c (mips_target_format): Handle vxworks targets.
 820         (md_begin): Complain about -G being used for PIC.  Don't change
 821         the text, data and bss alignments on VxWorks.
 822         (reloc_needs_lo_p): Don't return true for R_MIPS_GOT16 when
 823         generating VxWorks PIC.
 824         (load_address): Extend SVR4_PIC handling to VXWORKS_PIC.
 825         (macro): Likewise, but do not treat la $25 specially for
 826         VxWorks PIC, and do not handle jal.
 827         (OPTION_MVXWORKS_PIC): New macro.
 828         (md_longopts): Add -mvxworks-pic.
 829         (md_parse_option): Don't complain about using PIC and -G together here.
 830         Handle OPTION_MVXWORKS_PIC.
 831         (md_estimate_size_before_relax): Always use the first relaxation
 832         sequence on VxWorks.
 833         * config/tc-mips.h (VXWORKS_PIC): New.
 834
 835 2006-03-21  Paul Brook  <paul@codesourcery.com>
 836
 837         * config/tc-arm.c (md_apply_fix): Fix typo in offset mask.
 838
 839 2006-03-21  Sterling Augustine  <sterling@tensilica.com>
 840
 841         * config/tc-xtensa.c (enforce_three_byte_loop_align): New flag.
 842         (xtensa_setup_hw_workarounds): Set this new flag for older hardware.
 843         (get_loop_align_size): New.
 844         (xtensa_end): Skip xtensa_mark_narrow_branches when not aligning.
 845         (xtensa_mark_zcl_first_insns): Prevent widening of first loop frag.
 846         (get_text_align_power): Rewrite to handle inputs in the range 2-8.
 847         (get_noop_aligned_address): Use get_loop_align_size.
 848         (get_aligned_diff): Likewise.
 849
 850 2006-03-21  Paul Brook  <paul@codesourcery.com>
 851
 852         * config/tc-arm.c (insns): Correct opcodes for ldrbt and strbt.
 853
 854 2006-03-20  Paul Brook  <paul@codesourcery.com>
 855
 856         * config/tc-arm.c (BAD_BRANCH, BAD_NOT_IT): Define.
 857         (do_t_branch): Encode branches inside IT blocks as unconditional.
 858         (do_t_cps): New function.
 859         (do_t_blx, do_t_bkpt, do_t_branch23, do_t_bx, do_t_bxj, do_t_cpsi,
 860         do_t_czb, do_t_it, do_t_setend, do_t_tb): Add IT constaints.
 861         (opcode_lookup): Allow conditional suffixes on all instructions in
 862         Thumb mode.
 863         (md_assemble): Advance condexec state before checking for errors.
 864         (insns): Use do_t_cps.
 865
 866 2006-03-20  Paul Brook  <paul@codesourcery.com>
 867
 868         * config/tc-arm.c (output_relax_insn): Call dwarf2_emit_insn before
 869         outputting the insn.
 870
 871 2006-03-18  Jan-Benedict Glaw  <jbglaw@lug-owl.de>
 872
 873         * config/tc-vax.c: Update copyright year.
 874         * config/tc-vax.h: Likewise.
 875
 876 2006-03-18  Jan-Benedict Glaw  <jbglaw@lug-owl.de>
 877
 878         * config/tc-vax.c (md_chars_to_number): Used only locally, so
 879         make it static.
 880         * config/tc-vax.h (md_chars_to_number): Remove obsolete declaration.
 881
 882 2006-03-17  Paul Brook  <paul@codesourcery.com>
 883
 884         * config/tc-arm.c (insns): Add ldm and stm.
 885
 886 2006-03-17  Ben Elliston  <bje@au.ibm.com>
 887
 888         PR gas/2446
 889         * doc/as.texinfo (Ident): Document this directive more thoroughly.
 890
 891 2006-03-16  Paul Brook  <paul@codesourcery.com>
 892
 893         * config/tc-arm.c (insns): Add "svc".
 894
 895 2006-03-13  Bob Wilson  <bob.wilson@acm.org>
 896
 897         * config/tc-xtensa.c (xg_translate_sysreg_op): Remove has_underbar
 898         flag and avoid double underscore prefixes.
 899
 900 2006-03-10  Paul Brook  <paul@codesourcery.com>
 901
 902         * config/tc-arm.c (md_begin): Handle EABIv5.
 903         (arm_eabis): Add EF_ARM_EABI_VER5.
 904         * doc/c-arm.texi: Document -meabi=5.
 905
 906 2006-03-10  Ben Elliston  <bje@au.ibm.com>
 907
 908         * app.c (do_scrub_chars): Simplify string handling.
 909
 910 2006-03-07  Richard Sandiford  <richard@codesourcery.com>
 911             Daniel Jacobowitz  <dan@codesourcery.com>
 912             Zack Weinberg  <zack@codesourcery.com>
 913             Nathan Sidwell  <nathan@codesourcery.com>
 914             Paul Brook  <paul@codesourcery.com>
 915             Ricardo Anguiano  <anguiano@codesourcery.com>
 916             Phil Edwards  <phil@codesourcery.com>
 917
 918         * config/tc-arm.c (md_apply_fix): Install a value of zero into a
 919         BFD_RELOC_ARM_OFFSET_IMM field if we're going to generate a RELA
 920         R_ARM_ABS12 reloc.
 921         (tc_gen_reloc): Keep the original fx_offset for RELA pc-relative
 922         relocs, but adjust by md_pcrel_from_section.  Create R_ARM_ABS12
 923         relocations for BFD_RELOC_ARM_OFFSET_IMM on RELA targets.
 924
 925 2006-03-06  Bob Wilson  <bob.wilson@acm.org>
 926
 927         * config/tc-xtensa.c (xtensa_post_relax_hook): Generate literal tables
 928         even when using the text-section-literals option.
 929
 930 2006-03-06  Nathan Sidwell  <nathan@codesourcery.com>
 931
 932         * config/tc-m68k.c (m68k_extensions): Allow 'float' on both m68k
 933         and cf.
 934         (m68k_ip): <case 'J'> Check we have some control regs.
 935         (md_parse_option): Allow raw arch switch.
 936         (m68k_init_arch): Better detection of arch/cpu mismatch.  Detect
 937         whether 68881 or cfloat was meant by -mfloat.
 938         (md_show_usage): Adjust extension display.
 939         (m68k_elf_final_processing): Adjust.
 940
 941 2006-03-03  Bjoern Haase  <bjoern.m.haase@web.de>
 942
 943         * config/tc-avr.c (avr_mod_hash_value): New function.
 944         (md_apply_fix, exp_mod): Use BFD_RELOC_HH8_LDI and
 945         BFD_RELOC_MS8_LDI for hlo8() and hhi8()
 946         (md_begin): Set linkrelax variable to 1, use avr_mod_hash_value
 947         instead of int avr_ldi_expression: use avr_mod_hash_value instead
 948         of (int).
 949         (tc_gen_reloc): Handle substractions of symbols, if possible do
 950         fixups, abort otherwise.
 951         * config/tc-avr.h (TC_LINKRELAX_FIXUP, TC_VALIDATE_FIX,
 952         tc_fix_adjustable): Define.
 953
 954 2006-03-02  James E Wilson  <wilson@specifix.com>
 955
 956         * config/tc-ia64.c (emit_one_bundle): For IA64_OPCODE_LAST, if we
 957         change the template, then clear md.slot[curr].end_of_insn_group.
 958
 959 2006-02-28  Jan Beulich  <jbeulich@novell.com>
 960
 961         * macro.c (get_any_string): Don't insert quotes for <>-quoted input.
 962
 963 2006-02-28  Jan Beulich  <jbeulich@novell.com>
 964
 965         PR/1070
 966         * macro.c (getstring): Don't treat parentheses special anymore.
 967         (get_any_string): Don't consider '(' and ')' as quoting anymore.
 968         Special-case '(', ')', '[', and ']' when dealing with non-quoting
 969         characters.
 970
 971 2006-02-28  Mat <mat@csail.mit.edu>
 972
 973         * dwarf2dbg.c (get_filenum): Don't inadvertently decrease files_in_use.
 974
 975 2006-02-27  Jakub Jelinek  <jakub@redhat.com>
 976
 977         * dw2gencfi.c (struct fde_entry, struct cie_entry): Add signal_frame
 978         field.
 979         (CFI_signal_frame): Define.
 980         (cfi_pseudo_table): Add .cfi_signal_frame.
 981         (dot_cfi): Handle CFI_signal_frame.
 982         (output_cie): Handle cie->signal_frame.
 983         (select_cie_for_fde): Don't share CIE if signal_frame flag is
 984         different.  Copy signal_frame from FDE to newly created CIE.
 985         * doc/as.texinfo: Document .cfi_signal_frame.
 986
 987 2006-02-27  Carlos O'Donell  <carlos@codesourcery.com>
 988
 989         * doc/Makefile.am: Add html target.
 990         * doc/Makefile.in: Regenerate.
 991         * po/Make-in: Add html target.
 992
 993 2006-02-27  H.J. Lu <hongjiu.lu@intel.com>
 994
 995         * config/tc-i386.c (output_insn): Support Intel Merom New
 996         Instructions.
 997
 998         * config/tc-i386.h (CpuMNI): New.
 999         (CpuUnknownFlags): Add CpuMNI.
1000
1001 2006-02-24  David S. Miller  <davem@sunset.davemloft.net>
1002
1003         * config/tc-sparc.c (priv_reg_table): Add entry for "gl".
1004         (hpriv_reg_table): New table for hyperprivileged registers.
1005         (sparc_ip): New cases '$' and '%' for wrhpr/rdhpr hyperprivileged
1006         register encoding.
1007
1008 2006-02-24  DJ Delorie  <dj@redhat.com>
1009
1010         * config/tc-m32c.h (md_apply_fix): Define to m32c_apply_fix.
1011         (tc_gen_reloc): Don't define.
1012         * config/tc-m32c.c (rl_for, relaxable): New convenience macros.
1013         (OPTION_LINKRELAX): New.
1014         (md_longopts): Add it.
1015         (m32c_relax): New.
1016         (md_parse_options): Set it.
1017         (md_assemble): Emit relaxation relocs as needed.
1018         (md_convert_frag): Emit relaxation relocs as needed.
1019         (md_cgen_lookup_reloc): Add LAB_8_8 and LAB_8_16.
1020         (m32c_apply_fix): New.
1021         (tc_gen_reloc): New.
1022         (m32c_force_relocation): Force out jump relocs when relaxing.
1023         (m32c_fix_adjustable): Return false if relaxing.
1024
1025 2006-02-24  Paul Brook  <paul@codesourcery.com>
1026
1027         * config/arm/tc-arm.c (arm_ext_v6_notm, arm_ext_div, arm_ext_v7,
1028         arm_ext_v7a, arm_ext_v7r, arm_ext_v7m): New variables.
1029         (struct asm_barrier_opt): Define.
1030         (arm_v7m_psr_hsh, arm_barrier_opt_hsh): New variables.
1031         (parse_psr): Accept V7M psr names.
1032         (parse_barrier): New function.
1033         (enum operand_parse_code): Add OP_oBARRIER.
1034         (parse_operands): Implement OP_oBARRIER.
1035         (do_barrier): New function.
1036         (do_dbg, do_pli, do_t_barrier, do_t_dbg, do_t_div): New functions.
1037         (do_t_cpsi): Add V7M restrictions.
1038         (do_t_mrs, do_t_msr): Validate V7M variants.
1039         (md_assemble): Check for NULL variants.
1040         (v7m_psrs, barrier_opt_names): New tables.
1041         (insns): Add V7 instructions.  Mark V6 instructions absent from V7M.
1042         (md_begin): Initialize arm_v7m_psr_hsh and arm_barrier_opt_hsh.
1043         (arm_cpu_option_table): Add Cortex-M3, R4 and A8.
1044         (arm_arch_option_table): Add armv7, armv7a, armv7r and armv7m.
1045         (struct cpu_arch_ver_table): Define.
1046         (cpu_arch_ver): New.
1047         (aeabi_set_public_attributes): Use cpu_arch_ver.  Set
1048         Tag_CPU_arch_profile.
1049         * doc/c-arm.texi: Document new cpu and arch options.
1050
1051 2006-02-23  H.J. Lu  <hongjiu.lu@intel.com>
1052
1053         * config/tc-ia64.c (operand_match): Handle IA64_OPND_IMMU5b.
1054
1055 2006-02-23  H.J. Lu  <hongjiu.lu@intel.com>
1056
1057         * config/tc-ia64.c: Update copyright years.
1058
1059 2006-02-22  H.J. Lu  <hongjiu.lu@intel.com>
1060
1061         * config/tc-ia64.c (specify_resource): Add the rule 17 from
1062         SDM 2.2.
1063
1064 2005-02-22  Paul Brook  <paul@codesourcery.com>
1065
1066         * config/tc-arm.c (do_pld): Remove incorrect write to
1067         inst.instruction.
1068         (encode_thumb32_addr_mode): Use correct operand.
1069
1070 2006-02-21  Paul Brook  <paul@codesourcery.com>
1071
1072         * config/tc-arm.c (md_apply_fix): Fix off-by-one errors.
1073
1074 2006-02-17  Shrirang Khisti  <shrirangk@kpitcummins.com>
1075             Anil Paranjape   <anilp1@kpitcummins.com>
1076             Shilin Shakti    <shilins@kpitcummins.com>
1077
1078         * Makefile.am: Add xc16x related entry.
1079         * Makefile.in: Regenerate.
1080         * configure.in: Added xc16x related entry.
1081         * configure: Regenerate.
1082         * config/tc-xc16x.h: New file
1083         * config/tc-xc16x.c: New file
1084         * doc/c-xc16x.texi: New file for xc16x
1085         * doc/all.texi: Entry for xc16x
1086         * doc/Makefile.texi: Added c-xc16x.texi
1087         * NEWS: Announce the support for the new target.
1088
1089 2006-02-16  Nick Hudson  <nick.hudson@dsl.pipex.com>
1090
1091         * configure.tgt: set emulation for mips-*-netbsd*
1092
1093 2006-02-14  Jakub Jelinek  <jakub@redhat.com>
1094
1095         * config.in: Rebuilt.
1096
1097 2006-02-13  Bob Wilson  <bob.wilson@acm.org>
1098
1099         * config/tc-xtensa.c (xg_add_opcode_fix): Number operands starting
1100         from 1, not 0, in error messages.
1101         (md_assemble): Simplify special-case check for ENTRY instructions.
1102         (tinsn_has_invalid_symbolic_operands): Do not include opcode and
1103         operand in error message.
1104
1105 2006-02-13  Joseph S. Myers  <joseph@codesourcery.com>
1106
1107         * configure.tgt (arm-*-linux-gnueabi*): Change to
1108         arm-*-linux-*eabi*.
1109
1110 2006-02-10  Nick Clifton  <nickc@redhat.com>
1111
1112         * config/tc-crx.c (check_range): Ensure that the sign bit of a
1113         32-bit value is propagated into the upper bits of a 64-bit long.
1114
1115         * config/tc-arc.c (init_opcode_tables): Fix cast.
1116         (arc_extoper, md_operand): Likewise.
1117
1118 2006-02-09  David Heine  <dlheine@tensilica.com>
1119
1120         * config/tc-xtensa.c (xg_assembly_relax): Increment steps_taken for
1121         each relaxation step.
1122
1123 2006-02-09  Eric Botcazou  <ebotcazou@libertysurf.fr>
1124
1125         * configure.in (CHECK_DECLS): Add vsnprintf.
1126         * configure: Regenerate.
1127         * messages.c (errno.h, stdarg.h, varargs.h, va_list): Do not
1128         include/declare here, but...
1129         * as.h: Move code detecting VARARGS idiom to the top.
1130         (errno.h, stdarg.h, varargs.h, va_list): ...here.
1131         (vsnprintf): Declare if not already declared.
1132
1133 2006-02-08  H.J. Lu  <hongjiu.lu@intel.com>
1134
1135         * as.c (close_output_file): New.
1136         (main): Register close_output_file with xatexit before
1137         dump_statistics. Don't call output_file_close.
1138
1139 2006-02-07  Nathan Sidwell  <nathan@codesourcery.com>
1140
1141         * config/tc-m68k.c (mcf5208_control_regs, mcf5213_control_regs,
1142         mcf5329_control_regs): New.
1143         (not_current_architecture, selected_arch, selected_cpu): New.
1144         (m68k_archs, m68k_extensions): New.
1145         (archs): Renamed to ...
1146         (m68k_cpus): ... here.  Adjust.
1147         (n_arches): Remove.
1148         (md_pseudo_table): Add arch and cpu directives.
1149         (find_cf_chip, m68k_ip): Adjust table scanning.
1150         (no_68851, no_68881): Remove.
1151         (md_assemble): Lazily initialize.
1152         (select_control_regs): Adjust cpu names. Add 5208, 5213, 5329.
1153         (md_init_after_args): Move functionality to m68k_init_arch.
1154         (mri_chip): Adjust table scanning.
1155         (md_parse_option): Reimplement 'm' processing to add -march & -mcpu
1156         options with saner parsing.
1157         (m68k_lookup_cpu, m68k_set_arch, m68k_set_cpu, m68k_set_extension,
1158         m68k_init_arch): New.
1159         (s_m68k_cpu, s_m68k_arch): New.
1160         (md_show_usage): Adjust.
1161         (m68k_elf_final_processing): Set CF EF flags.
1162         * config/tc-m68k.h (m68k_init_after_args): Remove.
1163         (tc_init_after_args): Remove.
1164         * doc/c-m68k.texi (M68K-Opts): Document -march, -mcpu options.
1165         (M68k-Directives): Document .arch and .cpu directives.
1166
1167 2006-02-05  Arnold Metselaar  <arnold.metselaar@planet.nl>
1168
1169         * config/tc-z80.c (z80_start_line_hook): allow .equ and .defl as
1170         synonyms for equ and defl.
1171         (z80_cons_fix_new): New function.
1172         (emit_byte): Disallow relative jumps to absolute locations.
1173         (emit_data): Only handle defb, prototype changed, because defb is
1174         now handled as pseudo-op rather than an instruction.
1175         (instab): Entries for defb,defw,db,dw moved from here...
1176         (md_pseudo_table): ... to here, use generic cons() for defw,dw.
1177         Add entries for def24,def32,d24,d32.
1178         (md_assemble): Improved error handling.
1179         (md_apply_fix): New case BFD_RELOC_24, set fixP->fx_no_overflow to one.
1180         * config/tc-z80.h (TC_CONS_FIX_NEW): Define.
1181         (z80_cons_fix_new): Declare.
1182         * doc/c-z80.texi (defb, db): Mention warning on overflow.
1183         (def24,d24,def32,d32): New pseudo-ops.
1184
1185 2006-02-02  Paul Brook  <paul@codesourcery.com>
1186
1187         * config/tc-arm.c (do_shift): Remove Thumb-1 constraint.
1188
1189 2005-02-02  Paul Brook  <paul@codesourcery.com>
1190
1191         * config/tc-arm.c (T2_OPCODE_MASK, T2_DATA_OP_SHIFT, T2_OPCODE_AND,
1192         T2_OPCODE_BIC, T2_OPCODE_ORR, T2_OPCODE_ORN, T2_OPCODE_EOR,
1193         T2_OPCODE_ADD, T2_OPCODE_ADC, T2_OPCODE_SBC, T2_OPCODE_SUB,
1194         T2_OPCODE_RSB): Define.
1195         (thumb32_negate_data_op): New function.
1196         (md_apply_fix): Use it.
1197
1198 2006-01-31  Bob Wilson  <bob.wilson@acm.org>
1199
1200         * config/xtensa-istack.h (TInsn): Remove record_fix and sub_symbol
1201         fields.
1202         * config/tc-xtensa.h (xtensa_frag_type): Remove slot_sub_symbols field.
1203         * config/tc-xtensa.c (md_apply_fix): Check for unexpected uses of
1204         subtracted symbols.
1205         (relaxation_requirements): Add pfinish_frag argument and use it to
1206         replace setting tinsn->record_fix fields.
1207         (xg_assemble_vliw_tokens): Adjust calls to relaxation_requirements
1208         and vinsn_to_insnbuf.  Remove references to record_fix and
1209         slot_sub_symbols fields.
1210         (xtensa_mark_narrow_branches): Delete unused code.
1211         (is_narrow_branch_guaranteed_in_range): Handle expr that is not just
1212         a symbol.
1213         (convert_frag_immed): Adjust vinsn_to_insnbuf call and do not set
1214         record_fix fields.
1215         (tinsn_immed_from_frag): Remove code for handling slot_sub_symbols.
1216         (vinsn_to_insnbuf): Change use of record_fixup argument, replacing use
1217         of the record_fix field.  Simplify error messages for unexpected
1218         symbolic operands.
1219         (set_expr_symbol_offset_diff): Delete.
1220
1221 2006-01-31  Paul Brook  <paul@codesourcery.com>
1222
1223         * config/tc-arm.c (arm_reg_parse): Check if reg is non-NULL.
1224
1225 2006-01-31  Paul Brook  <paul@codesourcery.com>
1226         Richard Earnshaw <rearnsha@arm.com>
1227
1228         * config/tc-arm.c: Use arm_feature_set.
1229         (arm_ext_*, arm_arch_full, arm_arch_t2, arm_arch_none,
1230         arm_cext_iwmmxt, arm_cext_xscale, arm_cext_maverick, fpu_fpa_ext_v1,
1231         fpu_fpa_ext_v2, fpu_vfp_ext_v1xd, fpu_vfp_ext_v1, fpu_vfp_ext_v2):
1232         New variables.
1233         (insns): Use them.
1234         (md_atof, opcode_select, opcode_select, md_assemble, md_assemble,
1235         md_begin, arm_parse_extension, arm_parse_cpu, arm_parse_arch,
1236         arm_parse_fpu, arm_parse_float_abi, aeabi_set_public_attributes,
1237         s_arm_cpu, s_arm_arch, s_arm_fpu): Use macros for accessing CPU
1238         feature flags.
1239         (arm_legacy_option_table, arm_option_cpu_value_table): New types.
1240         (arm_opts): Move old cpu/arch options from here...
1241         (arm_legacy_opts): ... to here.
1242         (md_parse_option): Search arm_legacy_opts.
1243         (arm_cpus, arm_archs, arm_extensions, arm_fpus)
1244         (arm_float_abis, arm_eabis): Make const.
1245
1246 2006-01-25  Bob Wilson  <bob.wilson@acm.org>
1247
1248         * config/tc-xtensa.c (md_apply_fix): Set value to zero for PLT relocs.
1249
1250 2006-01-21  Jie Zhang  <jie.zhang@analog.com>
1251
1252         * config/bfin-parse.y (asm_1): Check value range for 16 bit immediate
1253         in load immediate intruction.
1254
1255 2006-01-21  Jie Zhang  <jie.zhang@analog.com>
1256
1257         * config/bfin-parse.y (value_match): Use correct conversion
1258         specifications in template string for __FILE__ and __LINE__.
1259         (binary): Ditto.
1260         (unary): Ditto.
1261
1262 2006-01-18  Alexandre Oliva  <aoliva@redhat.com>
1263
1264         Introduce TLS descriptors for i386 and x86_64.
1265         * config/tc-i386.c (tc_i386_fix_adjustable): Handle
1266         BFD_RELOC_386_TLS_GOTDESC, BFD_RELOC_386_TLS_DESC_CALL,
1267         BFD_RELOC_X86_64_GOTPC32_TLSDESC, BFD_RELOC_X86_64_TLSDESC_CALL.
1268         (optimize_disp): Emit fix up for BFD_RELOC_386_TLS_DESC_CALL and
1269         BFD_RELOC_X86_64_TLSDESC_CALL immediately, and clear the
1270         displacement bits.
1271         (build_modrm_byte): Set up zero modrm for TLS desc calls.
1272         (lex_got): Handle @tlsdesc and @tlscall.
1273         (md_apply_fix, tc_gen_reloc): Handle the new relocations.
1274
1275 2006-01-11  Nick Clifton  <nickc@redhat.com>
1276
1277         Fixes for building on 64-bit hosts:
1278         * config/tc-avr.c (mod_index): New union to allow conversion
1279         between pointers and integers.
1280         (md_begin, avr_ldi_expression): Use it.
1281         * config/tc-i370.c (md_assemble): Add cast for argument to print
1282         statement.
1283         * config/tc-tic54x.c (subsym_substitute): Likewise.
1284         * config/tc-mn10200.c (md_assemble): Use a union to convert the
1285         opindex field of fr_cgen structure into a pointer so that it can
1286         be stored in a frag.
1287         * config/tc-mn10300.c (md_assemble): Likewise.
1288         * config/tc-frv.c (frv_debug_tomcat): Use %p to print pointer
1289         types.
1290         * config/tc-v850.c: Replace uses of (int) casts with correct
1291         types.
1292
1293 2006-01-09  H.J. Lu  <hongjiu.lu@intel.com>
1294
1295         PR gas/2117
1296         * symbols.c (snapshot_symbol): Don't change a defined symbol.
1297
1298 2006-01-03  Hans-Peter Nilsson  <hp@bitrange.com>
1299
1300         PR gas/2101
1301         * config/tc-mmix.c (mmix_handle_mmixal): Don't treat #[0-9][FB] as
1302         a local-label reference.
1303
1304 For older changes see ChangeLog-2005
1305 \f
1306 Local Variables:
1307 mode: change-log
1308 left-margin: 8
1309 fill-column: 74
1310 version-control: never
1311 End: