gas/ChangeLog

   1 2006-05-23  Jie Zhang  <jie.zhang@analog.com>
   2
   3         * config/bfin-defs.h (bfin_equals): Remove declaration.
   4         * config/bfin-parse.y (asm_1): Remove "expr ASSIGN expr".
   5         * config/tc-bfin.c (bfin_name_is_register): Remove.
   6         (bfin_equals): Remove.
   7         * config/tc-bfin.h (TC_EQUAL_IN_INSN): Redefine as 1.
   8         (bfin_name_is_register): Remove declaration.
   9
  10 2006-05-19  Thiemo Seufer  <ths@mips.com>
  11             Nigel Stephens  <nigel@mips.com>
  12
  13         * config/tc-mipc.c (ISA_HAS_ODD_SINGLE_FPR): New define.
  14         (mips_oddfpreg_ok): New function.
  15         (mips_ip): Use it.
  16
  17 2006-05-19  Thiemo Seufer  <ths@mips.com>
  18             David Ung  <davidu@mips.com>
  19
  20         * config/tc-mips.h (tc_mips_regname_to_dw2regnum): Declare.
  21         * config/tc-mipc.c (ABI_NEEDS_64BIT_REGS, ISA_HAS_64BIT_REGS,
  22         ISA_HAS_DROR, ISA_HAS_ROR): Reformat.
  23         (regname, RTYPE_MASK, RTYPE_NUM, RTYPE_FPU, RTYPE_FCC, RTYPE_VEC,
  24         RTYPE_GP, RTYPE_CP0, RTYPE_PC, RTYPE_ACC, RTYPE_CCC, RNUM_MASK,
  25         RWARN, GENERIC_REGISTER_NUMBERS, FPU_REGISTER_NAMES,
  26         FPU_CONDITION_CODE_NAMES, COPROC_CONDITION_CODE_NAMES,
  27         N32N64_SYMBOLIC_REGISTER_NAMES, O32_SYMBOLIC_REGISTER_NAMES,
  28         SYMBOLIC_REGISTER_NAMES, MIPS16_SPECIAL_REGISTER_NAMES,
  29         MDMX_VECTOR_REGISTER_NAMES, MIPS_DSP_ACCUMULATOR_NAMES, reg_names,
  30         reg_names_o32, reg_names_n32n64): Define register classes.
  31         (reg_lookup): New function, use register classes.
  32         (md_begin): Reserve register names in the symbol table. Simplify
  33         OBJ_ELF defines.
  34         (mips_ip): Fix comment formatting. Handle symbolic COP0 registers.
  35         Use reg_lookup.
  36         (mips16_ip): Use reg_lookup.
  37         (tc_get_register): Likewise.
  38         (tc_mips_regname_to_dw2regnum): New function.
  39
  40 2006-05-19  Thiemo Seufer  <ths@mips.com>
  41
  42         * config/tc-arm.c, config/tc-arm.h (tc_arm_regname_to_dw2regnum):
  43         Un-constify string argument.
  44         * config/tc-i386.c, config/tc-i386.h (tc_x86_regname_to_dw2regnum):
  45         Likewise.
  46         * config/tc-m68k.c, config/tc-m68k.h (tc_m68k_regname_to_dw2regnum):
  47         Likewise.
  48         * config/tc-ppc.c, config/tc-ppc.h (tc_ppc_regname_to_dw2regnum):
  49         Likewise.
  50         * config/tc-s390.c, config/tc-s390.h (tc_s390_regname_to_dw2regnum):
  51         Likewise.
  52         * config/tc-sh.c, config/tc-sh.h (sh_regname_to_dw2regnum):
  53         Likewise.
  54         * config/tc-sparc.c, config/tc-sparc.h (sparc_regname_to_dw2regnum):
  55         Likewise.
  56
  57 2006-05-19  Nathan Sidwell  <nathan@codesourcery.com>
  58
  59         * gas/config/tc-m68k.c (m68k_init_arch): Move checking of
  60         cfloat/m68881 to correct architecture before using it.
  61
  62 2006-05-16  Bjoern Haase  <bjoern.m.haase@web.de>
  63
  64         * config/tc-avr.h (TC_VALIDATE_FIX): Allow fixups for immediate
  65         constant values.
  66
  67 2006-05-15  Paul Brook  <paul@codesourcery.com>
  68
  69         * config/tc-arm.c (arm_adjust_symtab): Use
  70         bfd_is_arm_special_symbol_name.
  71
  72 2006-05-15  Bob Wilson  <bob.wilson@acm.org>
  73
  74         * config/tc-xtensa.c (is_direct_call_opcode, is_branch_jmp_to_next,
  75         xg_assemble_vliw_tokens, xtensa_mark_narrow_branches,
  76         xtensa_fix_short_loop_frags, is_local_forward_loop, relax_frag_immed):
  77         Handle errors from calls to xtensa_opcode_is_* functions.
  78
  79 2006-05-14  Thiemo Seufer  <ths@mips.com>
  80
  81         * config/tc-mips.c (macro_build): Test for currently active
  82         mips16 option.
  83         (mips16_ip): Reject invalid opcodes.
  84
  85 2006-05-11  Carlos O'Donell  <carlos@codesourcery.com>
  86
  87         * doc/as.texinfo: Rename "Index" to "AS Index",
  88         and "ABORT" to "ABORT (COFF)".
  89
  90 2006-05-11  Paul Brook  <paul@codesourcery.com>
  91
  92         * config/tc-arm.c (parse_half): New function.
  93         (operand_parse_code): Remove OP_Iffff.  Add OP_HALF.
  94         (parse_operands): Ditto.
  95         (do_mov16): Reject invalid relocations.
  96         (do_t_mov16): Ditto.  Use Thumb reloc numbers.
  97         (insns): Replace Iffff with HALF.
  98         (md_apply_fix): Add MOVW and MOVT relocs.
  99         (tc_gen_reloc): Ditto.
 100         * doc/c-arm.texi: Document relocation operators
 101
 102 2006-05-11  Paul Brook  <paul@codesourcery.com>
 103
 104         * config/tc-arm.c (arm_fix_adjustable): Return 0 for function symbols.
 105
 106 2006-05-11  Thiemo Seufer  <ths@mips.com>
 107
 108         * config/tc-mips.c (append_insn): Don't check the range of j or
 109         jal addresses.
 110
 111 2006-05-11  Pedro Alves  <pedro_alves@portugalmail.pt>
 112
 113         * config/tc-arm.c (md_pcrel_from_section): Force a bias for
 114         relocs against external symbols for WinCE targets.
 115         (md_apply_fix): Likewise.
 116
 117 2006-05-09  David Ung  <davidu@mips.com>
 118
 119         * config/tc-mips.c (append_insn): Only warn about an out-of-range
 120         j or jal address.
 121
 122 2006-05-09  Nick Clifton  <nickc@redhat.com>
 123
 124         * config/tc-arm.c (arm_fix_adjustable): For COFF, convert fixups
 125         against symbols which are not going to be placed into the symbol
 126         table.
 127
 128 2006-05-09  Ben Elliston  <bje@au.ibm.com>
 129
 130         * expr.c (operand): Remove `if (0 && ..)' statement and
 131         subsequently unused target_op label.  Collapse `if (1 || ..)'
 132         statement.
 133         * app.c (do_scrub_chars): Remove unused case 0, as it is handled
 134         separately above the switch.
 135
 136 2006-05-08  Nick Clifton  <nickc@redhat.com>
 137
 138         PR gas/2623
 139         * config/tc-msp430.c (line_separator_character): Define as |.
 140
 141 2006-05-08  Thiemo Seufer  <ths@mips.com>
 142             Nigel Stephens  <nigel@mips.com>
 143             David Ung  <davidu@mips.com>
 144
 145         * config/tc-mips.c (mips_set_options): Add ase_smartmips flag.
 146         (mips_opts): Likewise.
 147         (file_ase_smartmips): New variable.
 148         (ISA_HAS_ROR): SmartMIPS implements rotate instructions.
 149         (macro_build): Handle SmartMIPS instructions.
 150         (mips_ip): Likewise.
 151         (md_longopts): Add argument handling for smartmips.
 152         (md_parse_options, mips_after_parse_args): Likewise.
 153         (s_mipsset): Add .set smartmips support.
 154         (md_show_usage): Document -msmartmips/-mno-smartmips.
 155         * doc/as.texinfo: Document -msmartmips/-mno-smartmips and
 156         .set smartmips.
 157         * doc/c-mips.texi: Likewise.
 158
 159 2006-05-08  Alan Modra  <amodra@bigpond.net.au>
 160
 161         * write.c (relax_segment): Add pass count arg.  Don't error on
 162         negative org/space on first two passes.
 163         (relax_seg_info): New struct.
 164         (relax_seg, write_object_file): Adjust.
 165         * write.h (relax_segment): Update prototype.
 166
 167 2006-05-05  Julian Brown  <julian@codesourcery.com>
 168
 169         * config/tc-arm.c (parse_vfp_reg_list): Improve register bounds
 170         checking.
 171         (do_neon_mov): Enable several VMOV variants for VFP. Add suitable
 172         architecture version checks.
 173         (insns): Allow overlapping instructions to be used in VFP mode.
 174
 175 2006-05-05  H.J. Lu  <hongjiu.lu@intel.com>
 176
 177         PR gas/2598
 178         * config/obj-elf.c (obj_elf_change_section): Allow user
 179         specified SHF_ALPHA_GPREL.
 180
 181 2006-05-05  Bjoern Haase  <bjoern.m.haase@web.de>
 182
 183         * gas/config/tc-avr.h (TC_VALIDATE_FIX): Define.  Disable fixups
 184         for PMEM related expressions.
 185
 186 2006-05-05  Nick Clifton  <nickc@redhat.com>
 187
 188         PR gas/2582
 189         * dwarf2dbg.c (INSERT_DIR_SEPARATOR): New macro.  Handles the
 190         insertion of a directory separator character into a string at a
 191         given offset.  Uses heuristics to decide when to use a backslash
 192         character rather than a forward-slash character.
 193         (dwarf2_directive_loc): Use the macro.
 194         (out_debug_info): Likewise.
 195
 196 2006-05-05  Thiemo Seufer  <ths@mips.com>
 197             David Ung  <davidu@mips.com>
 198
 199         * config/tc-mips.c (macro_build): Add case 'k' to handle cache
 200         instruction.
 201         (macro): Add new case M_CACHE_AB.
 202
 203 2006-05-04  Kazu Hirata  <kazu@codesourcery.com>
 204
 205         * config/tc-arm.c (opcode_tag): Add OT_cinfix3_deprecated.
 206         (opcode_lookup): Issue a warning for opcode with
 207         OT_cinfix3_deprecated.  Otherwise treat OT_cinfix3_deprecated
 208         identical to OT_cinfix3.
 209         (TxC3w, TC3w, tC3w): New.
 210         (insns): Use tC3w and TC3w for comparison instructions with
 211         's' suffix.
 212
 213 2006-05-04  Alan Modra  <amodra@bigpond.net.au>
 214
 215         * subsegs.h (struct frchain): Delete frch_seg.
 216         (frchain_root): Delete.
 217         (seg_info): Define as macro.
 218         * subsegs.c (frchain_root): Delete.
 219         (abs_seg_info, und_seg_info, absolute_frchain): Delete.
 220         (subsegs_begin, subseg_change): Adjust for above.
 221         (subseg_set_rest): Likewise.  Add new frchain structs to seginfo
 222         rather than to one big list.
 223         (subseg_get): Don't special case abs, und sections.
 224         (subseg_new, subseg_force_new): Don't set frchainP here.
 225         (seg_info): Delete.
 226         (subsegs_print_statistics): Adjust frag chain control list traversal.
 227         * debug.c (dmp_frags):  Likewise.
 228         * dwarf2dbg.c (first_frag_for_seg): Don't start looking for frag
 229         at frchain_root.  Make use of known frchain ordering.
 230         (last_frag_for_seg): Likewise.
 231         (get_frag_fix): Likewise.  Add seg param.
 232         (process_entries, out_debug_aranges): Adjust get_frag_fix calls.
 233         * write.c (chain_frchains_together_1): Adjust for struct frchain.
 234         (SUB_SEGMENT_ALIGN): Likewise.
 235         (subsegs_finish): Adjust frchain list traversal.
 236         * config/tc-xtensa.c (xtensa_cleanup_align_frags): Likewise.
 237         (xtensa_fix_target_frags, xtensa_mark_narrow_branches): Likewise.
 238         (xtensa_mark_zcl_first_insns, xtensa_fix_a0_b_retw_frags): Likewise.
 239         (xtensa_fix_b_j_loop_end_frags): Likewise.
 240         (xtensa_fix_close_loop_end_frags): Likewise.
 241         (xtensa_fix_short_loop_frags, xtensa_sanity_check): Likewise.
 242         (retrieve_segment_info): Delete frch_seg initialisation.
 243
 244 2006-05-03  Alan Modra  <amodra@bigpond.net.au>
 245
 246         * subsegs.c (subseg_get): Don't call obj_sec_set_private_data.
 247         * config/obj-elf.h (obj_sec_set_private_data): Delete.
 248         * config/tc-hppa.c (tc_gen_reloc): Don't use bfd_abs_symbol.
 249         * config/tc-mn10300.c (tc_gen_reloc): Likewise.
 250
 251 2006-05-02  Joseph Myers  <joseph@codesourcery.com>
 252
 253         * config/tc-arm.c (do_iwmmxt_wldstbh): Don't multiply offset by 4
 254         here.
 255         (md_apply_fix3): Multiply offset by 4 here for
 256         BFD_RELOC_ARM_CP_OFF_IMM_S2 and BFD_RELOC_ARM_T32_CP_OFF_IMM_S2.
 257
 258 2006-05-02  H.J. Lu  <hongjiu.lu@intel.com>
 259             Jan Beulich  <jbeulich@novell.com>
 260
 261         * config/tc-i386.c (output_invalid_buf): Change size for
 262         unsigned char.
 263         * config/tc-tic30.c (output_invalid_buf): Likewise.
 264
 265         * config/tc-i386.c (output_invalid): Cast none-ascii char to
 266         unsigned char.
 267         * config/tc-tic30.c (output_invalid): Likewise.
 268
 269 2006-05-02  Daniel Jacobowitz  <dan@codesourcery.com>
 270
 271         * doc/Makefile.am (AM_MAKEINFOFLAGS): New.
 272         (TEXI2POD): Use AM_MAKEINFOFLAGS.
 273         (asconfig.texi): Don't set top_srcdir.
 274         * doc/as.texinfo: Don't use top_srcdir.
 275         * aclocal.m4, Makefile.in, doc/Makefile.in: Regenerated.
 276
 277 2006-05-02  H.J. Lu  <hongjiu.lu@intel.com>
 278
 279         * config/tc-i386.c (output_invalid_buf): Change size to 16.
 280         * config/tc-tic30.c (output_invalid_buf): Likewise.
 281
 282         * config/tc-i386.c (output_invalid): Use snprintf instead of
 283         sprintf.
 284         * config/tc-ia64.c (declare_register_set): Likewise.
 285         (emit_one_bundle): Likewise.
 286         (check_dependencies): Likewise.
 287         * config/tc-tic30.c (output_invalid): Likewise.
 288
 289 2006-05-02  Paul Brook  <paul@codesourcery.com>
 290
 291         * config/tc-arm.c (arm_optimize_expr): New function.
 292         * config/tc-arm.h (md_optimize_expr): Define
 293         (arm_optimize_expr): Add prototype.
 294         (TC_FORCE_RELOCATION_SUB_SAME): Define.
 295
 296 2006-05-02  Ben Elliston  <bje@au.ibm.com>
 297
 298         * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Make single bit
 299         field unsigned.
 300
 301         * sb.h (sb_list_vector): Move to sb.c.
 302         * sb.c (free_list): Use type of sb_list_vector directly.
 303         (sb_build): Fix off-by-one error in assertion about `size'.
 304
 305 2006-05-01  Ben Elliston  <bje@au.ibm.com>
 306
 307         * listing.c (listing_listing): Remove useless loop.
 308         * macro.c (macro_expand): Remove is_positional local variable.
 309         * read.c (s_comm_internal): Simplify `if' condition 1 || x -> 1
 310         and simplify surrounding expressions, where possible.
 311         (assign_symbol): Likewise.
 312         (s_weakref): Likewise.
 313         * symbols.c (colon): Likewise.
 314
 315 2006-05-01  James Lemke  <jwlemke@wasabisystems.com>
 316
 317         * subsegs.c (subseg_set_rest): Always set seginfp->frchainP if NULL.
 318
 319 2006-04-30  Thiemo Seufer  <ths@mips.com>
 320             David Ung  <davidu@mips.com>
 321
 322         * config/tc-mips.c (validate_mips_insn): Handling of udi cases.
 323         (mips_immed): New table that records various handling of udi
 324         instruction patterns.
 325         (mips_ip): Adds udi handling.
 326
 327 2006-04-28  Alan Modra  <amodra@bigpond.net.au>
 328
 329         * dwarf2dbg.c (get_line_subseg): Attach new struct line_seg to end
 330         of list rather than beginning.
 331
 332 2006-04-26  Julian Brown  <julian@codesourcery.com>
 333
 334         * gas/config/tc-arm.c (neon_is_quarter_float): Move, and rename to...
 335         (is_quarter_float): Rename from above. Simplify slightly.
 336         (parse_qfloat_immediate): Parse a "quarter precision" floating-point
 337         number.
 338         (parse_neon_mov): Parse floating-point constants.
 339         (neon_qfloat_bits): Fix encoding.
 340         (neon_cmode_for_move_imm): Tweak to use floating-point encoding in
 341         preference to integer encoding when using the F32 type.
 342
 343 2006-04-26  Julian Brown  <julian@codesourcery.com>
 344
 345         * config/tc-arm.c (neon_el_type): Make NT_invtype be the zero (so
 346         zero-initialising structures containing it will lead to invalid types).
 347         (arm_it): Add vectype to each operand.
 348         (NTA_HASTYPE, NTA_HASINDEX): Constants used in neon_typed_alias
 349         defined field.
 350         (neon_typed_alias): New structure. Extra information for typed
 351         register aliases.
 352         (reg_entry): Add neon type info field.
 353         (arm_reg_parse): Remove RTYPE argument (revert to previous arguments).
 354         Break out alternative syntax for coprocessor registers, etc. into...
 355         (arm_reg_alt_syntax): New function. Alternate syntax handling broken
 356         out from arm_reg_parse.
 357         (parse_neon_type): Move. Return SUCCESS/FAIL.
 358         (first_error): New function. Call to ensure first error which occurs is
 359         reported.
 360         (parse_neon_operand_type): Parse exactly one type.
 361         (NEON_ALL_LANES, NEON_INTERLEAVE_LANES): Move.
 362         (parse_typed_reg_or_scalar): New function. Handle core of both
 363         arm_typed_reg_parse and parse_scalar.
 364         (arm_typed_reg_parse): Parse a register with an optional type.
 365         (NEON_SCALAR_REG, NEON_SCALAR_INDEX): Extract parts of parse_scalar
 366         result.
 367         (parse_scalar): Parse a Neon scalar with optional type.
 368         (parse_reg_list): Use first_error.
 369         (parse_vfp_reg_list): Use arm_typed_reg_parse instead of arm_reg_parse.
 370         (neon_alias_types_same): New function. Return true if two (alias) types
 371         are the same.
 372         (parse_neon_el_struct_list): Use parse_typed_reg_or_scalar. Return type
 373         of elements.
 374         (insert_reg_alias): Return new reg_entry not void.
 375         (insert_neon_reg_alias): New function. Insert type/index information as
 376         well as register for alias.
 377         (create_neon_reg_alias): New function. Parse .dn/.qn directives and
 378         make typed register aliases accordingly.
 379         (s_dn, s_qn): New functions. Handle incorrectly used .dn/.qn at start
 380         of line.
 381         (s_unreq): Delete type information if present.
 382         (s_arm_unwind_save_mmxwr): Remove arg 3 from arm_reg_parse calls.
 383         (s_arm_unwind_save_mmxwcg): Likewise.
 384         (s_arm_unwind_movsp): Likewise.
 385         (s_arm_unwind_setfp): Likewise.
 386         (parse_shift): Likewise.
 387         (parse_shifter_operand): Likewise.
 388         (parse_address): Likewise.
 389         (parse_tb): Likewise.
 390         (tc_arm_regname_to_dw2regnum): Likewise.
 391         (md_pseudo_table): Add dn, qn.
 392         (parse_neon_mov): Handle typed operands.
 393         (parse_operands): Likewise.
 394         (neon_type_mask): Add N_SIZ.
 395         (N_ALLMODS): New macro.
 396         (neon_check_shape): Fix typo in NS_DDD_QQQ case. Use first_error.
 397         (el_type_of_type_chk): Add some safeguards.
 398         (modify_types_allowed): Fix logic bug.
 399         (neon_check_type): Handle operands with types.
 400         (neon_three_same): Remove redundant optional arg handling.
 401         (do_neon_dyadic_i64_su, do_neon_shl_imm, do_neon_qshl_imm)
 402         (do_neon_logic, do_neon_qdmulh, do_neon_fcmp_absolute)
 403         (do_neon_step): Adjust accordingly.
 404         (neon_cmode_for_logic_imm): Use first_error.
 405         (do_neon_bitfield): Call neon_check_type.
 406         (neon_dyadic): Rename to...
 407         (neon_dyadic_misc): ...this. New name for neon_dyadic. Add bitfield
 408         to allow modification of type of the destination.
 409         (do_neon_dyadic_if_su, do_neon_dyadic_if_i, do_neon_dyadic_if_i_d)
 410         (do_neon_addsub_if_i, do_neon_mul): Adjust accordingly.
 411         (do_neon_compare): Make destination be an untyped bitfield.
 412         (neon_scalar_for_mul): Use NEON_SCALAR_REG, NEON_SCALAR_INDEX.
 413         (neon_mul_mac): Return early in case of errors.
 414         (neon_move_immediate): Use first_error.
 415         (neon_mac_reg_scalar_long): Fix type to include scalar.
 416         (do_neon_dup): Likewise.
 417         (do_neon_mov): Likewise (in several places).
 418         (do_neon_tbl_tbx): Fix type.
 419         (do_neon_ld_st_interleave, neon_alignment_bit, do_neon_ld_st_lane)
 420         (do_neon_ld_dup): Exit early in case of errors and/or use
 421         first_error.
 422         (opcode_lookup): Update for parse_neon_type returning SUCCESS/FAIL.
 423         Handle .dn/.qn directives.
 424         (REGDEF): Add zero for reg_entry neon field.
 425
 426 2006-04-26  Julian Brown  <julian@codesourcery.com>
 427
 428         * config/tc-arm.c (limits.h): Include.
 429         (fpu_arch_vfp_v3, fpu_vfp_ext_v3, fpu_neon_ext_v1)
 430         (fpu_vfp_v3_or_neon_ext): Declare constants.
 431         (neon_el_type): New enumeration of types for Neon vector elements.
 432         (neon_type_el): New struct. Define type and size of a vector element.
 433         (NEON_MAX_TYPE_ELS): Define constant. The maximum number of types per
 434         instruction.
 435         (neon_type): Define struct. The type of an instruction.
 436         (arm_it): Add 'vectype' for the current instruction.
 437         (isscalar, immisalign, regisimm, isquad): New predicates for operands.
 438         (vfp_sp_reg_pos): Rename to...
 439         (vfp_reg_pos): ...this, and add VFP_REG_Dd, VFP_REG_Dm, VFP_REG_Dn
 440         tags.
 441         (arm_reg_type): Add REG_TYPE_NQ (Neon Q register) and REG_TYPE_NDQ
 442         (Neon D or Q register).
 443         (reg_expected_msgs): Sync with above. Allow VFD to mean VFP or Neon D
 444         register.
 445         (GE_OPT_PREFIX_BIG): Define constant, for use in...
 446         (my_get_expression): Allow above constant as argument to accept
 447         64-bit constants with optional prefix.
 448         (arm_reg_parse): Add extra argument to return the specific type of
 449         register in when either a D or Q register (REG_TYPE_NDQ) is
 450         requested. Can be NULL.
 451         (parse_scalar): New function. Parse Neon scalar (vector reg and index).
 452         (parse_reg_list): Update for new arm_reg_parse args.
 453         (parse_vfp_reg_list): Allow parsing of Neon D/Q register lists.
 454         (parse_neon_el_struct_list): New function. Parse element/structure
 455         register lists for VLD<n>/VST<n> instructions.
 456         (s_arm_unwind_save_vfp): Update for new parse_vfp_reg_list args.
 457         (s_arm_unwind_save_mmxwr): Likewise.
 458         (s_arm_unwind_save_mmxwcg): Likewise.
 459         (s_arm_unwind_movsp): Likewise.
 460         (s_arm_unwind_setfp): Likewise.
 461         (parse_big_immediate): New function. Parse an immediate, which may be
 462         64 bits wide. Put results in inst.operands[i].
 463         (parse_shift): Update for new arm_reg_parse args.
 464         (parse_address): Likewise. Add parsing of alignment specifiers.
 465         (parse_neon_mov): Parse the operands of a VMOV instruction.
 466         (operand_parse_code): Add OP_RND, OP_RNQ, OP_RNDQ, OP_RNSC, OP_NRDLST,
 467         OP_NSTRLST, OP_NILO, OP_RNDQ_I0, OP_RR_RNSC, OP_RNDQ_RNSC, OP_RND_RNSC,
 468         OP_VMOV, OP_RNDQ_IMVNb, OP_RNDQ_I63b, OP_I0, OP_I16z, OP_I32z, OP_I64,
 469         OP_I64z, OP_oI32b, OP_oRND, OP_oRNQ, OP_oRNDQ.
 470         (parse_operands): Handle new codes above.
 471         (encode_arm_vfp_sp_reg): Rename to...
 472         (encode_arm_vfp_reg): ...this. Handle D regs (0-31) too. Complain if
 473         selected VFP version only supports D0-D15.
 474         (do_vfp_sp_monadic, do_vfp_sp_dyadic, do_vfp_sp_compare_z)
 475         (do_vfp_dp_sp_cvt, do_vfp_reg_from_sp, do_vfp_reg2_from_sp2)
 476         (do_vfp_sp_from_reg, do_vfp_sp2_from_reg2, do_vfp_sp_ldst)
 477         (do_vfp_dp_ldst, vfp_sp_ldstm, vfp_dp_ldstm): Update for new
 478         encode_arm_vfp_reg name, and allow 32 D regs.
 479         (do_vfp_dp_rd_rm, do_vfp_dp_rn_rd, do_vfp_dp_rd_rn, do_vfp_dp_rd_rn_rm)
 480         (do_vfp_rm_rd_rn): New functions to encode VFP insns allowing 32 D
 481         regs.
 482         (do_vfp_sp_const, do_vfp_dp_const, vfp_conv, do_vfp_sp_conv_16)
 483         (do_vfp_dp_conv_16, do_vfp_sp_conv_32, do_vfp_dp_conv_32): Handle
 484         constant-load and conversion insns introduced with VFPv3.
 485         (neon_tab_entry): New struct.
 486         (NEON_ENC_TAB): Bit patterns for overloaded Neon instructions, and
 487         those which are the targets of pseudo-instructions.
 488         (neon_opc): Enumerate opcodes, use as indices into...
 489         (neon_enc_tab): ...this. Hold data from NEON_ENC_TAB.
 490         (NEON_ENC_INTEGER, NEON_ENC_ARMREG, NEON_ENC_POLY, NEON_ENC_FLOAT)
 491         (NEON_ENC_SCALAR, NEON_ENC_IMMED, NEON_ENC_INTERLV, NEON_ENC_LANE)
 492         (NEON_ENC_DUP): Define meaningful helper macros to look up values in
 493         neon_enc_tab.
 494         (neon_shape): Enumerate shapes (permitted register widths, etc.) for
 495         Neon instructions.
 496         (neon_type_mask): New. Compact type representation for type checking.
 497         (N_SU_ALL, N_SU_32, N_SU_16_64, N_SUF_32, N_I_ALL, N_IF_32): Common
 498         permitted type combinations.
 499         (N_IGNORE_TYPE): New macro.
 500         (neon_check_shape): New function. Check an instruction shape for
 501         multiple alternatives. Return the specific shape for the current
 502         instruction.
 503         (neon_modify_type_size): New function. Modify a vector type and size,
 504         depending on the bit mask in argument 1.
 505         (neon_type_promote): New function. Convert a given "key" type (of an
 506         operand) into the correct type for a different operand, based on a bit
 507         mask.
 508         (type_chk_of_el_type): New function. Convert a type and size into the
 509         compact representation used for type checking.
 510         (el_type_of_type_ckh): New function. Reverse of above (only when a
 511         single bit is set in the bit mask).
 512         (modify_types_allowed): New function. Alter a mask of allowed types
 513         based on a bit mask of modifications.
 514         (neon_check_type): New function. Check the type of the current
 515         instruction against the variable argument list. The "key" type of the
 516         instruction is returned.
 517         (neon_dp_fixup): New function. Fill in and modify instruction bits for
 518         a Neon data-processing instruction depending on whether we're in ARM
 519         mode or Thumb-2 mode.
 520         (neon_logbits): New function.
 521         (neon_three_same, neon_two_same, do_neon_dyadic_i_su)
 522         (do_neon_dyadic_i64_su, neon_imm_shift, do_neon_shl_imm)
 523         (do_neon_qshl_imm, neon_cmode_for_logic_imm, neon_bits_same_in_bytes)
 524         (neon_squash_bits, neon_is_quarter_float, neon_qfloat_bits)
 525         (neon_cmode_for_move_imm, neon_write_immbits, neon_invert_size)
 526         (do_neon_logic, do_neon_bitfield, neon_dyadic, do_neon_dyadic_if_su)
 527         (do_neon_dyadic_if_su_d, do_neon_dyadic_if_i, do_neon_dyadic_if_i_d)
 528         (do_neon_addsub_if_i, neon_exchange_operands, neon_compare)
 529         (do_neon_cmp, do_neon_cmp_inv, do_neon_ceq, neon_scalar_for_mul)
 530         (neon_mul_mac, do_neon_mac_maybe_scalar, do_neon_tst, do_neon_mul)
 531         (do_neon_qdmulh, do_neon_fcmp_absolute, do_neon_fcmp_absolute_inv)
 532         (do_neon_step, do_neon_abs_neg, do_neon_sli, do_neon_sri)
 533         (do_neon_qshlu_imm, do_neon_qmovn, do_neon_qmovun)
 534         (do_neon_rshift_sat_narrow, do_neon_rshift_sat_narrow_u, do_neon_movn)
 535         (do_neon_rshift_narrow, do_neon_shll, neon_cvt_flavour, do_neon_cvt)
 536         (neon_move_immediate, do_neon_mvn, neon_mixed_length)
 537         (do_neon_dyadic_long, do_neon_abal, neon_mac_reg_scalar_long)
 538         (do_neon_mac_maybe_scalar_long, do_neon_dyadic_wide, do_neon_vmull)
 539         (do_neon_ext, do_neon_rev, do_neon_dup, do_neon_mov)
 540         (do_neon_rshift_round_imm, do_neon_movl, do_neon_trn, do_neon_zip_uzp)
 541         (do_neon_sat_abs_neg, do_neon_pair_long, do_neon_recip_est)
 542         (do_neon_cls, do_neon_clz, do_neon_cnt, do_neon_swp, do_neon_tbl_tbx)
 543         (do_neon_ldm_stm, do_neon_ldr_str, do_neon_ld_st_interleave)
 544         (neon_alignment_bit, do_neon_ld_st_lane, do_neon_ld_dup)
 545         (do_neon_ldx_stx): New functions. Neon bit encoding and encoding
 546         helpers.
 547         (parse_neon_type): New function. Parse Neon type specifier.
 548         (opcode_lookup): Allow parsing of Neon type specifiers.
 549         (REGNUM2, REGSETH, REGSET2): New macros.
 550         (reg_names): Add new VFPv3 and Neon registers.
 551         (NUF, nUF, NCE, nCE): New macros for opcode table.
 552         (insns): More VFP registers allowed in fcpyd, fmdhr, fmdlr, fmrdh,
 553         fmrdl, fabsd, fnegd, fsqrtd, faddd, fsubd, fmuld, fdivd, fmacd, fmscd,
 554         fnmuld, fnmacd, fnmscd, fcmpd, fcmpzd, fcmped, fcmpezd, fmdrr, fmrrd.
 555         Add Neon instructions vaba, vhadd, vrhadd, vhsub, vqadd, vqsub, vrshl,
 556         vqrshl, vshl, vqshl{u}, vand, vbic, vorr, vorn, veor, vbsl, vbit, vbif,
 557         vabd, vmax, vmin, vcge, vcgt, vclt, vcle, vceq, vpmax, vpmin, vmla,
 558         vmls, vpadd, vadd, vsub, vtst, vmul, vqdmulh, vqrdmulh, vacge, vacgt,
 559         vaclt, vacle, vrecps, vrsqrts, vabs, vneg, v{r}shr,  v{r}sra, vsli,
 560         vsri, vqshrn, vq{r}shr{u}n, v{r}shrn, vshll, vcvt, vmov, vmvn, vabal,
 561         vabdl, vaddl, vsubl, vmlal, vmlsl, vaddw, vsubw, v{r}addhn, v{r}subhn,
 562         vqdmlal, vqdmlsl, vqdmull, vmull, vext, vrev64, vrev32, vrev16, vdup,
 563         vmovl, v{q}movn, vzip, vuzp, vqabs, vqneg, vpadal, vpaddl, vrecpe,
 564         vrsqrte, vcls, vclz, vcnt, vswp, vtrn, vtbl, vtbx, vldm, vstm, vldr,
 565         vstr, vld[1234], vst[1234], fconst[sd], f[us][lh]to[sd],
 566         fto[us][lh][sd].
 567         (tc_arm_regname_to_dw2regnum): Update for arm_reg_parse args.
 568         (arm_cpu_option_table): Add Neon and VFPv3 to Cortex-A8.
 569         (arm_option_cpu_value): Add vfp3 and neon.
 570         (aeabi_set_public_attributes): Support VFPv3 and NEON attributes. Fix
 571         VFPv1 attribute.
 572
 573 2006-04-25  Bob Wilson  <bob.wilson@acm.org>
 574
 575         * config/xtensa-relax.c (widen_spec_list): Use new "WIDE.<opcode>"
 576         syntax instead of hardcoded opcodes with ".w18" suffixes.
 577         (wide_branch_opcode): New.
 578         (build_transition): Use it to check for wide branch opcodes with
 579         either ".w18" or ".w15" suffixes.
 580
 581 2006-04-25  Bob Wilson  <bob.wilson@acm.org>
 582
 583         * config/tc-xtensa.c (xtensa_create_literal_symbol,
 584         xg_assemble_literal, xg_assemble_literal_space): Do not set the
 585         frag's is_literal flag.
 586
 587 2006-04-25  Bob Wilson  <bob.wilson@acm.org>
 588
 589         * config/xtensa-relax.c (XCHAL_HAVE_WIDE_BRANCHES): Provide default.
 590
 591 2006-04-23  Kazu Hirata  <kazu@codesourcery.com>
 592
 593         * config/obj-coff.c, config/tc-arm.c, config/tc-bfin.c,
 594         config/tc-cris.c, config/tc-crx.c, config/tc-i386.c,
 595         config/tc-ia64.c, config/tc-maxq.c, config/tc-maxq.h,
 596         config/tc-mips.c, config/tc-msp430.c, config/tc-sh.c,
 597         config/tc-tic4x.c, config/tc-xtensa.c: Fix comment typos.
 598
 599 2005-04-20  Paul Brook  <paul@codesourcery.com>
 600
 601         * config/tc-arm.c (s_arm_arch, s_arm_cpu, s_arm_fpu): Enable for
 602         all targets.
 603         (md_pseudo_table): Enable .arch, .cpu and .fpu for all targets.
 604
 605 2006-04-19  Alan Modra  <amodra@bigpond.net.au>
 606
 607         * Makefile.am (CPU_TYPES): Add maxq and mt.  Sort.
 608         (CPU_OBJ_VALID): Change sense of COFF test to default to invalid.
 609         Make some cpus unsupported on ELF.  Run "make dep-am".
 610         * Makefile.in: Regenerate.
 611
 612 2006-04-19  Alan Modra  <amodra@bigpond.net.au>
 613
 614         * configure.in (--enable-targets): Indent help message.
 615         * configure: Regenerate.
 616
 617 2006-04-18  H.J. Lu  <hongjiu.lu@intel.com>
 618
 619         PR gas/2533
 620         * config/tc-i386.c (i386_immediate): Check illegal immediate
 621         register operand.
 622
 623 2006-04-18  Alan Modra  <amodra@bigpond.net.au>
 624
 625         * config/tc-i386.c: Formatting.
 626         (output_disp, output_imm): ISO C90 params.
 627
 628         * frags.c (frag_offset_fixed_p): Constify args.
 629         * frags.h (frag_offset_fixed_p): Ditto.
 630
 631         * config/tc-dlx.h (tc_coff_symbol_emit_hook): Delete.
 632         (COFF_MAGIC): Delete.
 633
 634         * config/tc-xc16x.h (TC_LINKRELAX_FIXUP): Delete.
 635
 636 2006-04-16  Daniel Jacobowitz  <dan@codesourcery.com>
 637
 638         * po/POTFILES.in: Regenerated.
 639
 640 2006-04-16  Mark Mitchell  <mark@codesourcery.com>
 641
 642         * doc/as.texinfo: Mention that some .type syntaxes are not
 643         supported on all architectures.
 644
 645 2006-04-14  Sterling Augustine  <sterling@tensilica.com>
 646
 647         * config/tc-xtensa.c (emit_single_op): Do not relax MOVI
 648         instructions when such transformations have been disabled.
 649
 650 2006-04-10  Sterling Augustine  <sterling@tensilica.com>
 651
 652         * config/tc-xtensa.c (xg_assemble_vliw_tokens): Record loop target
 653         symbols in RELAX[_CHECK]_ALIGN_NEXT_OPCODE frags.
 654         (xtensa_fix_close_loop_end_frags): Use the recorded values instead of
 655         decoding the loop instructions.  Remove current_offset variable.
 656         (xtensa_fix_short_loop_frags): Likewise.
 657         (min_bytes_to_other_loop_end): Remove current_offset argument.
 658
 659 2006-04-09  Arnold Metselaar  <arnold.metselaar@planet.nl>
 660
 661         * config/tc-z80.c (z80_optimize_expr): Removed.
 662         * config/tc-z80.h (z80_optimize_expr, md_optimize_expr): Removed.
 663
 664 2006-04-07  Joerg Wunsch <j.gnu@uriah.heep.sax.de>
 665
 666         * gas/config/tc-avr.c (mcu_types): Add support for attiny261,
 667         attiny461, attiny861, attiny25, attiny45, attiny85,attiny24,
 668         attiny44, attiny84, at90pwm2, at90pwm3, atmega164, atmega324,
 669         atmega644, atmega329, atmega3290, atmega649, atmega6490,
 670         atmega406, atmega640, atmega1280, atmega1281, at90can32,
 671         at90can64, at90usb646, at90usb647, at90usb1286 and
 672         at90usb1287.
 673         Move atmega48 and atmega88 from AVR_ISA_M8 to AVR_ISA_PWMx.
 674
 675 2006-04-07  Paul Brook  <paul@codesourcery.com>
 676
 677         * config/tc-arm.c (parse_operands): Set default error message.
 678
 679 2006-04-07  Paul Brook  <paul@codesourcery.com>
 680
 681         * config/tc-arm.c (parse_tb): Set inst.error before returning FAIL.
 682
 683 2006-04-07  Paul Brook  <paul@codesourcery.com>
 684
 685         * config/tc-arm.c (md_apply_fix): Set H bit on blx instruction.
 686
 687 2006-04-07  Paul Brook  <paul@codesourcery.com>
 688
 689         * config/tc-arm.c (THUMB2_LOAD_BIT): Define.
 690         (move_or_literal_pool): Handle Thumb-2 instructions.
 691         (do_t_ldst): Call move_or_literal_pool for =N addressing modes.
 692
 693 2006-04-07  Alan Modra  <amodra@bigpond.net.au>
 694
 695         PR 2512.
 696         * config/tc-i386.c (match_template): Move 64-bit operand tests
 697         inside loop.
 698
 699 2006-04-06  Carlos O'Donell  <carlos@codesourcery.com>
 700
 701         * po/Make-in: Add install-html target.
 702         * Makefile.am: Add install-html and install-html-recursive targets.
 703         * Makefile.in: Regenerate.
 704         * configure.in: AC_SUBST datarootdir, docdir, htmldir.
 705         * configure: Regenerate.
 706         * doc/Makefile.am: Add install-html and install-html-am targets.
 707         * doc/Makefile.in: Regenerate.
 708
 709 2006-04-06  Alan Modra  <amodra@bigpond.net.au>
 710
 711         * frags.c (frag_offset_fixed_p): Reinitialise offset before
 712         second scan.
 713
 714 2006-04-05  Richard Sandiford  <richard@codesourcery.com>
 715             Daniel Jacobowitz  <dan@codesourcery.com>
 716
 717         * config/tc-sparc.c (sparc_target_format): Handle TE_VXWORKS.
 718         (GOTT_BASE, GOTT_INDEX): New.
 719         (tc_gen_reloc): Don't alter relocations against GOTT_BASE and
 720         GOTT_INDEX when generating VxWorks PIC.
 721         * configure.tgt (sparc*-*-vxworks*): Remove this special case;
 722         use the generic *-*-vxworks* stanza instead.
 723
 724 2006-04-04  Alan Modra  <amodra@bigpond.net.au>
 725
 726         PR 997
 727         * frags.c (frag_offset_fixed_p): New function.
 728         * frags.h (frag_offset_fixed_p): Declare.
 729         * expr.c (expr): Use frag_offset_fixed_p when simplifying subtraction.
 730         (resolve_expression): Likewise.
 731
 732 2006-04-03  Sterling Augustine  <sterling@tensilica.com>
 733
 734         * config/tc-xtensa.c (init_op_placement_info_table): Check for formats
 735         of the same length but different numbers of slots.
 736
 737 2006-03-30  Andreas Schwab  <schwab@suse.de>
 738
 739         * configure.in: Fix help string for --enable-targets option.
 740         * configure: Regenerate.
 741
 742 2006-03-28  Nathan Sidwell  <nathan@codesourcery.com>
 743
 744         * gas/config/tc-m68k.c (find_cf_chip): Merge into ...
 745         (m68k_ip): ... here.  Use for all chips.  Protect against buffer
 746         overrun and avoid excessive copying.
 747
 748         * config/tc-m68k.c (m68000_control_regs, m68010_control_regs,
 749         m68020_control_regs, m68040_control_regs, m68060_control_regs,
 750         mcf_control_regs, mcf5208_control_regs, mcf5213_control_regs,
 751         mcf5329_control_regs, mcf5249_control_regs, mcf528x_control_regs,
 752         mcfv4e_control_regs, m68010_control_regs): Rename and reorder to ...
 753         (m68000_ctrl, m68010_ctrl, m68020_ctrl, m68040_ctrl, m68060_ctrl,
 754         mcf_ctrl, mcf5208_ctrl, mcf5213_ctrl, mcf5235_ctrl, mcf5249_ctrl,
 755         mcf5216_ctrl, mcf5250_ctrl, mcf5271_ctrl, mcf5272_ctrl,
 756         mcf5282_ctrl, mcfv4e_ctrl): ... these.
 757         (mcf5275_ctrl, mcf5329_ctrl, mcf5373_ctrl): New.
 758         (struct m68k_cpu): Change chip field to control_regs.
 759         (current_chip): Remove.
 760         (control_regs): New.
 761         (m68k_archs, m68k_extensions): Adjust.
 762         (m68k_cpus): Reorder to be in cpu number order.  Adjust.
 763         (CPU_ALLOW_MC, CPU_ALLOW_NEGATION): Remove.
 764         (find_cf_chip): Reimplement for new organization of cpu table.
 765         (select_control_regs): Remove.
 766         (mri_chip): Adjust.
 767         (struct save_opts): Save control regs, not chip.
 768         (s_save, s_restore): Adjust.
 769         (m68k_lookup_cpu): Give deprecated warning when necessary.
 770         (m68k_init_arch): Adjust.
 771         (md_show_usage): Adjust for new cpu table organization.
 772
 773 2006-03-25  Bernd Schmidt  <bernd.schmidt@analog.com>
 774
 775         * config/bfin-defs.h (Expr_Node_Type enum): Add Expr_Node_GOT_Reloc.
 776         * config/bfin-lex.l: Recognize GOT17M4 and FUNCDESC_GOT17M4.
 777         * config/bfin-parse.y: Include "libbfd.h", "elf/common.h" and
 778         "elf/bfin.h".
 779         (GOT17M4, FUNCDESC_GOT17M4): New tokens of type <value>.
 780         (any_gotrel): New rule.
 781         (got): Use it, and create Expr_Node_GOT_Reloc nodes.
 782         * config/tc-bfin.c: Include "libbfd.h", "elf/common.h" and
 783         "elf/bfin.h".
 784         (DEFAULT_FLAGS, bfin_flags, bfin_pic_flag): New.
 785         (bfin_pic_ptr): New function.
 786         (md_pseudo_table): Add it for ".picptr".
 787         (OPTION_FDPIC): New macro.
 788         (md_longopts): Add -mfdpic.
 789         (md_parse_option): Handle it.
 790         (md_begin): Set BFD flags.
 791         (md_apply_fix3, bfin_fix_adjustable): Handle new relocs.
 792         (bfin_gen_ldstidxi): Adjust to match the trees that the parser gives
 793         us for GOT relocs.
 794         * Makefile.am (bfin-parse.o): Update dependencies.
 795         (DEPTC_bfin_elf): Likewise.
 796         * Makefile.in: Regenerate.
 797
 798 2006-03-25  Richard Sandiford  <richard@codesourcery.com>
 799
 800         * config/tc-m68k.c (m68k_cpus): Change cpu_cf5208 entries to use
 801         mcfemac instead of mcfmac.
 802
 803 2006-03-23  Michael Matz  <matz@suse.de>
 804
 805         * config/tc-i386.c (type_names): Correct placement of 'static'.
 806         (reloc): Map some more relocs to their 64 bit counterpart when
 807         size is 8.
 808         (output_insn): Work around breakage if DEBUG386 is defined.
 809         (output_disp): A BFD_RELOC_64 with GOT_symbol as operand also
 810         needs to be mapped to BFD_RELOC_X86_64_GOTPC64 or
 811         BFD_RELOC_X86_64_GOTPC32.  Also x86-64 handles pcrel addressing
 812         different from i386.
 813         (output_imm): Ditto.
 814         (lex_got): Recognize @PLTOFF and @GOTPLT.  Make @GOT accept also
 815         Imm64.
 816         (md_convert_frag): Jumps can now be larger than 2GB away, error
 817         out in that case.
 818         (tc_gen_reloc): New relocs are passed through.  BFD_RELOC_64
 819         and BFD_RELOC_64_PCREL are mapped to BFD_RELOC_X86_64_GOTPC64.
 820
 821 2006-03-22  Richard Sandiford  <richard@codesourcery.com>
 822             Daniel Jacobowitz  <dan@codesourcery.com>
 823             Phil Edwards  <phil@codesourcery.com>
 824             Zack Weinberg  <zack@codesourcery.com>
 825             Mark Mitchell  <mark@codesourcery.com>
 826             Nathan Sidwell  <nathan@codesourcery.com>
 827
 828         * config/tc-mips.c (mips_target_format): Handle vxworks targets.
 829         (md_begin): Complain about -G being used for PIC.  Don't change
 830         the text, data and bss alignments on VxWorks.
 831         (reloc_needs_lo_p): Don't return true for R_MIPS_GOT16 when
 832         generating VxWorks PIC.
 833         (load_address): Extend SVR4_PIC handling to VXWORKS_PIC.
 834         (macro): Likewise, but do not treat la $25 specially for
 835         VxWorks PIC, and do not handle jal.
 836         (OPTION_MVXWORKS_PIC): New macro.
 837         (md_longopts): Add -mvxworks-pic.
 838         (md_parse_option): Don't complain about using PIC and -G together here.
 839         Handle OPTION_MVXWORKS_PIC.
 840         (md_estimate_size_before_relax): Always use the first relaxation
 841         sequence on VxWorks.
 842         * config/tc-mips.h (VXWORKS_PIC): New.
 843
 844 2006-03-21  Paul Brook  <paul@codesourcery.com>
 845
 846         * config/tc-arm.c (md_apply_fix): Fix typo in offset mask.
 847
 848 2006-03-21  Sterling Augustine  <sterling@tensilica.com>
 849
 850         * config/tc-xtensa.c (enforce_three_byte_loop_align): New flag.
 851         (xtensa_setup_hw_workarounds): Set this new flag for older hardware.
 852         (get_loop_align_size): New.
 853         (xtensa_end): Skip xtensa_mark_narrow_branches when not aligning.
 854         (xtensa_mark_zcl_first_insns): Prevent widening of first loop frag.
 855         (get_text_align_power): Rewrite to handle inputs in the range 2-8.
 856         (get_noop_aligned_address): Use get_loop_align_size.
 857         (get_aligned_diff): Likewise.
 858
 859 2006-03-21  Paul Brook  <paul@codesourcery.com>
 860
 861         * config/tc-arm.c (insns): Correct opcodes for ldrbt and strbt.
 862
 863 2006-03-20  Paul Brook  <paul@codesourcery.com>
 864
 865         * config/tc-arm.c (BAD_BRANCH, BAD_NOT_IT): Define.
 866         (do_t_branch): Encode branches inside IT blocks as unconditional.
 867         (do_t_cps): New function.
 868         (do_t_blx, do_t_bkpt, do_t_branch23, do_t_bx, do_t_bxj, do_t_cpsi,
 869         do_t_czb, do_t_it, do_t_setend, do_t_tb): Add IT constaints.
 870         (opcode_lookup): Allow conditional suffixes on all instructions in
 871         Thumb mode.
 872         (md_assemble): Advance condexec state before checking for errors.
 873         (insns): Use do_t_cps.
 874
 875 2006-03-20  Paul Brook  <paul@codesourcery.com>
 876
 877         * config/tc-arm.c (output_relax_insn): Call dwarf2_emit_insn before
 878         outputting the insn.
 879
 880 2006-03-18  Jan-Benedict Glaw  <jbglaw@lug-owl.de>
 881
 882         * config/tc-vax.c: Update copyright year.
 883         * config/tc-vax.h: Likewise.
 884
 885 2006-03-18  Jan-Benedict Glaw  <jbglaw@lug-owl.de>
 886
 887         * config/tc-vax.c (md_chars_to_number): Used only locally, so
 888         make it static.
 889         * config/tc-vax.h (md_chars_to_number): Remove obsolete declaration.
 890
 891 2006-03-17  Paul Brook  <paul@codesourcery.com>
 892
 893         * config/tc-arm.c (insns): Add ldm and stm.
 894
 895 2006-03-17  Ben Elliston  <bje@au.ibm.com>
 896
 897         PR gas/2446
 898         * doc/as.texinfo (Ident): Document this directive more thoroughly.
 899
 900 2006-03-16  Paul Brook  <paul@codesourcery.com>
 901
 902         * config/tc-arm.c (insns): Add "svc".
 903
 904 2006-03-13  Bob Wilson  <bob.wilson@acm.org>
 905
 906         * config/tc-xtensa.c (xg_translate_sysreg_op): Remove has_underbar
 907         flag and avoid double underscore prefixes.
 908
 909 2006-03-10  Paul Brook  <paul@codesourcery.com>
 910
 911         * config/tc-arm.c (md_begin): Handle EABIv5.
 912         (arm_eabis): Add EF_ARM_EABI_VER5.
 913         * doc/c-arm.texi: Document -meabi=5.
 914
 915 2006-03-10  Ben Elliston  <bje@au.ibm.com>
 916
 917         * app.c (do_scrub_chars): Simplify string handling.
 918
 919 2006-03-07  Richard Sandiford  <richard@codesourcery.com>
 920             Daniel Jacobowitz  <dan@codesourcery.com>
 921             Zack Weinberg  <zack@codesourcery.com>
 922             Nathan Sidwell  <nathan@codesourcery.com>
 923             Paul Brook  <paul@codesourcery.com>
 924             Ricardo Anguiano  <anguiano@codesourcery.com>
 925             Phil Edwards  <phil@codesourcery.com>
 926
 927         * config/tc-arm.c (md_apply_fix): Install a value of zero into a
 928         BFD_RELOC_ARM_OFFSET_IMM field if we're going to generate a RELA
 929         R_ARM_ABS12 reloc.
 930         (tc_gen_reloc): Keep the original fx_offset for RELA pc-relative
 931         relocs, but adjust by md_pcrel_from_section.  Create R_ARM_ABS12
 932         relocations for BFD_RELOC_ARM_OFFSET_IMM on RELA targets.
 933
 934 2006-03-06  Bob Wilson  <bob.wilson@acm.org>
 935
 936         * config/tc-xtensa.c (xtensa_post_relax_hook): Generate literal tables
 937         even when using the text-section-literals option.
 938
 939 2006-03-06  Nathan Sidwell  <nathan@codesourcery.com>
 940
 941         * config/tc-m68k.c (m68k_extensions): Allow 'float' on both m68k
 942         and cf.
 943         (m68k_ip): <case 'J'> Check we have some control regs.
 944         (md_parse_option): Allow raw arch switch.
 945         (m68k_init_arch): Better detection of arch/cpu mismatch.  Detect
 946         whether 68881 or cfloat was meant by -mfloat.
 947         (md_show_usage): Adjust extension display.
 948         (m68k_elf_final_processing): Adjust.
 949
 950 2006-03-03  Bjoern Haase  <bjoern.m.haase@web.de>
 951
 952         * config/tc-avr.c (avr_mod_hash_value): New function.
 953         (md_apply_fix, exp_mod): Use BFD_RELOC_HH8_LDI and
 954         BFD_RELOC_MS8_LDI for hlo8() and hhi8()
 955         (md_begin): Set linkrelax variable to 1, use avr_mod_hash_value
 956         instead of int avr_ldi_expression: use avr_mod_hash_value instead
 957         of (int).
 958         (tc_gen_reloc): Handle substractions of symbols, if possible do
 959         fixups, abort otherwise.
 960         * config/tc-avr.h (TC_LINKRELAX_FIXUP, TC_VALIDATE_FIX,
 961         tc_fix_adjustable): Define.
 962
 963 2006-03-02  James E Wilson  <wilson@specifix.com>
 964
 965         * config/tc-ia64.c (emit_one_bundle): For IA64_OPCODE_LAST, if we
 966         change the template, then clear md.slot[curr].end_of_insn_group.
 967
 968 2006-02-28  Jan Beulich  <jbeulich@novell.com>
 969
 970         * macro.c (get_any_string): Don't insert quotes for <>-quoted input.
 971
 972 2006-02-28  Jan Beulich  <jbeulich@novell.com>
 973
 974         PR/1070
 975         * macro.c (getstring): Don't treat parentheses special anymore.
 976         (get_any_string): Don't consider '(' and ')' as quoting anymore.
 977         Special-case '(', ')', '[', and ']' when dealing with non-quoting
 978         characters.
 979
 980 2006-02-28  Mat <mat@csail.mit.edu>
 981
 982         * dwarf2dbg.c (get_filenum): Don't inadvertently decrease files_in_use.
 983
 984 2006-02-27  Jakub Jelinek  <jakub@redhat.com>
 985
 986         * dw2gencfi.c (struct fde_entry, struct cie_entry): Add signal_frame
 987         field.
 988         (CFI_signal_frame): Define.
 989         (cfi_pseudo_table): Add .cfi_signal_frame.
 990         (dot_cfi): Handle CFI_signal_frame.
 991         (output_cie): Handle cie->signal_frame.
 992         (select_cie_for_fde): Don't share CIE if signal_frame flag is
 993         different.  Copy signal_frame from FDE to newly created CIE.
 994         * doc/as.texinfo: Document .cfi_signal_frame.
 995
 996 2006-02-27  Carlos O'Donell  <carlos@codesourcery.com>
 997
 998         * doc/Makefile.am: Add html target.
 999         * doc/Makefile.in: Regenerate.
1000         * po/Make-in: Add html target.
1001
1002 2006-02-27  H.J. Lu <hongjiu.lu@intel.com>
1003
1004         * config/tc-i386.c (output_insn): Support Intel Merom New
1005         Instructions.
1006
1007         * config/tc-i386.h (CpuMNI): New.
1008         (CpuUnknownFlags): Add CpuMNI.
1009
1010 2006-02-24  David S. Miller  <davem@sunset.davemloft.net>
1011
1012         * config/tc-sparc.c (priv_reg_table): Add entry for "gl".
1013         (hpriv_reg_table): New table for hyperprivileged registers.
1014         (sparc_ip): New cases '$' and '%' for wrhpr/rdhpr hyperprivileged
1015         register encoding.
1016
1017 2006-02-24  DJ Delorie  <dj@redhat.com>
1018
1019         * config/tc-m32c.h (md_apply_fix): Define to m32c_apply_fix.
1020         (tc_gen_reloc): Don't define.
1021         * config/tc-m32c.c (rl_for, relaxable): New convenience macros.
1022         (OPTION_LINKRELAX): New.
1023         (md_longopts): Add it.
1024         (m32c_relax): New.
1025         (md_parse_options): Set it.
1026         (md_assemble): Emit relaxation relocs as needed.
1027         (md_convert_frag): Emit relaxation relocs as needed.
1028         (md_cgen_lookup_reloc): Add LAB_8_8 and LAB_8_16.
1029         (m32c_apply_fix): New.
1030         (tc_gen_reloc): New.
1031         (m32c_force_relocation): Force out jump relocs when relaxing.
1032         (m32c_fix_adjustable): Return false if relaxing.
1033
1034 2006-02-24  Paul Brook  <paul@codesourcery.com>
1035
1036         * config/arm/tc-arm.c (arm_ext_v6_notm, arm_ext_div, arm_ext_v7,
1037         arm_ext_v7a, arm_ext_v7r, arm_ext_v7m): New variables.
1038         (struct asm_barrier_opt): Define.
1039         (arm_v7m_psr_hsh, arm_barrier_opt_hsh): New variables.
1040         (parse_psr): Accept V7M psr names.
1041         (parse_barrier): New function.
1042         (enum operand_parse_code): Add OP_oBARRIER.
1043         (parse_operands): Implement OP_oBARRIER.
1044         (do_barrier): New function.
1045         (do_dbg, do_pli, do_t_barrier, do_t_dbg, do_t_div): New functions.
1046         (do_t_cpsi): Add V7M restrictions.
1047         (do_t_mrs, do_t_msr): Validate V7M variants.
1048         (md_assemble): Check for NULL variants.
1049         (v7m_psrs, barrier_opt_names): New tables.
1050         (insns): Add V7 instructions.  Mark V6 instructions absent from V7M.
1051         (md_begin): Initialize arm_v7m_psr_hsh and arm_barrier_opt_hsh.
1052         (arm_cpu_option_table): Add Cortex-M3, R4 and A8.
1053         (arm_arch_option_table): Add armv7, armv7a, armv7r and armv7m.
1054         (struct cpu_arch_ver_table): Define.
1055         (cpu_arch_ver): New.
1056         (aeabi_set_public_attributes): Use cpu_arch_ver.  Set
1057         Tag_CPU_arch_profile.
1058         * doc/c-arm.texi: Document new cpu and arch options.
1059
1060 2006-02-23  H.J. Lu  <hongjiu.lu@intel.com>
1061
1062         * config/tc-ia64.c (operand_match): Handle IA64_OPND_IMMU5b.
1063
1064 2006-02-23  H.J. Lu  <hongjiu.lu@intel.com>
1065
1066         * config/tc-ia64.c: Update copyright years.
1067
1068 2006-02-22  H.J. Lu  <hongjiu.lu@intel.com>
1069
1070         * config/tc-ia64.c (specify_resource): Add the rule 17 from
1071         SDM 2.2.
1072
1073 2005-02-22  Paul Brook  <paul@codesourcery.com>
1074
1075         * config/tc-arm.c (do_pld): Remove incorrect write to
1076         inst.instruction.
1077         (encode_thumb32_addr_mode): Use correct operand.
1078
1079 2006-02-21  Paul Brook  <paul@codesourcery.com>
1080
1081         * config/tc-arm.c (md_apply_fix): Fix off-by-one errors.
1082
1083 2006-02-17  Shrirang Khisti  <shrirangk@kpitcummins.com>
1084             Anil Paranjape   <anilp1@kpitcummins.com>
1085             Shilin Shakti    <shilins@kpitcummins.com>
1086
1087         * Makefile.am: Add xc16x related entry.
1088         * Makefile.in: Regenerate.
1089         * configure.in: Added xc16x related entry.
1090         * configure: Regenerate.
1091         * config/tc-xc16x.h: New file
1092         * config/tc-xc16x.c: New file
1093         * doc/c-xc16x.texi: New file for xc16x
1094         * doc/all.texi: Entry for xc16x
1095         * doc/Makefile.texi: Added c-xc16x.texi
1096         * NEWS: Announce the support for the new target.
1097
1098 2006-02-16  Nick Hudson  <nick.hudson@dsl.pipex.com>
1099
1100         * configure.tgt: set emulation for mips-*-netbsd*
1101
1102 2006-02-14  Jakub Jelinek  <jakub@redhat.com>
1103
1104         * config.in: Rebuilt.
1105
1106 2006-02-13  Bob Wilson  <bob.wilson@acm.org>
1107
1108         * config/tc-xtensa.c (xg_add_opcode_fix): Number operands starting
1109         from 1, not 0, in error messages.
1110         (md_assemble): Simplify special-case check for ENTRY instructions.
1111         (tinsn_has_invalid_symbolic_operands): Do not include opcode and
1112         operand in error message.
1113
1114 2006-02-13  Joseph S. Myers  <joseph@codesourcery.com>
1115
1116         * configure.tgt (arm-*-linux-gnueabi*): Change to
1117         arm-*-linux-*eabi*.
1118
1119 2006-02-10  Nick Clifton  <nickc@redhat.com>
1120
1121         * config/tc-crx.c (check_range): Ensure that the sign bit of a
1122         32-bit value is propagated into the upper bits of a 64-bit long.
1123
1124         * config/tc-arc.c (init_opcode_tables): Fix cast.
1125         (arc_extoper, md_operand): Likewise.
1126
1127 2006-02-09  David Heine  <dlheine@tensilica.com>
1128
1129         * config/tc-xtensa.c (xg_assembly_relax): Increment steps_taken for
1130         each relaxation step.
1131
1132 2006-02-09  Eric Botcazou  <ebotcazou@libertysurf.fr>
1133
1134         * configure.in (CHECK_DECLS): Add vsnprintf.
1135         * configure: Regenerate.
1136         * messages.c (errno.h, stdarg.h, varargs.h, va_list): Do not
1137         include/declare here, but...
1138         * as.h: Move code detecting VARARGS idiom to the top.
1139         (errno.h, stdarg.h, varargs.h, va_list): ...here.
1140         (vsnprintf): Declare if not already declared.
1141
1142 2006-02-08  H.J. Lu  <hongjiu.lu@intel.com>
1143
1144         * as.c (close_output_file): New.
1145         (main): Register close_output_file with xatexit before
1146         dump_statistics. Don't call output_file_close.
1147
1148 2006-02-07  Nathan Sidwell  <nathan@codesourcery.com>
1149
1150         * config/tc-m68k.c (mcf5208_control_regs, mcf5213_control_regs,
1151         mcf5329_control_regs): New.
1152         (not_current_architecture, selected_arch, selected_cpu): New.
1153         (m68k_archs, m68k_extensions): New.
1154         (archs): Renamed to ...
1155         (m68k_cpus): ... here.  Adjust.
1156         (n_arches): Remove.
1157         (md_pseudo_table): Add arch and cpu directives.
1158         (find_cf_chip, m68k_ip): Adjust table scanning.
1159         (no_68851, no_68881): Remove.
1160         (md_assemble): Lazily initialize.
1161         (select_control_regs): Adjust cpu names. Add 5208, 5213, 5329.
1162         (md_init_after_args): Move functionality to m68k_init_arch.
1163         (mri_chip): Adjust table scanning.
1164         (md_parse_option): Reimplement 'm' processing to add -march & -mcpu
1165         options with saner parsing.
1166         (m68k_lookup_cpu, m68k_set_arch, m68k_set_cpu, m68k_set_extension,
1167         m68k_init_arch): New.
1168         (s_m68k_cpu, s_m68k_arch): New.
1169         (md_show_usage): Adjust.
1170         (m68k_elf_final_processing): Set CF EF flags.
1171         * config/tc-m68k.h (m68k_init_after_args): Remove.
1172         (tc_init_after_args): Remove.
1173         * doc/c-m68k.texi (M68K-Opts): Document -march, -mcpu options.
1174         (M68k-Directives): Document .arch and .cpu directives.
1175
1176 2006-02-05  Arnold Metselaar  <arnold.metselaar@planet.nl>
1177
1178         * config/tc-z80.c (z80_start_line_hook): allow .equ and .defl as
1179         synonyms for equ and defl.
1180         (z80_cons_fix_new): New function.
1181         (emit_byte): Disallow relative jumps to absolute locations.
1182         (emit_data): Only handle defb, prototype changed, because defb is
1183         now handled as pseudo-op rather than an instruction.
1184         (instab): Entries for defb,defw,db,dw moved from here...
1185         (md_pseudo_table): ... to here, use generic cons() for defw,dw.
1186         Add entries for def24,def32,d24,d32.
1187         (md_assemble): Improved error handling.
1188         (md_apply_fix): New case BFD_RELOC_24, set fixP->fx_no_overflow to one.
1189         * config/tc-z80.h (TC_CONS_FIX_NEW): Define.
1190         (z80_cons_fix_new): Declare.
1191         * doc/c-z80.texi (defb, db): Mention warning on overflow.
1192         (def24,d24,def32,d32): New pseudo-ops.
1193
1194 2006-02-02  Paul Brook  <paul@codesourcery.com>
1195
1196         * config/tc-arm.c (do_shift): Remove Thumb-1 constraint.
1197
1198 2005-02-02  Paul Brook  <paul@codesourcery.com>
1199
1200         * config/tc-arm.c (T2_OPCODE_MASK, T2_DATA_OP_SHIFT, T2_OPCODE_AND,
1201         T2_OPCODE_BIC, T2_OPCODE_ORR, T2_OPCODE_ORN, T2_OPCODE_EOR,
1202         T2_OPCODE_ADD, T2_OPCODE_ADC, T2_OPCODE_SBC, T2_OPCODE_SUB,
1203         T2_OPCODE_RSB): Define.
1204         (thumb32_negate_data_op): New function.
1205         (md_apply_fix): Use it.
1206
1207 2006-01-31  Bob Wilson  <bob.wilson@acm.org>
1208
1209         * config/xtensa-istack.h (TInsn): Remove record_fix and sub_symbol
1210         fields.
1211         * config/tc-xtensa.h (xtensa_frag_type): Remove slot_sub_symbols field.
1212         * config/tc-xtensa.c (md_apply_fix): Check for unexpected uses of
1213         subtracted symbols.
1214         (relaxation_requirements): Add pfinish_frag argument and use it to
1215         replace setting tinsn->record_fix fields.
1216         (xg_assemble_vliw_tokens): Adjust calls to relaxation_requirements
1217         and vinsn_to_insnbuf.  Remove references to record_fix and
1218         slot_sub_symbols fields.
1219         (xtensa_mark_narrow_branches): Delete unused code.
1220         (is_narrow_branch_guaranteed_in_range): Handle expr that is not just
1221         a symbol.
1222         (convert_frag_immed): Adjust vinsn_to_insnbuf call and do not set
1223         record_fix fields.
1224         (tinsn_immed_from_frag): Remove code for handling slot_sub_symbols.
1225         (vinsn_to_insnbuf): Change use of record_fixup argument, replacing use
1226         of the record_fix field.  Simplify error messages for unexpected
1227         symbolic operands.
1228         (set_expr_symbol_offset_diff): Delete.
1229
1230 2006-01-31  Paul Brook  <paul@codesourcery.com>
1231
1232         * config/tc-arm.c (arm_reg_parse): Check if reg is non-NULL.
1233
1234 2006-01-31  Paul Brook  <paul@codesourcery.com>
1235         Richard Earnshaw <rearnsha@arm.com>
1236
1237         * config/tc-arm.c: Use arm_feature_set.
1238         (arm_ext_*, arm_arch_full, arm_arch_t2, arm_arch_none,
1239         arm_cext_iwmmxt, arm_cext_xscale, arm_cext_maverick, fpu_fpa_ext_v1,
1240         fpu_fpa_ext_v2, fpu_vfp_ext_v1xd, fpu_vfp_ext_v1, fpu_vfp_ext_v2):
1241         New variables.
1242         (insns): Use them.
1243         (md_atof, opcode_select, opcode_select, md_assemble, md_assemble,
1244         md_begin, arm_parse_extension, arm_parse_cpu, arm_parse_arch,
1245         arm_parse_fpu, arm_parse_float_abi, aeabi_set_public_attributes,
1246         s_arm_cpu, s_arm_arch, s_arm_fpu): Use macros for accessing CPU
1247         feature flags.
1248         (arm_legacy_option_table, arm_option_cpu_value_table): New types.
1249         (arm_opts): Move old cpu/arch options from here...
1250         (arm_legacy_opts): ... to here.
1251         (md_parse_option): Search arm_legacy_opts.
1252         (arm_cpus, arm_archs, arm_extensions, arm_fpus)
1253         (arm_float_abis, arm_eabis): Make const.
1254
1255 2006-01-25  Bob Wilson  <bob.wilson@acm.org>
1256
1257         * config/tc-xtensa.c (md_apply_fix): Set value to zero for PLT relocs.
1258
1259 2006-01-21  Jie Zhang  <jie.zhang@analog.com>
1260
1261         * config/bfin-parse.y (asm_1): Check value range for 16 bit immediate
1262         in load immediate intruction.
1263
1264 2006-01-21  Jie Zhang  <jie.zhang@analog.com>
1265
1266         * config/bfin-parse.y (value_match): Use correct conversion
1267         specifications in template string for __FILE__ and __LINE__.
1268         (binary): Ditto.
1269         (unary): Ditto.
1270
1271 2006-01-18  Alexandre Oliva  <aoliva@redhat.com>
1272
1273         Introduce TLS descriptors for i386 and x86_64.
1274         * config/tc-i386.c (tc_i386_fix_adjustable): Handle
1275         BFD_RELOC_386_TLS_GOTDESC, BFD_RELOC_386_TLS_DESC_CALL,
1276         BFD_RELOC_X86_64_GOTPC32_TLSDESC, BFD_RELOC_X86_64_TLSDESC_CALL.
1277         (optimize_disp): Emit fix up for BFD_RELOC_386_TLS_DESC_CALL and
1278         BFD_RELOC_X86_64_TLSDESC_CALL immediately, and clear the
1279         displacement bits.
1280         (build_modrm_byte): Set up zero modrm for TLS desc calls.
1281         (lex_got): Handle @tlsdesc and @tlscall.
1282         (md_apply_fix, tc_gen_reloc): Handle the new relocations.
1283
1284 2006-01-11  Nick Clifton  <nickc@redhat.com>
1285
1286         Fixes for building on 64-bit hosts:
1287         * config/tc-avr.c (mod_index): New union to allow conversion
1288         between pointers and integers.
1289         (md_begin, avr_ldi_expression): Use it.
1290         * config/tc-i370.c (md_assemble): Add cast for argument to print
1291         statement.
1292         * config/tc-tic54x.c (subsym_substitute): Likewise.
1293         * config/tc-mn10200.c (md_assemble): Use a union to convert the
1294         opindex field of fr_cgen structure into a pointer so that it can
1295         be stored in a frag.
1296         * config/tc-mn10300.c (md_assemble): Likewise.
1297         * config/tc-frv.c (frv_debug_tomcat): Use %p to print pointer
1298         types.
1299         * config/tc-v850.c: Replace uses of (int) casts with correct
1300         types.
1301
1302 2006-01-09  H.J. Lu  <hongjiu.lu@intel.com>
1303
1304         PR gas/2117
1305         * symbols.c (snapshot_symbol): Don't change a defined symbol.
1306
1307 2006-01-03  Hans-Peter Nilsson  <hp@bitrange.com>
1308
1309         PR gas/2101
1310         * config/tc-mmix.c (mmix_handle_mmixal): Don't treat #[0-9][FB] as
1311         a local-label reference.
1312
1313 For older changes see ChangeLog-2005
1314 \f
1315 Local Variables:
1316 mode: change-log
1317 left-margin: 8
1318 fill-column: 74
1319 version-control: never
1320 End: