gas/ChangeLog

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