gas/ChangeLog

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