gas/ChangeLog

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