gas/ChangeLog

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