gas/ChangeLog

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