gas/ChangeLog

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