gas/ChangeLog

   1 2005-02-11  H.J. Lu  <hongjiu.lu@intel.com>
   2
   3         * NEWS: Mention "-munwind-check=[warning|error]".
   4
   5         * config/tc-ia64.c (md): Add unwind_check.
   6         (unwind_diagnostic): New.
   7         (in_procedure): Call unwind_diagnostic when a directive isn't
   8         in procedure.
   9         (in_prologue): Call unwind_diagnostic when a directive isn't in
  10         prologue.
  11         (in_body): Call unwind_diagnostic when a directive isn't in
  12         body region.
  13         (dot_endp): Set md.unwind_check to error before calling
  14         in_procedure and restore it after. When the name is missing or
  15         couldn't be found, use the one from the last .proc if
  16         md.unwind_check isn't error. Warn if md.unwind_check is
  17         warning.
  18         (md_parse_option): Handle "-munwind-check=[warning|error]".
  19         (md_show_usage): Add "-munwind-check=[warning|error]".
  20         (ia64_init): Set md.unwind_check to warning.
  21
  22         * doc/as.texinfo: Add "-munwind-check=[none|warning|error]".
  23         * doc/c-ia64.texi: Likewise.
  24
  25 2005-02-11  Jan Beulich  <jbeulich@novell.com>
  26
  27         * config/tc-ia64.h (LEX_AT): Include LEX_BEGIN_NAME.
  28         (LEX_QM): Likewise.
  29         (ia64_parse_name): New third parameter.
  30         (md_parse_name): Pass third argument.
  31         * config/tc-ia64.c (pseudo_func): Placeholders use NULL as name.
  32         (md_operand): Handling of '@'-prefixed symbols moved from here...
  33         (ia64_parse_name): ...to here.
  34
  35 2005-02-11  Jan Beulich  <jbeulich@novell.com>
  36
  37         * config/tc-ia64.c (md): Remove last_groups and group_idx.
  38         (errata_nop_necessary_p): Remove declaraction and definition.
  39         (emit_one_bundle): Don't call errata_nop_necessary_p. Don't
  40         update md.group_idx. Don't reset md.last_groups.
  41
  42 2005-02-11  Jan Beulich  <jbeulich@novell.com>
  43
  44         * config/tc-ia64.c (parse_section_name): Handle non-quoted first
  45         argument.
  46         (dot_xdata): Free section name after use.
  47         (dot_float_cons): Likewise.
  48         (dot_xstringer): Likewise.
  49         (dot_xdata_ua): Likewise.
  50         (dot_float_cons_ua): Likewise.
  51         (md_pseudo_table): Add xdata16 and xdata16.ua.
  52
  53 2005-02-10  H.J. Lu  <hongjiu.lu@intel.com>
  54
  55         * doc/all.texi: Add IA64.
  56         * doc/as.texinfo: Likewise.
  57
  58         * doc/c-ia64.texi: Fix typos.
  59
  60 2005-02-10  Julian Brown  <julian@codesourcery.com>
  61
  62         * config/tc-arm.c (do_t_ldmstm): Change BFD_RELOC_NONE to
  63         BFD_RELOC_UNUSED.
  64         (do_t_push_pop): Likewise.
  65         (md_assemble): Likewise.
  66         (md_apply_fix3): Handle BFD_RELOC_NONE correctly, make
  67         BFD_RELOC_UNUSED same as previous meaning of BFD_RELOC_NONE.
  68         (create_unwind_entry): Output dependency on the required personality
  69         routines.
  70         testsuite/gas/arm/unwind.d: Alter expected output to include
  71         dependency on __aeabi_unwind_cpp_pr[01].
  72
  73 2005-02-07  Nathan Sidwell  <nathan@codesourcery.com>
  74
  75         * as.h (seg_not_empty_p): Return int, not bfd_boolean.
  76         * subsegs.c (seg_not_empty_p): Likewise.
  77
  78 2005-02-07  Inderpreet Singh <inderpreetb@noida.hcltech.com>
  79
  80         * config/tc-maxq.c (md_estimate_size_before_relax): Correct the
  81         relative jump calculation.
  82         <md_convert_frag) : Likewise.
  83         <output_disp): Likewise.
  84
  85 2005-02-07  Hans-Peter Nilsson  <hp@axis.com>
  86
  87         * write.c (write_object_file): Recognize warning-symbol construct
  88         and skip object- and target- handling for the second symbol.
  89
  90 2005-02-02  Jan Beulich  <jbeulich@novell.com>
  91
  92         * config/tc-ia64.c (dot_pred_rel): Update comment. Handle @-prefixed
  93         designators along with quoted ones. Free copy of quoted designator
  94         when done.
  95
  96 2005-02-01  Ben Elliston  <bje@au.ibm.com>
  97
  98         * config/atof-ieee.c, config/obj-coff.c, config/obj-elf.c,
  99         config/obj-ieee.c, config/obj-som.c, config/obj-vms.c,
 100         config/tc-a29k.c, config/tc-alpha.c, config/tc-arc.c,
 101         config/tc-arm.c, config/tc-d30v.c, config/tc-dlx.c,
 102         config/tc-fr30.c, config/tc-h8300.c, config/tc-h8500.c,
 103         config/tc-i370.c, config/tc-i386.c, config/tc-i960.c,
 104         config/tc-ia64.c, config/tc-m32r.c, config/tc-m32r.h,
 105         config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-mips.c,
 106         config/tc-mn10200.c, config/tc-msp430.c, config/tc-ns32k.c,
 107         config/tc-openrisc.c, config/tc-or32.c, config/tc-pdp11.c,
 108         config/tc-pj.c, config/tc-sparc.h, config/tc-tic54x.c,
 109         config/tc-tic80.c, config/tc-v850.c, config/tc-w65.c,
 110         config/tc-xtensa.c, config/tc-z8k.c, config/xtensa-relax.c: Remove
 111         #if 0'd code throughout.
 112
 113 2005-01-31  Nick Clifton  <nickc@redhat.com>
 114
 115         * as.c (parse_args): Bump copyright date reported by --version to
 116         2005.
 117
 118 2005-01-31  Nick Clifton  <nickc@redhat.com>
 119
 120         * configure.tgt: Remove obsolete targets m68k-lynxos, sparc-lynxos
 121         and vax-vms.
 122
 123 2005-01-31  Jan Beulich  <jbeulich@novell.com>
 124
 125         * macro.c (buffer_and_nest): Allow 'from' being NULL; handle anything
 126         that can end with .endr in that case. Make requiring/permitting
 127         pseudo-ops without leading dot closer to the logic in read.c serving
 128         the same purpose.
 129         (expand_irp): Don't pass a mnemonic to buffer_and_nest as it will be
 130         ignored.
 131
 132 2005-01-31  Jan Beulich  <jbeulich@novell.com>
 133
 134         * macro.c (do_formals): Adjust to no longer accept empty parameter
 135         names.
 136         (define_macro): Adjust to no longer accept empty macro name, garbage
 137         following the parameters, or macros that were previously defined.
 138         * read.c (s_bad_end): Declare.
 139         (potable): Add endm. Handler for endr and endm is s_bad_end.
 140         (s_bad_end): Rename from s_bad_endr. Adjust to handle both .endm
 141         and .endr.
 142         * read.h (s_bad_endr): Remove.
 143
 144 2005-01-31  Jan Beulich  <jbeulich@novell.com>
 145
 146         * config/tc-ia64.c (parse_operands): Parse all specified operands,
 147         immediately discarding (but counting) those exceeding the maximum
 148         possible amount. Track whether output and input operand counts ever
 149         matched, and use this to better indicate which of the operands/
 150         operand types was wrong; specifically don't default to pointing to
 151         the first operand.
 152
 153 2005-01-31  Jan Beulich  <jbeulich@novell.com>
 154
 155         * config/tc-ia64.c (unwind): Remove proc_end (now an automatic
 156         variable in dot_endp). Add body and insn. Make prologue,
 157         prologue_mask, body, and insn bitfields.
 158         (fixup_unw_records): Remove spurious new-lines from end of diagnostic
 159         messages.
 160         (in_procedure, in_prologue, in_body): New.
 161         (dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
 162         dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
 163         dot_unwentry, dot_altrp, dot_savemem, dot_saveg, dot_savef, dot_saveb,
 164         dot_savegf, dot_spill, dot_spillreg, dot_spillmem, dot_spillreg_p,
 165         dot_spillmem_p, dot_label_state, dot_copy_state, dot_unwabi,
 166         dot_personality): Use the appropriate one of the above.
 167         (dot_proc): Clear unwind.proc_start; set to current location only if
 168         none of the entry points were valid. Check for non-zero-length entry
 169         point names. Check that entry points aren't defined, yet. Clear
 170         unwind.prologue, unwind.body, and unwind.insn.
 171         (dot_body): Call in_procedure. Check that first directive in procedure
 172         had no insns emitted before. Set unwind.body.
 173         (dot_prologue): Call in_procedure. Check that not already in prologue.
 174         Check that first directive in procedure had no insns emitted before.
 175         Clear unwind.body.
 176         (dot_endp): Call in_procedure. Declare proc_end. Check for non-zero-
 177         length entry point names. Check that entry points became defined.
 178         (md_assemble): Set unwind.insn once unwind.proc_start is defined.
 179
 180 2005-01-31  Jan Beulich  <jbeulich@novell.com>
 181
 182         * config/tc-ia64.c (emit_one_bundle): Snapshot manual bundling state
 183         before actually using it. Don't generate an error in manual bundling
 184         mode when looking at an insn requiring slot 2 but not yet at slot 2.
 185         Don't generate an error in manual bundling mode when looking at an
 186         insn required to be last in its group but the required slot hasn't
 187         been reached, yet. Allow conversion from MII to MI;I for bundle
 188         consisting of only 2 insns with the stop between them. Suppress
 189         various meaningless errors resulting from detecting earlier ones.
 190
 191 2005-01-31  Jan Beulich  <jbeulich@novell.com>
 192
 193         * config/tc-ia64.c (parse_operands): Also handle alloc without first
 194         input being ar.pfs.
 195
 196 2005-01-28  Christian Groessler  <chris@groessler.org>
 197
 198         * config/tc-z8k.c (md_assemble): Improve error detection.
 199
 200 2005-01-28  Jan Beulich  <jbeulich@novell.com>
 201
 202         * config/tc-ia64.c (ia64_estimate_size_before_relax): Allocate space
 203         for personality routine pointer only if there is one.
 204         (ia64_convert_frag): Likewise.
 205         (generate_unwind_image): Likewise.
 206
 207 2005-01-27  Christian Groessler  <chris@groessler.org>
 208
 209         * config/tc-z8k.c (INSERT): Remove, not used anywhere.
 210         (md_apply_fix3): Make relative branches out of range an error
 211         instead of a warning.  Display correct line number for out of
 212         range branches/calls/memory accesses.
 213
 214 2005-01-27  Nathan Sidwell  <nathan@codesourcery.com>
 215
 216         * dwarf2dbg.c (dwarf2_finish): Correct logic for determining when
 217         to emit .debug_line and other debug sections.
 218         * as.h (seg_not_empty_p): Declare.
 219         * subsegs.c (seg_not_empty_p): New predicate.
 220
 221 2005-01-27  Andrew Cagney  <cagney@gnu.org>
 222
 223         * configure: Regenerate to track ../gettext.m4 change.
 224
 225 2005-01-27  Jan Beulich  <jbeulich@novell.com>
 226
 227         * config/tc-ia64.c (emit_one_bundle): Change "?imbf??" to "?ibmfxx".
 228
 229 2005-01-27  Jan Beulich  <jbeulich@novell.com>
 230
 231         * config/tc-ia64.c (emit_one_bundle): Add late resolution of move
 232         to/from application registers dynamic insns.
 233         (md_assemble): Defer resolution of move to/from application registers
 234         dynamic insns when they can be issued on either the I- or M-units.
 235
 236 2005-01-25  Alexandre Oliva  <aoliva@redhat.com>
 237
 238         * config/tc-frv.c (md_apply_fix3): Mark TLS symbols as such.
 239         2004-12-10  Alexandre Oliva  <aoliva@redhat.com>
 240         * config/tc-frv.c (frv_pic_ptr): Add tlsmoff support.
 241         2004-11-10  Alexandre Oliva  <aoliva@redhat.com>
 242         * cgen.c (gas_cgen_parse_operand): Handle
 243         CGEN_PARSE_OPERAND_SYMBOLIC.
 244         * config/tc-frv.c (md_cgen_lookup_reloc): Handle TLS relocations.
 245         (frv_force_relocation): Likewise.  Fix handling of PIC
 246         relocations.
 247         (md_apply_fix3): Likewise.
 248
 249 2005-01-21  Ben Elliston  <bje@au.ibm.com>
 250
 251         * as.h: Remove #if 0'd code.
 252         * atof-generic.c (atof_generic): Likewise.
 253         * ecoff.c (ecoff_directive_frame): Likewise.
 254         * frags.h (FRAG_APPEND_1_CHAR): Likewise.
 255         * itbl-ops.c (itbl_add_reg): Likewise.
 256         * listing.c (calc_hex): Likewise.
 257         * read.c (MASK_CHAR): Likewise.
 258         * subsegs.c (subsegs_print_statistics): Likewise.
 259         * symbols.c (indent): Likewise.
 260         * write.c (write_relocs): Likewise.
 261         (write_object_file): Likewise.
 262         (relax_frag): Likewise.
 263
 264 2005-01-20  Nick Clifton  <nickc@redhat.com>
 265
 266         * as.c (std_longopts): Add an entry for "--a" in order to prevent
 267         getopt_long_only() from considering -a as an abbreviation for
 268         --alternate.
 269         (parse_args): Fix the parsing of -a=<file>.
 270
 271 2005-01-20  Alan Modra  <amodra@bigpond.net.au>
 272
 273         PR gas/684
 274         * read.c (s_incbin): Adjust default count for skip.  Check validity
 275         of count and skip rigorously.
 276
 277 2005-01-19  Fred Fish  <fnf@specifixinc.com>
 278
 279         * config/tc-mips.c (dummy_opcode): Add init for new struct member.
 280
 281 2005-01-19  Richard Sandiford  <rsandifo@redhat.com>
 282
 283         * read.c (convert_to_bignum): New function, split out from...
 284         (emit_expr): ...here.  Handle the case where X_add_number is
 285         positive and the input value is negative.
 286         (output_big_sleb128): Fix setting of continuation bit.  Check whether
 287         the final byte needs to be sign-extended.  Fix size-shrinking loop.
 288         (emit_leb128_expr): When generating a signed leb128, see whether the
 289         sign of an O_constant's X_add_number matches the sign of the input
 290         value.  Use a bignum if not.
 291
 292 2005-01-17  Andrew Stubbs  <andrew.stubbs@st.com>
 293
 294         * tc-sh.c (md_begin,md_parse_option): Change arch_sh1_up to
 295         arch_sh_up in order to match the external name and make the
 296         testsuite's job easier.
 297
 298 2005-01-14  H.J. Lu  <hongjiu.lu@intel.com>
 299
 300         PR 659
 301         * config/tc-i386.c (i386_scale): Disallow 0 scale.
 302
 303 2005-01-12  Nick Clifton  <nickc@redhat.com>
 304
 305         * config/tc-iq2000.c (s_iq2000_set): Fix thinko parsing
 306         ignored_arguments array.
 307
 308 2005-01-10  H.J. Lu  <hongjiu.lu@intel.com>
 309
 310         * write.c (write_object_file): Disallow a symbol equated to
 311         common symbol.
 312
 313 2005-01-10  Inderpreet Singh <inderpreetb@noida.hcltech.com>
 314
 315         * tc-maxq.c: Replace constants 10 and 20 with bfd_mach_maxq10 and
 316         bfd_mach_maxq20.
 317         (md_pseudo_table): Add new pseudo ops for maxq10 and maxq20.
 318         (maxq_target): New function: Set the machine type.
 319
 320 2005-01-06  Paul Brook  <paul@codesourcery.com>
 321
 322         * config/tc-arm.c (FPU_DEFAULT): Define for TE_VXWORKS.
 323         (md_begin): Handle TE_VXWORKS for FP defaults.
 324         (md_apply_fix3): Correct rela offsets.
 325         (elf32_arm_target_format): Add VxWorks targets.
 326
 327 2005-01-06  Paul Brook  <paul@codesourcery.com>
 328
 329         * configure.tgt: Set em=vxworks for *-*-vxworks.
 330         * config/te-vxworks.h: New File.
 331
 332 2005-01-06  Paul Brook  <paul@codesourcery.com>
 333
 334         * config/tc-arm.c (arm_cpus): Correct arch field for arm1026ej-s.
 335
 336 2005-01-04  Dmitry Diky  <diwil@spec.ru>
 337
 338         * config/tc-msp430.c (md_apply_fix3): Fix offset calculation for
 339         global label.
 340
 341 2005-01-03  David Mosberger  <davidm@hpl.hp.com>
 342
 343         * config/tc-ia64.c (md): Add member "loc_directive_seen".
 344         (dot_loc): New function.
 345         (md_pseudo_table): Add entry to map .loc to dot_loc().
 346         (emit_one_bundle): Only call dwarf2_gen_line_info() if we have
 347         seen a .loc directive or we're generating DWARF2 debug info for
 348         assembly source.
 349
 350 2004-12-29  Alan Modra  <amodra@bigpond.net.au>
 351
 352         PR gas/619
 353         * read.c (s_comm_internal): Don't zero end of name until size
 354         expression has been parsed.
 355
 356 2004-12-25  Marek Michalkiewicz  <marekm@amelek.gda.pl>
 357
 358         * config/tc-avr.c (mcu_types): Move attiny{13,2313} from avr4 to avr2.
 359
 360 2004-12-23  Tomer Levi  <Tomer.Levi@nsc.com>
 361
 362         * config/tc-crx.c: Support 'bcop' relaxation (dealt as in 'cmp&branch'
 363         case).
 364
 365 2004-12-22  Ian Lance Taylor  <ian@airs.com>
 366
 367         * configure.tgt: New.
 368         * configure.in: Move setting of cpu_type, fmt, etc., to
 369         configure.tgt.
 370         * Makefile.am (CONFIG_STATUS_DEPENDENCIES): Add
 371         $(srcdir)/configure.tgt.
 372         * configure, Makefile.in: Rebuild.
 373
 374 2004-12-22  Klaus Rudolph  <lts-rudolph@gmx.de>
 375
 376         * config/tc-avr.c: Add support for the new R_AVR_LDI, R_AVR_6 and
 377         R_AVR_6_ADIW relocs for the LDI, ADIW/SBIW and LDD/STD
 378         instructions.
 379         (avr_offset_expression): New function to parse offsets for LDI
 380         instructions.
 381         (avr_operand): Use it.
 382         (md_apply_fix3): Generate the relocs.
 383
 384 2004-12-16  Andrew Stubbs  <andrew.stubbs@st.com>
 385
 386         * config/tc-sh64.c (shmedia_md_apply_fix3): Add missing
 387         BFD_RELOC_SH_IMMS10BY8 relocation.
 388
 389         * config/tc-sh64.c (shmedia_build_Mytes): Emit an error message rather
 390         than just ignoring bad code.
 391
 392 2004-12-16  Richard Sandiford  <rsandifo@redhat.com>
 393
 394         * config/tc-v850.c (handle_lo16): New function.
 395         (v850_reloc_prefix): Use it to check lo().
 396         (md_assemble, md_apply_fix3): Handle BFD_RELOC_V850_LO16_SPLIT_OFFSET.
 397
 398 2004-12-14  P.J. Darcy  <darcypj@us.ibm.com>
 399
 400         * configure.in: Add s390x-ibm-tpf support.
 401         * configure: Regenerate.
 402
 403 2004-12-15 Jan Beulich  <jbeulich@novell.com>
 404
 405         * config/obj-elf.c (obj_elf_change_section): Only set type and
 406         attributes on new sections. Emit warning when type of re-declared
 407         section doesn't match.
 408
 409 2004-12-15 Jan Beulich  <jbeulich@novell.com>
 410
 411         * dw2gencfi.c (dot.cfi.startproc): Clear cur_cfa_offset so
 412         '.cfi_startproc simple' doesn't inherit the old value.
 413
 414 2004-12-15  Jan Beulich  <jbeulich@novell.com>
 415
 416         * dw2gencfi.c (output_cfi_insn): Adjust DW_CFA_def_cfa_sf generation
 417         to emit a signed and factored offset. Adjust DW_CFA_def_cfa_offset_sf
 418         generation to emit a factored offset.
 419
 420 2004-12-10  Ian Lance Taylor  <ian@wasabisystems.com>
 421
 422         * config/tc-mips.c (macro) [M_LA_AB]: Give an error for a offset
 423         which is too large in the case of NO_PIC without 64-bit
 424         addresses.
 425
 426         * config/tc-mips.c (mips_in_shared): New static variable.
 427         (macro_build_lui): Permit "_gp" if !mips_in_shared.
 428         (md_longopts): Add -mshared and -mno-shared.
 429         (md_parse_option): Handle OPTION_MSHARED and OPTION_MNO_SHARED.
 430         (s_cpload): Implement !mips_in_shared case.
 431         (s_cpsetup): Likewise.
 432         * doc/c-mips.texi (MIPS Opts): Document -mno-shared.
 433         * NEWS: Mention -mno-shared.
 434
 435 2004-12-09  Paul Brook  <paul@codesourcery.com>
 436
 437         * config/tc-arm.c (s_arm_unwind_fnend): Use R_ARM_PREL31 relocation
 438         for function start.
 439
 440 2004-12-09  Ian Lance Taylor  <ian@wasabisystems.com>
 441
 442         * config/tc-mips.c (append_insn): If we emit a nop during a relax
 443         sequence, increase the size of the sequence.
 444
 445         * config/tc-mips.c (mips_cpu_info_table): Change "9000" entry to
 446         use CPU_RM9000.
 447
 448 2004-12-07  Ben Elliston  <bje@gnu.org>
 449
 450         * read.c (s_align): Use an align_limit temporary to allay a GCC
 451         signed/unsigned comparison warning.
 452
 453 2004-12-01  Mark Mitchell  <mark@codesourcery.com>
 454
 455         * Makefile.am (TARG_ENV_HFILES): Add te-armlinuxeabi.h.
 456         * configure.in: Use it for arm*-*-linux-gnueabi*.
 457         * config/tc-arm.c: Allow emulation file to set FPU_DEFAULT.
 458         * config/te-armlinuxeabi.h: New file.
 459         * Makefile.in: Regenerated.
 460         * aclocal.m4: Likewise.
 461         * configure: Likewise.
 462         * doc/Makefile.in: Regenerated.
 463
 464 2004-12-02  Bob Wilson  <bob.wilson@acm.org>
 465
 466         * config/tc-xtensa.c (xtensa_switch_section_emit_state): Use subseg_set.
 467         (xtensa_restore_emit_state): Likewise.
 468
 469 2004-12-02  Alan Modra  <amodra@bigpond.net.au>
 470
 471         * read.c (ALIGN_LIMIT): Define, increasing limit for BFD_ASSEMBLER.
 472         (s_align): Use it.
 473
 474 2004-11-30  Tero Niemela  <tero_niemela@yahoo.com>
 475
 476         * Makefile.am: Change LOCALEDIR to $(datadir)/share.
 477         * Makefile.in: Regenerate.
 478
 479 2004-11-29  Tomer Levi  <Tomer.Levi@nsc.com>
 480
 481         * config/tc-crx.c: Major code cleanup. Remove unused variables and
 482         functions, give functions a meaningful name, add comments.
 483         (check_range): New function - Replace operand size calculation
 484         with range checking.
 485         (assemble_insn): Update Algorithm, improve error issuing.
 486         (enum op_err): New.
 487         (process_label_constant): Bug fix regarding COP_BRANCH_INS relocation
 488         handling.
 489
 490 2004-11-29  Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
 491
 492         * config/tc-m32r.c (md_pcrel_from_section): Fixed a pcrel relocte
 493         miss between different section in the same module.
 494         (tc_gen_reloc): Likewise.
 495
 496 2004-11-25  Theodore A. Roth  <troth@openavr.org>
 497
 498         * gas/config/tc-avr.c (mcu_types): Add support for atmega165,
 499         atmega325, atmega3250, atmega645 and atmega6450.
 500
 501 2004-11-25 Jan Beulich  <jbeulich@novell.com>
 502
 503         * config/tc-i386.c (optimize_imm): Adjust immediates to only those
 504         permissible for the selected instruction suffix.
 505         (process_suffix): For DefaultSize instructions, suppressing the
 506         guessing of a 'q' suffix if the instruction doesn't support it is
 507         pointless, because only an 'l' suffix can be guessed in this place.
 508
 509 2004-11-24  Nick Clifton  <nickc@redhat.com>
 510
 511         * config/tc-iq2000.c: Remove support for IQ10 processor.
 512         Convert to ISO C90 formatting.
 513         * config/tc-iq2000.h: Likewise.
 514
 515 2004-11-23  Nick Clifton  <nickc@redhat.com>
 516
 517         * config/tc-mn10300.c (md_relax_table): More fixes to the offsets
 518         in this table.  They should be correct now.
 519
 520 2004-11-23 Jan Beulich <jbeulich@novell.com>
 521
 522         * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
 523         indicate the MMX extensions added by both SSE and 3DNow!A.
 524         (Cpu3dnowA): Declare.
 525         (CpuUnknownFlags): Update.
 526         * config/tc-i386.c (cpu_sub_arch_name): Declare.
 527         (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
 528         neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
 529         3DNow!. Athlon additionally implies 3DNow!A. Several new
 530         entries (those starting with a dot are for sub-arch specification).
 531         (set_cpu_arch): Handle sub-arch specifications.
 532         (parse_insn): Distinguish between instructions not supported because
 533         of insufficient CPU features and because of 64-bit mode.
 534         * doc/c-i386.texi: Describe enhanced .arch directive.
 535
 536 2004-11-22  Bob Wilson  <bob.wilson@acm.org>
 537
 538         * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
 539
 540 2004-11-22  Bob Wilson  <bob.wilson@acm.org>
 541
 542         * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
 543         without a corresponding .debug_info section.
 544
 545 2004-11-22  Hans-Peter Nilsson  <hp@axis.com>
 546
 547         * read.c (potable): Add "error" and "warning".
 548         (s_errwarn): New function.
 549         * read.h (s_errwarn): Declare.
 550         * doc/as.texinfo (Error, Warning): Document .error and .warning.
 551
 552 2004-11-22  Nick Clifton  <nickc@redhat.com>
 553
 554         * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
 555         c_dot_file_symbol.
 556
 557 2004-11-19  Alan Modra  <amodra@bigpond.net.au>
 558
 559         * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
 560         struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
 561
 562 2004-11-19  Alan Modra  <amodra@bigpond.net.au>
 563
 564         * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
 565         (coff_adjust_symtab): Adjust call.
 566         (crawl_symbols): Likewise.
 567         * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
 568         (obj_app_file): Adjust.
 569
 570 2004-11-18  Inderpreet Singh   <inderpreetb@nioda.hcltech.com>
 571
 572         * configure.in: Enable bfd_assember for the MAXQ port.
 573         * configure: Regenerate.
 574
 575 2004-11-12  Bob Wilson  <bob.wilson@acm.org>
 576             Sterling Augustine  <sterling@tensilica.com>
 577
 578         * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
 579         there is a conflict.
 580         (check_t1_t2_reads_and_writes): Check for both reads and writes to
 581         interfaces that are related as determined by xtensa_interface_class_id.
 582
 583 2004-11-12  Nick Clifton  <nickc@redhat.com>
 584
 585         * config/tc-mn10300.c (md_relax_table): Fix off by one negative
 586         offsets for conditional branches.
 587
 588 2004-11-11  Bob Wilson  <bob.wilson@acm.org>
 589
 590         * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
 591
 592 2004-11-10  Bob Wilson  <bob.wilson@acm.org>
 593
 594         * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
 595         relaxing at the end of a loop.  Don't mark frags as UNREACHABLE or
 596         MAYBE_UNREACHABLE.
 597         (relax_frag_immed): Update call to update_next_frag_state.
 598
 599 2004-11-10  Alan Modra  <amodra@bigpond.net.au>
 600
 601         * obj.h (struct format_ops <app_file>): Add int param.
 602         * read.h (s_app_file_string): Likewise.
 603         * read.c (s_app_file_string): Likewise.
 604         (s_app_file): Adjust s_app_file_string call.
 605         * config/tc-mips.c (s_mips_file): Likewise.
 606         * config/obj-coff.h (obj_app_file): Add app param.
 607         * config/obj-ecoff.h (obj_app_file): Likewise.
 608         * config/obj-multi.h (obj_app_file): Likewise.
 609         * config/obj-elf.h (elf_file_symbol): Likewise.
 610         * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
 611         if called for # preprocessor lines.
 612
 613 2004-11-08  H.J. Lu  <hongjiu.lu@intel.com>
 614
 615         PR 528
 616         * symbols.c (resolve_symbol_value): Convert weak symbols only
 617         for Windows PECOFF.
 618         (symbol_equated_reloc_p): Don't equate weaks when relocating
 619         only for Windows PECOFF.
 620
 621 2004-11-08  Tomer Levi  <Tomer.Levi@nsc.com>
 622
 623         * config/tc-crx.c (print_insn): Check and set insn_addr.
 624         * config/tc-crx.h (md_frag_check): Define.
 625
 626 2004-11-08  Inderpreet Singh   <inderpreetb@nioda.hcltech.com>
 627             Vineet Sharma      <vineets@noida.hcltech.com>
 628
 629         * configure.in: Add support for new target maxq-coff.
 630         * configure: Regenerate.
 631         * NEWS: Mention new support.
 632         * config/tc-maxq.c: New file.
 633         * config/tc-maxq.h: New file.
 634         * config/obj-coff.h: Add support for maxq-coff.
 635
 636 2004-11-08  Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
 637
 638         * symbols.c (any_external_name): Define.
 639         (resolve_symbol_value): Convert weak symbols.
 640         (S_SET_EXTERNAL): Support any_external_name.
 641         (S_SET_NAME): Qualify parameter const.
 642         (symbol_equated_reloc_p): Equate to weaks when relocating.
 643         * symbols.h (S_SET_NAME): Qualfiy parameter const.
 644         * tc.h (any_external_name): Declare.
 645         * config/obj-coff.c ("coff/pe.h"): Include for BFD
 646         assemblers also.
 647         (weak_is_altname): Declare and define.
 648         (weak_name2altname): Same.
 649         (weak_altname2name): Same.
 650         (weak_uniquify): Same.
 651         (weak_altprefix): Define.
 652         (obj_coff_weak): Change .weak syntax and handling.
 653         (coff_frob_symbol): Fix PE weak symbol alternates.
 654         * config/obj-coff.h (USE_UNIQUE): Define.
 655         * config/tc-i386.c (md_apply_fix3): Assume weak symbols
 656         are in another segment.
 657         (tc_gen_reloc): Remove broken addend hack.
 658         doc/as.texinfo: Update.
 659
 660 2004-11-05  Sterling Augustine  <sterling@tensilica.com>
 661
 662         * config/tc-xtensa.c (total_frag_text_expansion): New.
 663         (md_estimate_size_before_relax): Use it.
 664         (find_address_of_next_align_frag): Likewise.
 665
 666 2004-11-05  Tomer Levi  <Tomer.Levi@nsc.com>
 667
 668         * config/tc-crx.c: Rename argument types.
 669         (processing_arg_number): Rename to 'cur_arg_num'.
 670         (get_number_of_bits): Rename to 'set_operand_size'.
 671         (get_operandtype): Rename to 'parse_operand', totally rewrite.
 672         (set_cons_rparams): Rename to 'set_operand', totally rewrite.
 673         (set_indexmode_parameters): Remove function, integrate its code into
 674         'set_operand'.
 675         (set_operand_size): Get rid of 'Operand Number' function parameter -
 676         use global variable 'cur_arg_num' instead.
 677         Use a local 'argument' pointer to reference the current argument.
 678         (parse_operand): Likewise.
 679         (set_operand): Likewise.
 680         (process_label_constant): Likewise.
 681
 682 2004-11-04  Bob Wilson  <bob.wilson@acm.org>
 683
 684         * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
 685         (add_section_rename): Delete.  Inlined into...
 686         (build_section_rename): ...here.  Use xstrdup instead of strdup.
 687         (xtensa_section_rename): Drop "const" from argument and return types.
 688         (md_show_usage): Indent to match show_usage().
 689         * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
 690         (tc_canonicalize_section_name): Define.
 691         (md_elf_section_rename): Remove unused macro.
 692         * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
 693         * doc/c-xtensa.texi (Xtensa Options): Likewise.
 694         (Frame Directive): Delete.
 695
 696 2004-11-04  Daniel Jacobowitz  <dan@debian.org>
 697
 698         * configure.in: Remove arm-*-oabi and thumb-*-oabi.
 699         * config/tc-arm.c (target_oabi): Delete.
 700         (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
 701         (arm_opts): Remove moabi.
 702         * doc/as.texinfo (Overview): Remove documentation of -moabi.
 703         * doc/c-arm.texi (ARM Options): Likewise.
 704         * configure: Regenerated.
 705
 706 2004-11-04  Hans-Peter Nilsson  <hp@axis.com>
 707
 708         * configure.in (crisv32): Recognize.  AC_DEFINE_UNQUOTED
 709         DEFAULT_CRIS_ARCH.  Handle crisv32-*-linux-gnu* like
 710         cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
 711         * configure: Regenerate.
 712         * config/tc-cris.c (enum cris_archs): New.
 713         (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
 714         (cris_insn_ver_valid_for_arch): New functions.
 715         (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
 716         (cris_arch): New variable.
 717         (md_pseudo_table): New pseudo .arch.
 718         (err_for_dangerous_mul_placement): Initialize according to
 719         DEFAULT_CRIS_ARCH.
 720         (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
 721         All users changed.
 722         (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
 723         (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
 724         (BRANCH_WF_V32, BRANCH_WB_V32): New.
 725         (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
 726         use in md_cris_relax_table.
 727         (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
 728         STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
 729         Update and improve head comment.
 730         (OPTION_PIC): Define in terms of previous option, OPTION_US.
 731         (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
 732         (OPTION_ARCH): New.
 733         (md_longopts): New option --march=...
 734         (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
 735         macros.
 736         (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
 737         (HANDLE_RELAXABLE): New macro.
 738         (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
 739         cases.  Check for weak symbols and assume not relaxable.  Handle
 740         STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
 741         STATE_ABS_BRANCH_V32, STATE_LAPC.  Use new variable symbolP, not
 742         fragP->fr_symbol.
 743         (md_convert_frag): Handle STATE_COND_BRANCH_V32,
 744         STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
 745         (cris_create_short_jump): Adjust for CRISv32.
 746         (cris_relax_frag): Handle new states.
 747         (md_create_long_jump): Ditto.  Emit error for common_v10_v32.
 748         (md_begin): Define symbols "..asm.arch.cris.v32",
 749         "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
 750         "..asm.arch.cris.any_v0_v10".  Use cris_insn_ver_valid_for_arch
 751         when entering opcode table entry points.
 752         (md_assemble): Adjust branch handling for CRISv32.  Handle LAPC
 753         relaxation.  In fix_new_exp call for main insn, pass 1 for pcrel
 754         parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
 755         (cris_process_instruction): Initialize out_insnp->insn_type to
 756         CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
 757         <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
 758         cases.
 759         <case 'm'>: Check that modified_char == '.'.
 760         <invalid operands>: Consume the rest of the line.
 761         When operands don't match, skip over subsequent insns with
 762         non-matching version specifier but same mnemonic.
 763         <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
 764         special registers in CRISv32 are always 32 bit long.
 765         <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
 766         New cases.
 767         (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
 768         and compatible.  Recognize "ACR" for v32, unless followed by "+".
 769         (get_spec_reg): Consider cris_arch when looking up register.
 770         (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
 771         v32 or compatible.
 772         (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
 773         (cris_get_expression): Restore input_line_pointer if failing "early".
 774         (get_flags): Consider cris_arch and recognize flags accordingly.
 775         (branch_disp): Adjust for CRISv32.
 776         (gen_cond_branch_32): Similar.  Emit error for common_v10_v32.
 777         (cris_number_to_imm): Use as_bad_where, not as_bad.  Remove
 778         related FIXME.  Don't insist on BFD_RELOC_32_PCREL fixup to be
 779         resolved.  Don't enter zeros in object file for
 780         BFD_RELOC_32_PCREL.
 781         <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
 782         <case BFD_RELOC_CRIS_SIGNED_8>: New case.
 783         (md_parse_option): Break out "return 1".
 784         <OPTION_ARCH> New case.
 785         (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
 786         <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
 787         <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
 788         <case BFD_RELOC_32_PCREL>: New cases.
 789         Addends for non-zero fx_pcrel are too in fx_offset.
 790         (md_show_usage): Show --march=<arch>.
 791         (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
 792         (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
 793         (s_syntax) <struct syntaxes>: Properly constify member operand.
 794         * config/tc-cris.h (TARGET_MACH): Define.
 795         (cris_mach): Declare.
 796         * doc/as.texinfo (Overview) <CRIS>: Add --march=...
 797         * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
 798         (CRIS-Opts): Document --march=...
 799         (CRIS-Pseudos): Document .arch.
 800
 801 2004-11-04 Jan Beulich <jbeulich@novell.com>
 802
 803         * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
 804         intel syntax and no register prefix, allow $ in symbol names when
 805         intel syntax.
 806         (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
 807         (intel_float_operand): Add fourth return value indicating math control
 808         operations. Make classification more precise.
 809         (md_assemble): Complain if memory operand of mov[sz]x has no size
 810         specified.
 811         (parse_insn): Translate word operands to floating point instructions
 812         operating on integers as well as control instructions to short ones
 813         as expected by AT&T syntax. Translate 'd' suffix to short one only for
 814         floating point instructions operating on non-integer operands.
 815         (match_template): Remove fldcw special case. Adjust q-suffix handling
 816         to permit it on fild/fistp/fisttp in AT&T mode.
 817         (process_suffix): Don't guess DefaultSize insns' suffix from
 818         stackop_size for certain floating point control instructions. Guess
 819         suffix for branch and [ls][gi]dt based on flag_code. Split error
 820         messages for Intel and AT&T syntax, and make the condition more strict
 821         for the former. Adjust suppressing of generation of operand size
 822         overrides.
 823         (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
 824         OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
 825         more error checking.
 826         * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
 827         SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
 828
 829 2004-11-03  Hans-Peter Nilsson  <hp@axis.com>
 830
 831         * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
 832         md_short_jump_size, md_long_jump_size.
 833         * write.c [!WORKING_DOT_WORD]: Ditto.
 834         * tc.h [!WORKING_DOT_WORD]: Declare them here.  Drop const
 835         qualifier.
 836         * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
 837         declare.
 838         * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
 839         const qualifier in these definitions.
 840         * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
 841         config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
 842
 843 2004-11-02  Nick Clifton  <nickc@redhat.com>
 844
 845         * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
 846         table before deciding to produce a .debug_line section to match up
 847         with a user provided .debug_info section.
 848
 849 2004-10-28  Tomer Levi  <Tomer.Levi@nsc.com>
 850
 851         * config/tc-crx.c (getreg_image): Bug fix, a return value was
 852         mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
 853
 854 2004-10-27  Tomer Levi  <Tomer.Levi@nsc.com>
 855
 856         * config/tc-crx.c: Remove global variable 'post_inc_mode'.
 857         (get_flags): New function.
 858         (get_number_of_bits): Edit comments, update numeric values to
 859         supported sizes.
 860         (process_label_constant): Don't support the colon
 861         format (SYMBOL:[s|m|l]).
 862         (set_cons_rparams): Support argument type 'arg_rbase'.
 863         (get_operandtype): Bug fix in 'rbase' operand type parsing.
 864         (handle_LoadStor): Bug fix, first handle post-increment mode.
 865         (getreg_image): Remove redundant code, update according to latest
 866         CRX spec.
 867         (print_constant): Bug fix relate to 3-word instructions.
 868         (assemble_insn): Bug fix, when matching instructions, verify also
 869         instruction type (not only mnemonic).
 870         Add various error checking.
 871         (preprocess_reglist): Support HI/LO and user registers.
 872
 873 2004-10-25  David Mosberger-Tang  <davidm@hpl.hp.com>
 874
 875         * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
 876         in an epilogue directive go negative.
 877
 878 2004-10-25  H.J. Lu  <hongjiu.lu@intel.com>
 879
 880         PR 474
 881         * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
 882         after reporting template error during manual bundling. Reported
 883         by Michael Dupont, michaelx.dupont@intel.com.
 884
 885 2004-10-25  Daniel Jacobowitz  <dan@debian.org>
 886
 887         * Makefile.am: Run dep-am.
 888         * aclocal.m4: Regenerate with automake 1.9.2.
 889         * Makefile.in: Regenerate with automake 1.9.2.
 890         * doc/Makefile.in: Likewise.
 891
 892         * config/tc-arm.c: Include "dw2gencfi.h".
 893         (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
 894         New functions.
 895         * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
 896         (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
 897         (tc_cfi_frame_initial_instructions): Define.
 898         (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
 899         Add prototypes.
 900
 901 2004-10-21  Tomer Levi  <Tomer.Levi@nsc.com>
 902
 903         * config/tc-crx.c (assemble_insn): Check unsigned immediate
 904         operands validity.
 905         Update coprocessor id to be unsigned immediate.
 906
 907 2004-10-18  Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
 908
 909         * config/tc-i386.c (O_secrel): Delete.
 910         (tc_pe_dwarf2_emit_offset): New function.
 911         * config/tc-i386.h (O_secrel): Define as O_md1.
 912         (TC_DWARF2_EMIT_OFFSET): Define.
 913
 914 2004-10-18  Nick Clifton  <nickc@redhat.com>
 915
 916         * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
 917         ignore @fptr() directives for 4-byte fixups.
 918
 919 2004-10-15  Alan Modra  <amodra@bigpond.net.au>
 920
 921         * Makefile.am: Run "make dep-am"
 922         * Makefile.in: Regenerate.
 923
 924 2004-10-14  Bob Wilson  <bob.wilson@acm.org>
 925
 926         * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
 927         Remove comments about placement of literal pools.
 928         (Literal Directive): Update description of literal placement.
 929         (Literal Prefix Directive): Remove statement that this does not apply
 930         to absolute-mode literals.  Describe new section naming scheme.
 931
 932 2004-10-12  Bob Wilson  <bob.wilson@acm.org>
 933
 934         * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
 935         (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
 936         is_windowed_return_opcode): Delete.
 937         (xtensa_frob_label): Use get_subseg_target_freq.
 938         (md_assemble): Inline call to is_entry_opcode.
 939         (xtensa_handle_align): Inline call to get_frag_is_literal.
 940         (relaxation_requirements): Inline call to is_jx_opcode.
 941         (emit_single_op): Inline call to is_movi_opcode.
 942         (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
 943         get_frag_is_no_transform, is_entry_opcode, and
 944         set_frag_is_specific_opcode.  Use get_subseg_total_freq.
 945         (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
 946         xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
 947         Inline calls to get_frag_is_no_transform.
 948         (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
 949         (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
 950         get_frag_is_no_transform.
 951         (convert_frag_immed_finish_loop): Inline calls to get_expression_value
 952         and set_frag_is_no_transform.
 953         (get_expression_value): Delete.
 954         (subseg_map struct): Rename cur_total_freq to total_freq.  Rename
 955         cur_target_freq to target_freq.
 956         (get_subseg_info): Split out code to create a new map entry into ...
 957         (add_subseg_info): ... this new function.
 958         (get_last_insn_flags): Check if get_subseg_info succeeded.
 959         (set_last_insn_flags): Call add_subseg_info if needed.
 960         (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
 961         (xtensa_reorder_segments): Compute last_sec while counting sections.
 962         Remove call to get_last_sec.
 963         (get_last_sec): Delete.
 964         (cache_literal_section): Inline call to retrieve_literal_seg and its
 965         callees, seg_present and add_seg_list.
 966         (retrieve_literal_seg, seg_present, add_seg_list): Delete.
 967         (get_frag_is_insn, get_frag_is_no_transform,
 968         set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
 969         * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
 970
 971 2004-10-12  Bob Wilson  <bob.wilson@acm.org>
 972
 973         * config/tc-xtensa.c: Use ISO C90 formatting.
 974         * config/tc-xtensa.h: Likewise.
 975         * config/xtensa-istack.h: Likewise.
 976         * config/xtensa-relax.c: Likewise.
 977         * config/xtensa-relax.h: Likewise.
 978
 979 2004-10-12  Paul Brook  <paul@codesourcery.com>
 980
 981         * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
 982         EF_ARM_EABI_VER4.
 983         (arm_eabis): Ditto.
 984         * doc/c-arm.texi: Document that we actually support -meabi=4, not
 985         -meabi=3.
 986
 987 2004-10-08  Bob Wilson  <bob.wilson@acm.org>
 988
 989         * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
 990         entries.
 991         (Acknowledgements): Use "GAS" instead of AS variable.
 992
 993 2004-10-08  Daniel Jacobowitz  <dan@debian.org>
 994
 995         * config/tc-i386.c: Include "elf/x86-64.h".
 996         (i386_elf_section_type): New function.
 997         * config/tc-i386.h (md_elf_section_type): Define.
 998         (i386_elf_section_type): New prototype.
 999
1000 2004-10-08  Linus Nielsen Feltzing  <linus@haxx.se>
1001
1002         * config/m68k-parse.h (enum m68k_register): New control register,
1003         MBAR2 (for MCF5249)
1004         * config/tc-m68k.c: Correct control register set for MCF5249.
1005
1006 2004-10-07  Bob Wilson  <bob.wilson@acm.org>
1007             Sterling Augustine  <sterling@tensilica.com>
1008
1009         * config/tc-xtensa.c (absolute_literals_supported): New global flag.
1010         (UNREACHABLE_MAX_WIDTH): Define.
1011         (XTENSA_FETCH_WIDTH): Delete.
1012         (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
1013         prefer_const16, prefer_l32r): New global variables.
1014         (LIT4_SECTION_NAME): Define.
1015         (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
1016         (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
1017         (frag_flags struct): New.
1018         (xtensa_block_info struct): Move from tc-xtensa.h.  Add flags field.
1019         (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
1020         (bitfield, bit_is_set, set_bit, clear_bit): Define.
1021         (MAX_FORMATS): Define.
1022         (op_placement_info struct, op_placement_table): New.
1023         (O_pltrel, O_hi16, O_lo16): Define.
1024         (directiveE enum): Rename directive_generics to directive_transform.
1025         Delete directive_relax.  Add directive_schedule,
1026         directive_absolute_literals, and directive_last_directive.
1027         (directive_info): Rename "generics" to "transform".  Delete "relax".
1028         Add "schedule" and "absolute-literals".
1029         (directive_state): Adjust entries to match changes in directive_info.
1030         (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
1031         (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
1032         xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
1033         (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
1034         (align_only_targets, software_a0_b_retw_interlock,
1035         software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
1036         software_avoid_short_loop, software_avoid_close_loop_end,
1037         software_avoid_all_short_loops, specific_opcode): Delete.
1038         (warn_unaligned_branch_targets): New.
1039         (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
1040         workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
1041         (option_[no_]link_relax, option_[no_]transform,
1042         option_[no_]absolute_literals, option_warn_unaligned_targets,
1043         option_prefer_l32r, option_prefer_const16, option_target_hardware):
1044         New enum values.
1045         (option_[no_]align_only_targets, option_literal_section_name,
1046         option_text_section_name, option_data_section_name,
1047         option_bss_section_name, option_eb, option_el): Delete.
1048         (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
1049         warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
1050         and target-hardware.  Delete entries for [no-]target-align-only,
1051         literal-section-name, text-section-name, data-section-name, and
1052         bss-section-name.
1053         (md_parse_option): Handle new options and remove old ones.  Accept but
1054         ignore [no-]density options.  Warn for [no-]generics and [no-]relax
1055         and treat them as [no-]transform.
1056         (md_show_usage): Add new options and remove old ones.
1057         (xtensa_setup_hw_workarounds): New.
1058         (md_pseudo_table): Change "word" entry to use xtensa_elf_cons.  Add
1059         "long", "short", "loc" and "frequency" entries.
1060         (use_generics): Rename to ...
1061         (use_transform): ... this function.  Add past_xtensa_end check.
1062         (use_longcalls): Add past_xtensa_end check.
1063         (code_density_available, can_relax): Delete.
1064         (do_align_targets): New.
1065         (get_directive): Accept dashes in directive names.  Warn about
1066         [no-]generics and [no-]relax directives and treat them as
1067         [no-]transform.
1068         (xtensa_begin_directive): Call md_flush_pending_output only for some
1069         directives.  Check for directives inside instruction bundles.  Warn
1070         about deprecated ".begin literal" usage.  Warn and ignore [no-]density
1071         directives.  Handle new directives.  Check generating_literals flag
1072         for literal_prefix.
1073         (xtensa_end_directive): Check for directives inside instruction
1074         bundles.  Warn and ignore [no-]density directives.  Handle new
1075         directives.  Call xtensa_set_frag_assembly_state.
1076         (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
1077         xtensa_dwarf2_emit_insn): New.
1078         (xtensa_literal_position): Call md_flush_pending_output.  Do not check
1079         use_literal_section flag.
1080         (xtensa_literal_pseudo): Call md_flush_pending_output.  Handle absolute
1081         literals.  Use xtensa_elf_cons to parse the expression.
1082         (xtensa_literal_prefix): Do not check use_literal_section.  Support
1083         ".lit4" sections for absolute literals.  Change prefix convention to
1084         replace ".text" (or ".t" in a linkonce section).  No need to call
1085         subseg_set.
1086         (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
1087         (expression_end): Handle closing braces and colons.
1088         (PLT_SUFFIX, plt_suffix): Delete.
1089         (expression_maybe_register): Use new xtensa-isa.h functions.  Use
1090         xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
1091         and O_hi16 expressions as well.
1092         (tokenize_arguments): Handle closing braces and colons.
1093         (parse_arguments): Use new xtensa-isa.h functions.  Handle "invisible"
1094         operands and paired register syntax.
1095         (get_invisible_operands): New.
1096         (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax.  Use
1097         new xtensa-isa.h functions.
1098         (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
1099         (xg_translate_idioms): Check if inside bundle.  Use use_transform.
1100         Handle new Xtensa LX RSR/WSR/XSR syntax.  Remove code to widen density
1101         instructions.  Use xtensa_translate_zero_immed.
1102         (operand_is_immed, operand_is_pcrel_label): Delete.
1103         (get_relaxable_immed): Use new xtensa-isa.h functions.
1104         (get_opcode_from_buf): Add slot parameter.  Use new xtensa-isa.h
1105         functions.
1106         (xtensa_print_insn_table, print_vliw_insn): New.
1107         (is_direct_call_opcode): Use new xtensa-isa.h functions.
1108         (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
1109         is_branch_or_jump_opcode): Delete.
1110         (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
1111         (opnum_to_reloc, reloc_to_opnum): Delete.
1112         (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
1113         xtensa-isa.h functions.  Operate on one slot of an instruction.
1114         (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
1115         xg_get_insn_size): Delete.
1116         (xg_get_build_instr_size): Use xg_get_single_size.
1117         (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
1118         xg_build_widen_table.  Use xg_get_single_size.
1119         (xg_get_max_narrow_insn_size): Delete.
1120         (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
1121         xg_is_relaxable_insn): Update calls to xg_build_widen_table.  Use
1122         xg_get_single_size.
1123         (xg_build_to_insn): Record the loc field.  Handle OP_OPERAND_HI16U and
1124         OP_OPERAND_LOW16U.  Check xg_valid_literal_expression.
1125         (xg_expand_to_stack, xg_expand_narrow): Update calls to
1126         xg_build_widen_table.  Use xg_get_single_size.
1127         (xg_immeds_fit): Use new xtensa-isa.h functions.  Update call to
1128         xg_check_operand.
1129         (xg_symbolic_immeds_fit): Likewise.  Also handle O_lo16 and O_hi16, and
1130         treat weak symbols conservatively.
1131         (xg_check_operand): Use new xtensa-isa.h functions.
1132         (is_dnrange): Delete.
1133         (xg_assembly_relax): Inline previous calls to tinsn_copy.
1134         (xg_finish_frag): Specify separate relax states for the frag and slot0.
1135         (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
1136         xtensa-isa.h functions.
1137         (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
1138         xg_instruction_matches_options): New.
1139         (xg_instruction_matches_rule): Handle O_register expressions.  Call
1140         xg_instruction_matches_options.
1141         (transition_rule_cmp): New.
1142         (xg_instruction_match): Update call to xg_build_simplify_table.
1143         (xg_build_token_insn): Record loc fields.
1144         (xg_simplify_insn): Check is_specific_opcode field and
1145         density_supported flag.
1146         (xg_expand_assembly_insn): Skip checking code_density_available.  Use
1147         new xtensa-isa.h functions.  Call use_transform instead of can_relax.
1148         (xg_assemble_literal): Add error handling for O_big.  Call
1149         record_alignment.  Handle O_pltrel.
1150         (xg_valid_literal_expression): New.
1151         (xg_assemble_literal_space): Add slot parameter.  Remove call to
1152         set_expr_symbol_offset.  Add call to record_alignment.  Update call to
1153         xg_finish_frag.
1154         (xg_emit_insn): Delete.
1155         (xg_emit_insn_to_buf): Add format parameter.  Update calls to
1156         xg_add_opcode_fix and xtensa_insnbuf_to_chars.
1157         (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
1158         and slot parameters.  Handle new "alternate" relocations for absolute
1159         literals and CONST16 instructions.  Check for bad uses of O_lo16 and
1160         O_hi16.  Use new xtensa-isa.h functions.
1161         (xg_assemble_tokens): Delete.
1162         (is_register_writer): Use new xtensa-isa.h functions.
1163         (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
1164         old-style RSR from LCOUNT.
1165         (next_frag_opcode): Delete.
1166         (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
1167         update_next_frag_state): New.
1168         (update_next_frag_nop_state): Delete.
1169         (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
1170         (xtensa_mark_literal_pool_location): Check use_literal_section flag and
1171         the state of the absolute-literals directive.  Add calls to
1172         record_alignment and xtensa_set_frag_assembly_state.  Call
1173         xtensa_switch_to_non_abs_literal_fragment instead of
1174         xtensa_switch_to_literal_fragment.
1175         (build_nop): New.
1176         (assemble_nop): Use build_nop.  Update call to xtensa_insnbuf_to_chars.
1177         (get_expanded_loop_offset): Change check for undefined opcode to an
1178         assertion.
1179         (xtensa_set_frag_assembly_state, relaxable_section,
1180         xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
1181         xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
1182         (md_begin): Update call to xtensa_isa_init.  Initialize linkrelax to 1.
1183         Set lit4_seg_name.  Call xg_init_vinsn.  Initialize new global opcodes.
1184         Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
1185         (xtensa_init_fix_data): New.
1186         (xtensa_frob_label): Reset label symbol to the current frag.  Check
1187         do_align_targets and generating_literals flag.  Propagate frequency
1188         info to new alignment frag.  Call xtensa_set_frag_assembly_state.
1189         (xtensa_unrecognized_line): New.
1190         (xtensa_flush_pending_output): Check if inside a bundle.  Add a call
1191         to xtensa_set_frag_assembly_state.
1192         (error_reset_cur_vinsn): New.
1193         (md_assemble): Remove check for literal frag.  Remove call to
1194         istack_init.  Call use_transform instead of use_generics.  Parse
1195         explicit instruction format specifiers.  Move code for
1196         a0_b_retw_interlock workaround to xg_assemble_vliw_tokens.  Call
1197         error_reset_cur_vinsn on errors.  Add call to get_invisible_operands.
1198         Add dwarf2_where call.  Remote automatic alignment for ENTRY
1199         instructions.  Move call to xtensa_clear_insn_labels to the end.
1200         Rearrange to handle bundles.
1201         (xtensa_cons_fix_new): Delete.
1202         (xtensa_handle_align): New.
1203         (xtensa_frag_init): Call xtensa_set_frag_assembly_state.  Remove
1204         assignment to is_no_density field.
1205         (md_pcrel_from): Use new xtensa-isa.h functions.  Use decode_reloc
1206         instead of reloc_to_opnum.  Handle "alternate" relocations.
1207         (xtensa_force_relocation, xtensa_check_inside_bundle,
1208         xtensa_elf_section_change_hook): New.
1209         (xtensa_symbol_new_hook): Delete.
1210         (xtensa_fix_adjustable): Check for difference of symbols with an
1211         offset.  Check for external and weak symbols.
1212         (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
1213         (md_estimate_size_before_relax): Return expansion for the first slot.
1214         (tc_gen_reloc): Handle difference of symbols by producing
1215         XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
1216         into the output.  Handle new XTENSA_SLOT*_OP relocs by storing the
1217         tentative values into the output when linkrelax is set.
1218         (XTENSA_PROP_SEC_NAME): Define.
1219         (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
1220         Create literal tables only if using literal sections.  Create new
1221         property tables instead of old instruction tables.  Check for unaligned
1222         branch targets and loops.
1223         (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
1224         new_resource_table, clear_resource_table, resize_resource_table,
1225         resources_available, reserve_resources, release_resources,
1226         opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
1227         resources_conflict, xg_find_narrowest_format, relaxation_requirements,
1228         bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
1229         (xtensa_end): Call xtensa_flush_pending_output.  Set past_xtensa_end
1230         flag.  Update checks for workaround options.  Call
1231         xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
1232         (xtensa_cleanup_align_frags): Add special case for branch targets.
1233         Check for and mark unreachable frags.
1234         (xtensa_fix_target_frags): Remove use of align_only_targets flag.
1235         Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
1236         end of a zero-overhead loop body.
1237         (frag_can_negate_branch): Handle instructions with multiple slots.
1238         Use new xtensa-isa.h functions
1239         (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
1240         xtensa_mark_zcl_first_insns): New.
1241         (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
1242         transformations are disabled.
1243         (next_instrs_are_b_retw): Use new xtensa-isa.h functions.  Handle
1244         multislot instructions.
1245         (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
1246         Likewise.  Also error if transformations are disabled.
1247         (unrelaxed_frag_max_size): New.
1248         (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
1249         xtensa-isa.h functions.
1250         (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
1251         xtensa_opcode_is_loop instead of is_loop_opcode.
1252         (get_text_align_power): Replace as_fatal with assertion.
1253         (get_text_align_fill_size): Iterate instead of using modulus when
1254         use_nops is false.
1255         (get_noop_aligned_address): Assert that this is for a machine-dependent
1256         RELAX_ALIGN_NEXT_OPCODE frag.  Use next_frag_opcode_is_loop,
1257         xg_get_single_size, and frag_format_size.
1258         (get_widen_aligned_address): Rename to ...
1259         (get_aligned_diff): ... this function.  Add max_diff parameter.
1260         Remove handling of rs_align/rs_align_code frags.  Use
1261         next_frag_format_size, get_text_align_power, get_text_align_fill_size,
1262         next_frag_opcode_is_loop, and xg_get_single_size.  Compute max_diff
1263         and pass it back to caller.
1264         (xtensa_relax_frag): Use relax_frag_loop_align.  Add code for new
1265         RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
1266         RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types.  Check relax_seen.
1267         (relax_frag_text_align): Rename to ...
1268         (relax_frag_loop_align): ... this function.  Assume loops can only be
1269         in the first slot of an instruction.
1270         (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
1271         instruction.  Remove call to frag_wane.
1272         (relax_frag_narrow): Rename to ...
1273         (relax_frag_for_align): ... this function.  Extend to handle
1274         RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
1275         RELAX_NARROW for the first slot.
1276         (find_address_of_next_align_frag, bytes_to_stretch): New.
1277         (future_alignment_required): Use find_address_of_next_align_frag and
1278         bytes_to_stretch.  Look ahead to subsequent frags to make smarter
1279         alignment decisions.
1280         (relax_frag_immed): Add format, slot, and estimate_only parameters.
1281         Check if transformations are enabled for b_j_loop_end workaround.
1282         Use new xtensa-isa.h functions and handle multislot instructions.
1283         Update call to xg_assembly_relax.
1284         (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
1285         RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
1286         frag types.
1287         (convert_frag_narrow): Add segP, format and slot parameters.  Call
1288         convert_frag_immed for branch instructions.  Adjust calls to
1289         tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf.  Use
1290         xg_get_single_size and xg_get_single_format.
1291         (convert_frag_fill_nop): New.
1292         (convert_frag_immed): Add format and slot parameters.  Handle multislot
1293         instructions and use new xtensa-isa.h functions.  Update calls to
1294         tinsn_immed_from_frag and xg_assembly_relax.  Check if transformations
1295         enabled for b_j_loop_end workaround.  Use build_nop instead of
1296         assemble_nop.  Check is_specific_opcode flag.  Check for unreachable
1297         frags.  Use xg_get_single_size.  Handle O_pltrel.
1298         (fix_new_exp_in_seg): Remove check for old plt flag.
1299         (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
1300         xtensa_insnbuf_to_chars.  Call tinsn_immed_from_frag.  Change check
1301         for loop opcode to an assertion.  Mark all frags up to the end of the
1302         loop as not transformable.
1303         (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
1304         (get_subseg_info): New.
1305         (xtensa_move_literals): Call xtensa_set_frag_assembly_state.  Add null
1306         check for dest_seg.
1307         (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
1308         literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
1309         (xtensa_switch_to_non_abs_literal_fragment): New.
1310         (cache_literal_section): Add is_code parameter and pass it through to
1311         retrieve_literal_seg.
1312         (retrieve_literal_seg): Add is_code parameter and use it to set the
1313         flags on the literal section.  Handle case where head parameter is 0.
1314         (get_frag_is_no_transform, set_frag_is_specific_opcode,
1315         set_frag_is_no_transform): New.
1316         (xtensa_create_property_segments): Add end_property_function parameter
1317         and pass it through to add_xt_block_frags.  Call bfd_get_section_flags
1318         and skip SEC_DEBUGGING and !SEC_ALLOC sections.
1319         (xtensa_create_xproperty_segments, section_has_xproperty): New.
1320         (add_xt_block_frags): Add end_property_function parameter and call it
1321         if it is non-zero.  Call xtensa_frag_flags_init.
1322         (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
1323         get_frag_property_flags, frag_flags_to_number,
1324         xtensa_frag_flags_combinable, xt_block_aligned_size,
1325         xtensa_xt_block_combine, add_xt_prop_frags,
1326         init_op_placement_info_table, opcode_fits_format_slot,
1327         xg_get_single_size, xg_get_single_format): New.
1328         (istack_push): Inline call to tinsn_copy.
1329         (tinsn_copy): Delete.
1330         (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
1331         CONST16 opcodes.  Handle O_big, O_illegal, and O_absent.
1332         (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
1333         (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
1334         functions.  Handle invisible operands.
1335         (tinsn_to_slotbuf): New.
1336         (tinsn_check_arguments): Use new xtensa-isa.h functions.
1337         (tinsn_from_chars): Add slot parameter.  Rewrite using xg_init_vinsn,
1338         vinsn_from_chars, and xg_free_vinsn.
1339         (tinsn_from_insnbuf): New.
1340         (tinsn_immed_from_frag): Add slot parameter and handle multislot
1341         instructions.  Handle symbol differences.
1342         (get_num_stack_text_bytes): Use xg_get_single_size.
1343         (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
1344         xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
1345         get_expr_register, set_expr_symbol_offset_diff): New.
1346         * config/tc-xtensa.h (MAX_SLOTS): Define.
1347         (xtensa_relax_statesE): Move from tc-xtensa.c. Add
1348         RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
1349         RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
1350         RELAX_NONE types.
1351         (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
1352         (xtensa_frag_type struct): Add is_assembly_state_set,
1353         use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
1354         is_align, is_text_align, alignment, and is_first_loop_insn fields.
1355         Replace is_generics and is_relax fields by is_no_transform field.
1356         Delete is_text and is_longcalls fields.  Change text_expansion and
1357         literal_expansion to arrays of MAX_SLOTS entries.  Add arrays of
1358         per-slot information: literal_frags, slot_subtypes, slot_symbols,
1359         slot_sub_symbols, and slot_offsets.  Add fr_prev field.
1360         (xtensa_fix_data struct): New.
1361         (xtensa_symfield_type struct): Delete plt field.
1362         (xtensa_block_info struct): Move definition to tc-xtensa.h.  Add
1363         forward declaration here.
1364         (xt_section_type enum): Delete xt_insn_sec.  Add xt_prop_sec.
1365         (XTENSA_SECTION_RENAME): Undefine.
1366         (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
1367         tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
1368         HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
1369         (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
1370         (unit_num_copies_func, opcode_num_units_func,
1371         opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
1372         (resource_table struct): New.
1373         * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
1374         (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
1375         literal_space, symbol, sub_symbol, offset, and literal_frag fields.
1376         (tinsn_copy): Delete prototype.
1377         (vliw_insn struct): New.
1378         * config/xtensa-relax.c (insn_pattern_struct): Add options field.
1379         (widen_spec_list): Add option conditions for density and boolean
1380         instructions.  Add expansions using CONST16 and conditions for using
1381         CONST16 vs. L32R.  Use new Xtensa LX RSR/WSR syntax.  Add entries for
1382         predicted branches.
1383         (simplify_spec_list): Add option conditions for density instructions.
1384         Add entry for NOP instruction.
1385         (append_transition): Add cmp function pointer parameter and use it to
1386         insert the new entry in order.
1387         (operand_function_LOW16U, operand_function_HI16U): New.
1388         (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
1389         OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1390         (enter_opname, split_string): Use xstrdup instead of strdup.
1391         (init_insn_pattern): Initialize new options field.
1392         (clear_req_or_option_list, clear_req_option_list,
1393         clone_req_or_option_list, clone_req_option_list, parse_option_cond):
1394         New.
1395         (parse_insn_pattern): Parse option conditions.
1396         (transition_applies): New.
1397         (build_transition): Use new xtensa-isa.h functions.  Fix incorrectly
1398         swapped last arguments in calls to append_constant_value_condition.
1399         Call clone_req_option_list.  Add warning about invalid opcode.
1400         Handle LOW16U and HI16U function names.
1401         (build_transition_table): Add cmp parameter and use it in calls to
1402         append_transition.  Use new xtensa-isa.h functions.  Check
1403         transition_applies before adding entries.
1404         (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
1405         pass it through to build_transition_table.
1406         * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
1407         ReqOption, transition_cmp_fn): New types.
1408         (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1409         (transition_rule struct): Add options field.
1410         * doc/as.texinfo (Overview): Update Xtensa options.
1411         * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
1412         --[no-]relax, and --[no-]generics options.  Update descriptions of
1413         --text-section-literals and --[no-]longcalls.  Add
1414         --[no-]absolute-literals and --[no-]transform.
1415         (Xtensa Syntax): Add description of syntax for FLIX instructions.
1416         Remove use of "generic" and "specific" terminology for opcodes.
1417         (Xtensa Registers): Generalize the syntax description to include
1418         user-defined register files.
1419         (Xtensa Automatic Alignment): Update.
1420         (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
1421         (Xtensa Call Relaxation): Linker can now remove most of the overhead.
1422         (Xtensa Directives): Remove confusing rules about precedence.
1423         (Density Directive, Relax Directive): Delete.
1424         (Schedule Directive): New.
1425         (Generics Directive): Rename to ...
1426         (Transform Directive): ... this node.
1427         (Literal Directive): Update for absolute literals.  Missing
1428         literal_position directive is now an error.
1429         (Literal Position Directive): Update for absolute literals.
1430         (Freeregs Directive): Delete.
1431         (Absolute Literals Directive): New.
1432         (Frame Directive): Minor editing.
1433         * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
1434         Update dependencies.
1435         * Makefile.in: Regenerate.
1436
1437 2004-10-07  Richard Sandiford  <rsandifo@redhat.com>
1438
1439         * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
1440         to build the second and third fixups for a composite relocation.
1441         (macro_read_relocs): New function.
1442         (macro_build): Use it.
1443         (s_cpsetup): Pass all three composite relocation codes to macro_build.
1444         Simplify fragging code accordingly.
1445         (s_gpdword): Use fix_new rather than fix_new_exp for the second part
1446         of the composite relocation.  Set fx_tcbit in both fixups.
1447
1448 2004-10-07  Richard Sandiford  <rsandifo@redhat.com>
1449
1450         * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
1451         (md_apply_fix3): Don't treat composite relocs as done.
1452
1453 2004-10-07  Jan Beulich <jbeulich@novell.com>
1454
1455         * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
1456         prefix for symbol names generated from the LOCAL macro directive.
1457
1458         * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
1459         FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
1460
1461 2004-10-07  Tomer Levi  <Tomer.Levi@nsc.com>
1462
1463         * config/tc-crx.c (preprocess_reglist): Handle Co-processor
1464         Special registers.
1465         (md_assemble): Add error checking for Co-Processor instructions.
1466         (get_cinv_parameters): Add 'b' option to invalidate the
1467         branch-target cache.
1468
1469 2004-10-05  Paul Brook  <paul@codesourcery.com>
1470
1471         * config/tc-arm.c (unwind): New variable.
1472         (vfp_sp_encode_reg): New function.
1473         (vfp_sp_reg_required_here): Use it.
1474         (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
1475         (vfp_parse_reg_list): New function.
1476         (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
1477         s_arm_unwind_personality, s_arm_unwind_personalityindex,
1478         s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
1479         s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
1480         functions.
1481         (md_pseudo_table): Add them.
1482         (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
1483         (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
1484         (set_section, add_unwind_adjustsp, flush_pending_unwind,
1485         finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
1486         require_hashconst, add_unwind_opcode): New functions.
1487         * doc/tc-arm.text: Document unwinding opcodes.
1488         * NEWS: Mention the new feature.
1489
1490 2004-10-04  Eric Christopher  <echristo@redhat.com>
1491
1492         * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
1493
1494 2004-10-01  H.J. Lu  <hongjiu.lu@intel.com>
1495
1496         * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
1497         TLS relocations.
1498         * config/tc-s390.c (md_apply_fix3): Likewise.
1499         * config/tc-sparc.c (md_apply_fix3): Likewise.
1500
1501 2004-10-01  Paul Brook  <paul@codesourcery.com>
1502
1503         * config/tc-arm.c (arm_elf_section_type): New function.
1504         (arm_elf_change_section): Set section link for exidx sections.
1505         * config/tc-arm.h (arm_elf_section_type): Add prototype.
1506         (md_elf_section_type): Define.
1507
1508 2004-10-01  Bill Farmer  <Bill@the-farmers.freeserve.co.uk>
1509
1510         * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
1511         instruction's offset.
1512
1513 2004-10-01  Adam Nemet  <anemet@lnxw.com>
1514
1515         * (TARGET_FORMAT): Remove LynxOS COFF definition.
1516
1517 2004-10-01  Ravi Ramaseshan  <ravi.ramaseshan@codito.com>
1518
1519         * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
1520         asymbol *, instead use symbol_get_bfdsym.
1521
1522 2004-09-30  Linus Nielsen Feltzing  <linus@haxx.se>
1523
1524         * config/tc-m68k.c (select_control_regs): Add mcf5249.
1525
1526 2004-09-30  Paul Brook  <paul@codesourcery.com>
1527
1528         * config/tc-arm.c (do_smi, do_nop): New functions.
1529         (insns): Add ARMv6ZK instructions.
1530         (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
1531         (tc_gen_reloc): Ditto.
1532         (arm_cpus): Add mpcore and arm1176.
1533         (arm_archs): Add armv6{k,z,zk}.
1534         * doc/c-arm.texi: Document new cores and architectures.
1535
1536 2004-09-30  Nick Clifton  <nickc@redhat.com>
1537
1538         * config/tc-arm.c: Use ISO C90 formatting.
1539
1540 2004-09-30  Vladimir Ivanov  <vladitx@nucleusys.com>
1541
1542         * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
1543         as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
1544         REG_TYPE_MVDX is expected.
1545
1546 2004-09-29  Marc Bevand  <m.bevand@gmail.com>
1547
1548         * doc/c-i386.texi (i386-Mnemonics): Fix typo.
1549
1550 2004-09-21  James E Wilson  <wilson@specifixinc.com>
1551
1552         * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
1553         (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
1554         output_spill_base, output_unat_psprel, output_lc_psprel,
1555         output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
1556         output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
1557         output_spill_psprel_p): Use it.
1558
1559 2004-09-20  Tomer Levi  <Tomer.Levi@nsc.com>
1560
1561         * config/tc-crx.c (handle_LoadStor): New function.
1562         Handle load/stor unique instructions before parsing.
1563
1564 2004-09-17  Paul Brook  <paul@codesourcery.com>
1565
1566         * config/tc-arm.c (s_arm_rel31): New funciton.
1567         (md_pseudo_table): Add .rel31.
1568         (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1569         BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1570         (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1571         (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1572         (arm_parse_reloc): Add (target2).
1573
1574 2004-09-17  Alan Modra  <amodra@bigpond.net.au>
1575
1576         * Makefile.am: Run "make dep-am".
1577         * Makefile.in: Regenerate.
1578         * aclocal.m4: Regenerate.
1579         * configure: Regenerate.
1580         * doc/Makefile.in: Regenerate.
1581         * po/POTFILES.in: Regenerate.
1582         * po/gas.pot: Regenerate.
1583
1584 2004-09-14  Hideki IWAMOTO  <h-iwamoto@kit.hi-ho.ne.jp>
1585
1586         * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1587         [!LLONG_MAX]: Ditto.
1588
1589 2004-09-13  Paul Brook  <paul@codesourcery.com>
1590
1591         * config/tc-arm.c: Rename RELABS to TARGET1.
1592
1593 2004-09-13  Alan Modra  <amodra@bigpond.net.au>
1594
1595         * messages.c (as_internal_value_out_of_range): Cast values passed
1596         to as_bad_where or as_warn_where to proper type.
1597
1598 2004-09-11  Theodore A. Roth  <troth@openavr.org>
1599
1600         * config/tc-avr.c: Add support for
1601         atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1602
1603 2004-09-09  Alan Modra  <amodra@bigpond.net.au>
1604
1605         * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1606         from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1607
1608 2004-09-08  Paul Brook  <paul@codesourcery.com>
1609
1610         * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1611         fini_array and preinit_array section types.
1612         * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1613         and fini_array.
1614         * doc/as.texinfo: Document extra section types.
1615
1616 2004-09-02  Mark Mitchell  <mark@codesourcery.com>
1617
1618         * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1619         * Makefile.in: Regenerated.
1620         * configure.in: Set em for arm*-*-symbianelf*.
1621         * configure: Regenerated.
1622         * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1623         vectors when appropriate.
1624         * config/te-symbian.h: New file.
1625
1626 2004-09-03  Tomer Levi  <Tomer.Levi@nsc.com>
1627
1628         * config/tc-crx.c (gettrap): Exception vector can be case
1629         insensitive.
1630         (process_label_constant): Fix a 32-bit displacement bug in branch
1631         instructions.
1632         (get_operandtype) : Bug fix, wrong operand was used.
1633         (process_label_constant): Initialize relocation type to
1634         BFD_RELOC_NONE
1635
1636 2004-09-01  Richard Earnshaw < reanrsha@arm.com>
1637
1638         * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1639         for synthesizable cores.
1640
1641         * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1642
1643 2004-08-25  Dmitry Diky  <diwil@spec.ru>
1644
1645         * config/tc-msp430.c:   Clean-up the code.
1646         (md_relax_table): New relax table.
1647         (mcu_types): Sort MCU types.
1648         (md_pseudo_table): Add .profiler pseudo handler.
1649         (pow2value): New function.
1650         (msp430_profiler): New function.
1651         (msp430_operands): Add new insns handlers.
1652         (msp430_srcoperand): Add register operand handler, allow complex
1653         expressions.
1654         (md_estimate_size_before_relax): Rewritten.
1655         (md_convert_frag): Rewritten.
1656         (msp430_relax_frag): New function.
1657         * config/tc-msp430.h (md_relax_frag): define macro
1658         * doc/c-msp430.texi: Update information.
1659
1660 2004-08-24  Nick Clifton  <nickc@redhat.com>
1661
1662         * as.c (std_shortopts): Allow -g to take an optional argument.
1663         (parse_args): Pass any switch starting with -g on to the backend
1664         for parsing.
1665
1666 2004-08-18  Mark Mitchell  <mark@codesourcery.com>
1667
1668         * configure.in (arm*-*-symbianelf*): New target.
1669         (arm*-*-eabi*): Likewise.
1670         * configure: Regenerated.
1671
1672 2004-08-18  Thiemo Seufer  <seufer@csv.ica.uni-stuttgart.de>
1673         * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1674         correctly.
1675
1676 2004-08-18  Jakub Jelinek  <jakub@redhat.com>
1677
1678         * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1679         argument, don't do anything if current section is not
1680         .gnu.linkonce.t.* and linkonce_empty is set.
1681         (generate_unwind_image, dot_endp): Adjust callers, call
1682         start_unwind_section (*, 1) if nothing will be put into the
1683         section.
1684
1685 2004-08-17  Nick Clifton  <nickc@redhat.com>
1686
1687         * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1688         (show_usage): Add -g.
1689         (std_longopts): Add --gen-debug.  Alpha sort the table.
1690         (parse_args): Print an error message if a switch is not handled.
1691         Handle the -g switch, calling md_debug_format_selector() if
1692         necessary.
1693         * NEWS: Mention new feature.
1694         * doc/as.texinfo: Document new switch.
1695         * doc/internals.texi: Document behaviour of md_parse_option.
1696
1697         * config/tc-arm.c (md_parse_option): Do not issue an error message
1698         if the switch is not recognised.
1699         * config/tc-m68k.c (md_parse_option): Likewise.
1700         * config/tc-pdp11.c (md_parse_option): Likewise.
1701         * config/tc-v850.c (md_parse_option): Likewise.
1702
1703         * as.h: Fix up formatting.
1704         * tc.h: Likewise.
1705
1706 2004-08-16  Nick Clifton  <nickc@redhat.com>
1707
1708         * macro.c (macro_set_alternate): Use ISO C90 formatting.
1709
1710         * configure.in: Sort architecture based tables alphabetically.
1711         * configure: Regenerate.
1712
1713 2004-08-16  Alan Modra  <amodra@bigpond.net.au>
1714
1715         * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1716         { "cc", 68 }, with { "cr", 70 }.
1717
1718 2004-08-13 Jan Beulich <jbeulich@novell.com>
1719            Nick Clifton <nickc@redhat.com>
1720
1721         * as.c: Add and handle new --alternate command line option.
1722         * macro.c (macro_set_alternate): New.
1723         * macro.h (macro_set_alternate): Declare.
1724         * read.c: Add and handle new .altmacro and .noaltmacro directives.
1725         * doc/as.texinfo: Document new command line option and pseudo-ops
1726         as well as insert documentation originating from gasp about
1727         alternate macro syntax.
1728         * NEWS: Mention new command line option and pseudo-ops.
1729
1730 2004-08-10  Mark Mitchell  <mark@codesourcery.com>
1731
1732         * expr.c (operand): Handle the "~", "-", and "!" operators applied
1733         to bignums.
1734
1735 2004-08-06  Paul Brook  <paul@codesourcery.com>
1736
1737         * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1738         Handle new relocations.
1739         * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1740
1741 2004-08-05  Bob Wilson  <bob.wilson@acm.org>
1742
1743         * write.c (relax_segment): Use was_address instead of address when
1744         setting fr_fix field for align frag due to backwards .org.
1745
1746 2004-07-29  Alexandre Oliva  <aoliva@redhat.com>
1747
1748         Introduce SH2a support.
1749         2004-02-24  Corinna Vinschen  <vinschen@redhat.com>
1750         * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1751         arch_sh2a_nofpu_up.
1752         2004-02-24  Corinna Vinschen  <vinschen@redhat.com>
1753         * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1754         2004-02-20  Corinna Vinschen  <vinschen@redhat.com>
1755         * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1756         to end of conditional expression.
1757         2004-02-20  Corinna Vinschen  <vinschen@redhat.com>
1758         * config/tc-sh.c: Add sh2a-nofpu support.
1759         2003-12-29  DJ Delorie  <dj@redhat.com>
1760         * tc-sh.c: Add sh2a support.
1761         (parse_reg): Add tbr.
1762         (parse_at): Support @@(disp,tbr).
1763         (get_specific): Support sh2a opcodes.
1764         (insert4): New, for 4 byte relocs.
1765         (build_Mytes): Support sh2a opcodes.
1766         (md_apply_fix3_Mytes): Support sh2a opcodes.
1767         2003-12-02  Michael Snyder  <msnyder@redhat.com>
1768         * config/tc-sh.c (md_parse_option): Handle sh2a.
1769         (sh_elf_final_processing): Ditto.
1770
1771 2004-07-27  Jason Thorpe  <thorpej@wasabisystems.com>
1772
1773         * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1774         for TE_NetBSD.
1775
1776 2004-07-27  Alan Modra  <amodra@bigpond.net.au>
1777
1778         * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1779         (ppc_arch): Expand comment.
1780
1781 2004-07-27  Tomer Levi  <Tomer.Levi@nsc.com>
1782
1783         * config/tc-crx.c: Support evaluating the difference between two
1784         symbols.
1785         * config/tc-crx.h: Likewise.
1786
1787 2004-07-26  H.J. Lu  <hongjiu.lu@intel.com>
1788
1789         * config/tc-ia64.c (start_unwind_section): Set the linked-to
1790         section.
1791         (ia64_elf_section_change_hook): Set the linked-to section for
1792         SHT_IA_64_UNWIND.
1793
1794 2004-07-26  Dmitry Diky  <diwil@spec.ru>
1795
1796         * config/tc-msp430.c: Add new subtargets: msp430x1610,
1797         msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1798         msp430xG438, msp430xG439.
1799
1800 2004-07-25  Daniel Jacobowitz  <dan@debian.org>
1801
1802         * doc/as.texinfo (Section, PushSection): Correct documentation
1803         for ELF.
1804
1805 2004-07-21  Jan Beulich  <jbeulich@novell.com>
1806
1807         * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1808         permissible for the selected instruction suffix.
1809         (match_template): Don't permit 64-bit general purpose operands in
1810         32-bit mode.
1811         (finalize_imm): Permit 64-bit immediates.
1812         (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1813         specially except for the width of the used base and/or index
1814         registers.  For 32-bit displacements, use sign-extended
1815         relocations only when using 64-bit addressing.
1816         Force zero displacement on rip-relative addressing when there is
1817         no other displacement.
1818         (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1819         specially except for the width of the used base and/or index
1820         registers.
1821         (parse_register): Disallow Reg64 registers in 32-bit mode.
1822
1823         * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1824         suffix if the instruction doesn't support it.
1825
1826 2004-07-20  Maciej W. Rozycki  <macro@linux-mips.org>
1827
1828         * config/tc-mips.c (append_insn): Handle constant expressions with
1829         no associated relocation.
1830         (mips_ip): Cancel the expression after use for the Q format
1831         specifier.
1832         (parse_relocation): Return no relocation for unsupported
1833         operators.
1834         (my_getSmallExpression): Return no relocation if no relocation
1835         operators are used.
1836
1837 2004-07-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
1838
1839         * config/obj-som.c (adjust_stab_sections): Add prototype.
1840         (obj_som_compiler, obj_som_version, obj_som_copyright,
1841         adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1842         * config/tc-hppa.c (update_subspace):  Likewise.
1843         (is_defined_subspace): Amplify comment.
1844         * config/obj-som.h (som_frob_file): Add prototype.
1845
1846 2004-07-19  Christopher Faylor  <cgf@timesys.com>
1847             H.J. Lu  <hongjiu.lu@intel.com>
1848
1849         * subsegs.c (section_symbol): Don't create a new segment when
1850         existing segment is undefined.
1851
1852 2004-07-16  Richard Earnshaw  <rearnsha@arm.com>
1853
1854         * config/tc-arm.c: Include include/opcode/arm.h.
1855         (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1856         (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1857         (FPU_ARCH_*): Delete.
1858         * Makefile.am: Update dependencies.
1859         * Makefile.in: Regenerate.
1860
1861 2004-07-15  Nick Clifton  <nickc@redhat.com>
1862
1863         * configure.in: Accept armbe as a big-endian arm configuration.
1864         * configure: Regenerate.
1865
1866 2004-07-13  Thomas Nystrom  <thn@saeab.se>
1867
1868         * config/tc-i386.c (T_SHIFTOP): New constant.
1869         (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1870         (intel_el1): Handle '~'.
1871         (intel_get_token): Handle '<>', '&', '|' and '~'.
1872
1873 2004-07-13  Nick Clifton <nickc@redhat.com>
1874
1875         (md_assemble): Remove spurious newline from end of as_bad error
1876         message.
1877         (intel_e05_1): Likewise.
1878         (intel_e11): Likewise.
1879         (intel_match_token): Likewise.
1880
1881 2004-07-11  Andreas Schwab  <schwab@suse.de>
1882
1883         * config/tc-m68k.c: Convert to C90.  Remove redundant
1884         declarations.  Indentation fixup.
1885         [M68KCOFF]: Include "obj-coff.h" instead of declaring
1886         obj_coff_section ourselves.
1887
1888 2004-07-09  James E Wilson  <wilson@specifixinc.com>
1889
1890         * config/tc-ia64.c (default_big_endian): New.
1891         (dot_byteorder, md_begin): Use it.
1892         (md_parse_option): Set it.
1893
1894 2004-07-09  Nick Clifton  <nickc@redhat.com>
1895
1896         * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1897         * configure: Regenerate.
1898         * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1899         * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1900         sh_find_elf_flags if necessary.
1901
1902 2004-07-08  Richard Sandiford  <rsandifo@redhat.com>
1903
1904         * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1905         going to be split into more than one in-place addend, return 0
1906         for relocations against mergeable sections.  Associate comments
1907         with code.
1908
1909 2004-07-07  Tomer Levi  <Tomer.Levi@nsc.com>
1910
1911         * Makefile.am (CPU_TYPES): Add crx.
1912         (TARGET_CPU_CFILES): Add config/tc-crx.c.
1913         (TARGET_CPU_HFILES): Add config/tc-crx.h.
1914         (DEPTC_crx_elf): New target.
1915         (DEPOBJ_crx_elf): Likewise.
1916         (DEP_crx_elf): Likewise.
1917         * Makefile.in: Regenerate.
1918         * configure.in: Add crx* target.
1919         * configure: Regenerate.
1920         * config/tc-crx.c: New file.
1921         * config/tc-crx.h: New file.
1922         * NEWS: Mention new target.
1923
1924 2004-07-06  Nick Clifton  <nickc@redhat.com>
1925
1926         * config.in: Undefine TARGET_SYMBIAN by default.
1927         * configure.in:
1928         * configure: Regenerate. Add sh-symbian-elf target.  If
1929         selected define TARGET_SYMBIAN.
1930         * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1931         format if TARGET_SYMBIAN has been defined.
1932
1933         * output-file.c (output_file_create): Report the target format
1934         chosen when bfd_openw reports that it is invalid.
1935
1936         * config/obj-coff.c (coff_pseudo_table): Only define the weak
1937         pseudo for BFD based assemblers.
1938
1939 2004-07-05  Andrew Stubbs <andrew.stubbs@superh.com>
1940
1941         gas:
1942         * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1943         (md_parse_option): Remove redundant -isa testing.
1944         Make bfd_arch variable const.
1945         (md_show_usage): Make bfd_arch variable const.
1946
1947 2004-07-03  James E Wilson  <wilson@specifixinc.com>
1948
1949         * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1950         * config/tc-ia64.h (md_frag_check): Define.
1951
1952 2004-07-03  Aaron W. LaFramboise  <aaron98wiridge9@aaronwl.com>
1953
1954         * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1955         externals.
1956         * doc/as.texinfo (Weak): Document PE weak symbols.
1957
1958 2004-07-03  Richard Sandiford  <rsandifo@redhat.com>
1959
1960         * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1961         (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1962         (mips_frob_file): Rework so that only a single pass through the
1963         relocs is needed.  Allow %lo()s to have higher offsets than their
1964         corresponding %hi()s or %got()s.
1965
1966 2004-07-02  Nick Clifton  <nickc@redhat.com>
1967
1968         * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1969         allow values which have come from undefined symbols.
1970         Always consider this fixup to have been processed as a reloc
1971         cannot be generated for it.
1972
1973 2004-07-02  Alan Modra  <amodra@bigpond.net.au>
1974
1975         * frags.h (struct frag): Add has_code and insn_addr fields.
1976         * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1977         * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1978         * config/tc-ppc.h (md_frag_check): Define.
1979
1980 2004-06-28  Maciej W. Rozycki  <macro@linux-mips.org>
1981
1982         * doc/Makefile.am (info): Rename goal to...
1983         (info-local): ... this, to preserve implicit dependencies.
1984         * doc/Makefile.in: Regenerate with automake 1.8.5.
1985
1986 2004-06-25  Kazuhiro Inaoka  <inaoka.kazuhiro@renesas.com>
1987
1988         * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1989         (m32r_cgen_record_fixup_exp): Changed for  @GOTOFF, @GOT.
1990         (m32r_fix_adjustable): Changed for  @GOTOFF, @GOT, @PLT.
1991         (tc_gen_reloc): Likewise.
1992         (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1993         (m32r_parse_name): Likewise.
1994         (m32r_cgen_parse_fix_exp): Likewise.
1995         * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1996         (O_PIC_reloc): Likewise.
1997         (TC_CGEN_PARSE_FIX_EXP): Likewise..
1998         * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1999         for @GOTOFF, @GOT, @PLT.
2000
2001 2004-06-21 Jan Beulich  <jbeulich@novell.com>
2002
2003         * gas/symbols.c: While discarding ordinary local absolute symbols
2004         when --strip-local-absolute is in effect, retain file symbols.
2005
2006 2004-06-20  Andreas Schwab  <schwab@suse.de>
2007
2008         * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
2009         (md_parse_option): Likewise.
2010
2011 2004-06-17  Jan Beulich <jbeulich@novell.com>
2012
2013         * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
2014         * config/te-netware.h: New file.
2015         * config/te-ppcnw.h: Delete: Obsolete.
2016         * configure.in: Eliminate ill NetWare targets. Make generic
2017         NetWare target use proper emulation.
2018         * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
2019         reference to new te-netware.h.
2020         * configure: Regenerate.
2021         * Makefile.in: Regenerate.
2022
2023 2004-06-15  Martin Schwidefsky  <schwidefsky@de.ibm.com>
2024
2025         * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
2026         comparison in .insn pseudo operation.
2027
2028 2004-06-15  Alan Modra  <amodra@bigpond.net.au>
2029
2030         * config/obj-coff.c (coff_adjust_section_syms): Use
2031         bfd_get_section_size instead of bfd_get_section_size_before_reloc.
2032         (coff_frob_section): Likewise.
2033         * config/tc-mips.c (md_apply_fix3): Likewise.
2034         * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
2035         (elf_frob_file_after_relocs): Likewise.
2036
2037 2004-06-10  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
2038
2039         * config/tc-hppa.c (log2): Rename to exact_log2.
2040         (pa_next_subseg): Delete unused function.
2041         (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
2042         (create_new_subspace): Likewise.
2043
2044         Bug gas/213
2045         * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
2046         labels.  Fix warning.
2047
2048 2004-05-28  DJ Delorie  <dj@redhat.com>
2049
2050         * config/tc-mn10300.h (tc_fix_adjustable): Define.
2051         * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
2052         or non-merged symbols.
2053
2054 2004-05-28  H.J. Lu  <hongjiu.lu@intel.com>
2055
2056         * config/tc-ia64.c (remove_marked_resource): Save, clear and
2057         restore the old slot when inserting srlz.i/srlz.d.
2058
2059 2004-05-28  Andrew Stubbs <andrew.stubbs@superh.com>
2060
2061         * Makefile.am: Regenerate dependecies.
2062         * Makefile.in: Regenerate.
2063         * config/tc-sh.c (valid_arch): Make unsigned.
2064         (preset_target_arch): Likewise.
2065         (md_begin): Use new architecture flags system.
2066         (get_specific): Likewise.
2067         (assemble_ppi): Likewise.
2068         (md_assemble): Likewise. Also fix error check for bad opcodes.
2069         (md_parse_option): Likewise. Also generate -isa values according
2070         to the table in bfd/cpu-sh.c instead of just constants. Also
2071         allow <arch>-up ISA variants.
2072         (sh_elf_final_processing): Replace if-else chain with a call to
2073         sh_find_elf_flags().
2074         * testsuite/gas/sh/arch: New directory.
2075         * testsuite/gas/sh/arch/arch.exp: New test script.
2076         * testsuite/gas/sh/arch/arch_expected.txt: New file.
2077         * testsuite/gas/sh/arch/sh.s: New file.
2078         * testsuite/gas/sh/arch/sh2.s: New file.
2079         * testsuite/gas/sh/arch/sh-dsp.s: New file.
2080         * testsuite/gas/sh/arch/sh2e.s: New file.
2081         * testsuite/gas/sh/arch/sh3-nommu.s: New file.
2082         * testsuite/gas/sh/arch/sh3.s: New file.
2083         * testsuite/gas/sh/arch/sh3-dsp.s: New file.
2084         * testsuite/gas/sh/arch/sh3e.s: New file.
2085         * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
2086         * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
2087         * testsuite/gas/sh/arch/sh4.s: New file.
2088         * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
2089         * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
2090         * testsuite/gas/sh/arch/sh4a.s: New file.
2091
2092 2004-05-28  Peter Barada <peter@the-baradas.com>
2093
2094         * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
2095         with zero offset into mode 2 addressing to save a word.
2096
2097 2004-05-27  H.J. Lu  <hongjiu.lu@intel.com>
2098
2099         * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
2100         (ar_is_only_in_integer_unit): New.
2101         (ar_is_only_in_memory_unit): New.
2102         (generate_unwind_image): Silence gcc on 32bit host.
2103         (md_assemble): Use ar_is_only_in_integer_unit instead of
2104         ar_is_in_integer_unit. Check AR access.
2105
2106 2004-05-27  Peter Barada  <peter@the-baradas.com>
2107
2108         * config/tc-m68k.c (md_begin): Sort the opcode table into
2109         alphabetical order.
2110         (m68k_compare_opcode): New function to do the sorting.
2111
2112 2004-05-24  Peter Barada  <peter@the-baradas.com>
2113
2114         * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
2115         addressing on MIT style operands.
2116         * config/m68k-parse.y(yylex): Allow '-&' for predecrement
2117         w/mask addressing.
2118         * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
2119         operands.
2120
2121 2004-05-23  Alan Modra  <amodra@bigpond.net.au>
2122
2123         * expr.c (operand, operator): Don't reject '++' and '--'.
2124
2125 2004-05-20  Richard Sandiford  <rsandifo@redhat.com>
2126
2127         * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
2128         dwarf tables.
2129
2130 2004-05-17  Adam Nemet  <anemet@lnxw.com>
2131
2132         * configure.in: Add ppc-*-lynxos*.  Update i386-*-lynxos* to ELF.
2133         * configure: Regenerate.
2134
2135 2004-05-13  Paul Brook  <paul@codesourcery.com>
2136
2137         * dw2gencfi.c (output_cie):  Handle dwarf3 format CIE entries.
2138
2139 2004-05-13  Joel Sherrill <joel@oarcorp.com>
2140
2141         * configure.in (or32-*-rtems*): Switch to elf.
2142         * configure: Regenerate.
2143
2144 2004-05-13  Nick Clifton  <nickc@redhat.com>
2145
2146         * po/fr.po: Updated French translation.
2147
2148 2004-05-11  Nick Clifton  <nickc@redhat.com>
2149
2150         * doc/as.texinfo (Section): Document G and T flags to .section
2151         directive.  Document the extra arguments that the G flag
2152         requires.  Document the #tls flag.
2153
2154 2004-05-11  H.J. Lu  <hongjiu.lu@intel.com>
2155
2156         * subsegs.c (section_symbol): Create a new section symbol if
2157         the existing one doesn't match.
2158         * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
2159
2160 2004-05-07  Richard Sandiford  <rsandifo@redhat.com>
2161
2162         * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
2163         to cope with VR4181A errata MD(1) and MD(4).
2164
2165 2004-05-07  Brian Ford  <ford@vss.fsi.com>
2166
2167         * NEWS: Mention .secrel32 for pe[i]-i386.
2168
2169 2004-05-07  Alexandre Oliva  <aoliva@redhat.com>
2170
2171         * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
2172         (HANDLE_ALIGN): New.
2173
2174 2004-05-06  Daniel Jacobowitz  <dan@debian.org>
2175
2176         * Makefile.am (DIST_SUBDIRS): Define.
2177         * aclocal.m4: Regenerate with automake 1.8.4.
2178         * Makefile.in: Likewise.
2179         * doc/Makefile.in: Likewise.
2180
2181 2004-05-06  David Mosberger-Tang  <davidm@hpl.hp.com>
2182
2183         * config/tc-ia64.c (dot_serialize): Declare.
2184         (dot_serialize): New function.
2185         (md_pseudo_table): Add ".serialize.data" and
2186         ".serialize.instruction" directives.
2187
2188 2004-05-06  Nick Clifton  <nickc@redhat.com>
2189
2190         * messages (as_internal_value_out_of_range): Print a message about
2191         a value being out of range.  Be consistent about whether the
2192         values are printed in decimal or hexadecimal.
2193         (as_warn_value_out_of_range): Generate a warning message about an
2194         out of range value.
2195         (as_bad_value_out_of_range): Generate an error message about an
2196         out of range value.
2197         * as.h: Prototype the new functions.
2198         * config/tc-alpha.c (insert_operand): Use new function.
2199         * config/tc-arc.c (arc_insert_operand): Likewise.
2200         * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
2201         * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
2202         * config/tc-ppc.c (ppc_insert_operand): Likewise.
2203         * config/tc-s390.c (s390_insert_operand): Likewise.
2204         * config/tc-v850.c (v850_insert_operand): Likewise.
2205
2206 2004-05-05  Alexandre Oliva  <aoliva@redhat.com>
2207
2208         * configure.in: Set em=linux for frv-*-*linux*.
2209         * configure: Rebuilt.
2210         * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
2211         (frv_md_fdpic_enabled): New.
2212         * config/tc-frv.c (frv_md_fdpic_enabled): New.
2213         (DEFAULT_FDPIC): New.
2214         (frv_flags): Use DEFAULT_FDPIC.
2215         (frv_pic_flag): Likewise.
2216         (OPTION_NOPIC): New.
2217         (md_longopts): Add -mnopic.
2218         (md_parse_option): Handle it.
2219         (md_show_usage): Add -mfdpic and -mnopic.
2220
2221 2004-05-05  Peter Barada  <peter@the-baradas.com>
2222
2223         * config/tc-m68k.c: Add find_cf_chip to print list of valid
2224         chips for invalid coldfire instructions, rename selectors
2225         for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
2226         add current_chip to track which chip is referred to(including save/restore),
2227         use current_chip to select control registers, not current_arch.
2228         (md_show_usage): Add new chips.
2229         * doc/c-m68k.texi: Document new command line switches.
2230
2231 2004-05-05  Jakub Jelinek  <jakub@redhat.com>
2232
2233         * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
2234         (NOP_OPCODE): Define.
2235         (s390_align_code): Remove prototype.
2236         * tc-s390.c (s390_align_code): Remove.
2237
2238 2004-05-04  H.J. Lu  <hongjiu.lu@intel.com>
2239
2240         * config/tc-ia64.c (make_unw_section_name): Removed.
2241         (start_unwind_section): New function.
2242         (generate_unwind_image): Take const segT instead of const
2243         char *.
2244         (dot_handlerdata): Adjusted.
2245         (dot_endp): Likewise.
2246
2247 2004-05-02  H.J. Lu  <hongjiu.lu@intel.com>
2248
2249         * config/obj-elf.c (obj_elf_change_section): Allow the
2250         ".note.GNU-stack" section has SHF_EXECINSTR.
2251
2252 2004-05-02  H.J. Lu  <hongjiu.lu@intel.com>
2253
2254         * config/obj-elf.c (get_section): Return bfd_boolean.
2255         (obj_elf_change_section): Call bfd_get_section_by_name_if
2256         instead of bfd_map_over_sections.
2257
2258 2004-04-30  H.J. Lu  <hongjiu.lu@intel.com>
2259
2260         * config/obj-elf.c (get_section): New function.
2261         (obj_elf_change_section): Support multiple sections with same
2262         name.
2263
2264 2004-04-30  Nick Clifton  <nickc@redhat.com>
2265
2266         * config/tc-arm.c (create_register_alias): Fix typo checking for
2267         case sensitive register aliases.
2268         (co_proc_number): Use error message string in all_reg_maps[]
2269         array.
2270         (cp_reg_required_here): Likewise.
2271         (fp_reg_required_here): Likewise.
2272
2273 2004-04-29  Brian Ford  <ford@vss.fsi.com>
2274
2275         * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
2276
2277 2004-04-28  Chris Demetriou  <cgd@broadcom.com>
2278
2279         * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
2280         (load_address, macro, mips_ip, md_parse_option)
2281         (mips_force_relocation, mips_validate_fix, md_apply_fix3)
2282         (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
2283         embedded-PIC handling, and update comments.
2284         (SWITCH_TABLE): Remove.
2285         * config/tc-mips.h (DIFF_EXPR_OK): Delete.
2286         (enum mips_pic_level): Remove EMBEDDED_PIC.
2287         (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
2288         (TC_FORCE_RELOCATION): Update comment.
2289         * ecoff.c (ecoff_build_lineno): Add comment about some code that
2290         might be safe to remove now that MIPS embedded-PIC is gone.
2291
2292 2004-04-28  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
2293
2294         * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
2295         call to obj_set_subsection_attributes.
2296         (obj_som_init_stab_section): Likewise.
2297         * config/tc-hppa.c (default_subspace_dict): Add comdat field.
2298         (pa_def_subspaces): Provide comdat default.
2299         (pa_subspace): Handle new "comdat" parameter.  Set SEC_LINK_ONCE and
2300         not SEC_IS_COMMON if section is comdat, common or dup_common.  Update
2301         calls to create_new_subspace and update_subspace to pass comdat flag.
2302         (create_new_subspace, update_subspace): Add new comdat argument.  Use
2303         it in calls to obj_set_subsection_attributes.
2304         * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
2305         and use of comdat, common and dup_comm parameters.
2306
2307 2004-04-26  H.J. Lu  <hongjiu.lu@intel.com>
2308
2309         * config/obj-elf.c (obj_elf_change_section): Check if the old
2310         group name is NULL before comparison.
2311
2312 2004-04-23  Chris Demetriou  <cgd@broadcom.com>
2313
2314         * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
2315
2316 2004-04-23  Thiemo Seufer  <seufer@csv.ica.uni-stuttgart.de>
2317
2318         * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
2319
2320 2004-04-23  Chris Demetriou  <cgd@broadcom.com>
2321
2322         * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
2323         (OPTION_MEMBEDDED_PIC): Remove.
2324         (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
2325         (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
2326         (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
2327         (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
2328         (OPTION_ELF_BASE): Renumber.
2329         (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
2330         (md_show_usage): Remove mention of -membedded-pic.
2331         * doc/as.texinfo: Remove mention of -membedded-pic.
2332
2333 2004-04-23  Thiemo Seufer  <seufer@csv.ica.uni-stuttgart.de>
2334
2335         * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
2336         * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
2337         a.out support.
2338         (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
2339         s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
2340         USE_GLOBAL_POINTER_OPT.
2341
2342 2004-04-22  Thiemo Seufer  <seufer@csv.ica.uni-stuttgart.de>
2343
2344         * config/tc-mips.c (macro): One more use of load_delay_nop.
2345
2346 2004-04-22  Atsushi Nemoto  <anemo@mba.ocn.ne.jp>
2347
2348         * config/tc-mips.c (load_delay_nop): New function.
2349         (load_address, macro): Use load_delay_nop() to build a nop
2350         which can be omitted with gpr_interlocks.
2351
2352 2004-04-22  Thiemo Seufer  <seufer@csv.ica.uni-stuttgart.de>
2353
2354         * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
2355         cop_interlocks): Remove superfluous CPU entries.
2356
2357 2004-04-22  Paul Brook  <paul@codesourcery.com>
2358
2359         * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
2360
2361 2004-04-22  Peter Barada <peter@the-baradas.com>
2362
2363         * NEWS: Added support for EMAC instructions and MAC/EMAC
2364         Motorola syntax.
2365         * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
2366         scale factor tokens, trailing_ampersand to mark mask addressing
2367         for MAC/EMAC instructions.
2368         * config/m68k-parse.y: Add options_ampersand clause, '<<',
2369         '>>'.
2370         (yylex): Handle '>', '<', and '&' following '+'.
2371         * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
2372         architectures in archs[].
2373         (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
2374         for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
2375         respectively.
2376         (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
2377         (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
2378         cases.
2379         Add EMAC operands to init_table[].
2380
2381 2004-04-22  Bruno De Bus <bdebus@elis.ugent.be>
2382
2383         * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
2384         state.
2385         (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
2386         * config/tc-arm.c (enum mstate): Delete from here.
2387         (mapping_state): Remove the static mapstate variable and instead
2388         store the state in the segment.  This allows a per-section mapping
2389         state.  Handle and ignore MAP_UNDEFINED states.
2390         (arm_elf_change_section): Get the current mapping state from the
2391         new section.
2392         (s_ltorg): Set the mapping state to MAP_DATA.
2393         (arm_cleanup): Use arm_elf_change_section to get the mapping state
2394         for each pool as it is emitted.
2395
2396 2004-04-22  Nick Clifton  <nickc@redhat.com>
2397
2398         * config/tc-arm.h: Formatting tidy ups.
2399
2400 2004-04-20  Chris Demetriou  <cgd@broadcom.com>
2401
2402         * NEWS: Note that MIPS -membedded-pic option is deprecated.
2403
2404 2004-04-20  DJ Delorie  <dj@redhat.com>
2405
2406         * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
2407         * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
2408         [TE_PE] (O_secrel): Define.
2409         [TE_PE] (x86_pe_cons_fix_new): New.
2410         [TE_PE] (pe_directive_secrel): Likewise.
2411         (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
2412
2413 2004-04-19  Eric Christopher  <echristo@redhat.com>
2414
2415         * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
2416         of previous patch for fix in gcc.
2417
2418 2004-04-19  Jakub Jelinek  <jakub@redhat.com>
2419
2420         * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
2421
2422 2004-04-19  Nathan Sidwell  <nathan@codesourcery.com>
2423
2424         * read.c (do_align): Call md_flush_pending_output, if defined.
2425
2426 2004-04-16  Alan Modra  <amodra@bigpond.net.au>
2427
2428         * expr.c (operand): Correct checks for ++ and --.
2429
2430 2004-04-14  H.J. Lu  <hongjiu.lu@intel.com>
2431
2432         * config/tc-generic.c: Add some comments.
2433
2434 2004-04-14  Richard Sandiford  <rsandifo@redhat.com>
2435
2436         * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
2437         -{no-}mfix-vr4122-bugs.
2438         * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
2439         (append_insn, mips_emit_delays): Update accordingly.
2440         (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
2441         (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
2442         (md_parse_option): Update after above changes.
2443         (md_show_usage): Add -mfix-vr4120.
2444
2445 2004-04-13  Bob Wilson  <bob.wilson@acm.org>
2446
2447         * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
2448         of .section directive; add a reference to the ELF .subsection
2449         directive.
2450
2451 2004-04-13  Kazuhiro Inaoka  <inaoka.kazuhiro@renesas.com>
2452
2453         * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
2454         in parallel.
2455
2456 2004-04-11  Thiemo Seufer  <seufer@csv.ica.uni-stuttgart.de>
2457
2458         * Makefile.am: Remove mips from aout targets.
2459         * Makefile.in: Regenerate.
2460         * configure.in: Remove mips-dec-bsd* target.
2461         * configure: Regenerate.
2462
2463 2004-04-07  Alan Modra  <amodra@bigpond.net.au>
2464
2465         PR 96
2466         * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
2467         to struct map_bfd.  Adjust MAP macro, and define MAP32, MAP64.
2468         Update "mapping".  Restrict some @ modifiers to 32 bit.
2469
2470 2004-04-01  Asgari Jinia  <asgarij@kpitcummins.com>
2471             Dhananjay Deshpande <dhananjayd@kpitcummins.com>
2472
2473         * config/tc-sh.c (dont_adjust_reloc_32): New variable.
2474         (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
2475         dont_adjust_reloc_32 is set.
2476         (md_longopts): Add option -renesas.
2477         (md_parse_option, md_show_usage): Likewise.
2478         * doc/c-sh.texi: Likewise.
2479
2480 2004-04-01  Dave Korn  <dk@artimi.com>
2481
2482         * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
2483         hi16 and lo16 fixS structs.
2484         (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
2485         exactly the same way as for RELOC_DLX_REL16.
2486         (machine_ip): properly respect LO flag in the_insn and
2487         output RELOC_DLX_LO16 rather than RELOC_DLX_16.
2488         (md_apply_fix3): apply RELOC_DLX_LO16.
2489
2490 2004-03-30  Stan Shebs  <shebs@apple.com>
2491
2492         Remove long-obsolete MPW support.
2493         * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
2494         * configure.in: Remove mention of ppc-*-mpw* config.
2495         * configure.in: Likewise.
2496
2497 2004-03-30  Nick Clifton  <nickc@redhat.com>
2498
2499         * config/tc-arm.c (meabi_flags): Make its use conditional upon
2500         OBJ_ELF being defined.
2501
2502 2004-03-27  Alan Modra  <amodra@bigpond.net.au>
2503
2504         * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
2505
2506 2004-03-23  Paul Brook  <paul@codesourcery.com>
2507
2508         * config/tc-arm.c (meabi_flags): New variable.
2509         (arm_parse_eabi): New function.
2510         (md_begin): Set flags for EABI v3.
2511         (arm_eabis): Add.
2512         (arm_long_opts): Add meabi.
2513         * doc/as.texinf <ARM>: Document -meabi.
2514         * doc/c-arm.texi: Ditto.
2515
2516 2004-03-22  Bob Wilson  <bob.wilson@acm.org>
2517
2518         * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
2519         tables even when use_literal_section flag is not set.
2520
2521 2004-03-22  Alan Modra  <amodra@bigpond.net.au>
2522
2523         * config/tc-sh.c: Remove trailing whitespace.
2524
2525 2004-03-22  Hans-Peter Nilsson  <hp@axis.com>
2526
2527         * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
2528         --mul-bug-abort and the default behavior.
2529         * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
2530         (err_for_dangerous_mul_placement): New variable.
2531         (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
2532         macros.
2533         (md_cris_relax_table): Have placeholder for STATE_MUL.
2534         (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
2535         (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
2536         case doing nothing.
2537         (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
2538         STATE_BYTE)>: Ditto.
2539         (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
2540         alignment and position of this frag, emit error message if
2541         suspicious.
2542         (md_assemble): For a multiply insn and when checking it,
2543         transform the current frag into a special frag for that purpose.
2544         (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
2545         OPTION_MULBUG_ABORT_ON>: Handle new options.
2546
2547 2004-03-19  Bob Wilson  <bob.wilson@acm.org>
2548
2549         * config/tc-xtensa.c (mark_literal_frags): New function.
2550         (xtensa_move_literals): Call mark_literal_frags for all literal
2551         segments, including init and fini literal segments.
2552         (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
2553
2554 2004-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
2555
2556         * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
2557         (pa_comm): Set BSF_OBJECT in symbol flags.
2558
2559 2004-03-19  Alan Modra  <amodra@bigpond.net.au>
2560
2561         * Makefile.am: Run "make dep-am".
2562         * Makefile.in: Regenerate.
2563         * doc/Makefile.in: Regenerate.
2564         * config.in: Regenerate.
2565         * po/gas.pot: Regenerate.
2566
2567 2004-03-18  Nathan Sidwell  <nathan@codesourcery.com>
2568
2569         * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2570         (demand_empty_rest_of_line): Issue an error here.
2571         (ignore_rest_of_line): Silently skip to end.
2572         (demand_copy_string): Issue an error, not warning.
2573         (equals): Likewise.
2574         * config/obj-elf.c (obj_elf_section_name): Likewise.
2575         (obj_elf_section): Likewise.
2576         * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2577         (arc_extinst): Likewise.
2578         * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2579         (dot_spill): Likewise.
2580         (dot_unwabi): Likewise.
2581         (dot_prologue): Likewise.
2582
2583         * expr.c (operand): Reject ++ and --.
2584         (operator): Likewise.
2585
2586 2004-03-17  Kaz Kojima  <kkojima@rr.iij4u.or.jp>
2587
2588         * config/tc-sh.c: Include dw2gencfi.h.
2589         (sh_cfi_frame_initial_instructions): New function.
2590         (sh_regname_to_dw2regnum): Likewise.
2591         * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2592         file.
2593         (TARGET_USE_CFIPOP): Define.
2594         (tc_cfi_frame_initial_instructions): Likewise.
2595         (tc_regname_to_dw2regnum): Likewise.
2596         (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2597         * Makefile.am: Update dependencies.
2598         * Makefile.in: Regenerate.
2599
2600 2004-03-17  Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2601
2602         * configure.in: Switch sh-*-rtems* to ELF.  Add sh-*-rtemscoff*.
2603         * configure: Regenerate.
2604
2605 2004-03-12  Bob Wilson  <bob.wilson@acm.org>
2606
2607         * read.c (s_leb128): Call md_flush_pending_output.
2608
2609 2004-03-12  Michal Ludvig  <mludvig@suse.cz>
2610
2611         * config/tc-i386.c (output_insn): Handle PadLock instructions.
2612         * config/tc-i386.h (CpuPadLock): New define.
2613         (CpuUnknownFlags): Added CpuPadLock.
2614
2615 2004-03-07  Andreas Schwab  <schwab@suse.de>
2616
2617         * doc/c-hppa.texi (HPPA Directives): Fix typo.
2618
2619 2004-03-07  Richard Henderson  <rth@redhat.com>
2620
2621         * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2622         (cfi_finish): Likewise for fde.
2623
2624 2004-03-05  H.J. Lu  <hongjiu.lu@intel.com>
2625
2626         * config/tc-ia64.c (md_assemble): Properly handle NULL
2627         align_frag.
2628         (ia64_handle_align): Don't abort if failed to add a stop bit.
2629
2630 2004-03-04  H.J. Lu  <hongjiu.lu@intel.com>
2631
2632         * Makefile.in: Regenerated.
2633         * aclocal.m4: Likewise.
2634         * configure: Likewise.
2635         * doc/Makefile.in: Likewise.
2636
2637 2004-03-03  H.J. Lu  <hongjiu.lu@intel.com>
2638
2639         * config/tc-ia64.c (dot_align): New.
2640         (ia64_do_align): Make it static.
2641         (md_pseudo_table): Use "dot_align" for "align".
2642         (ia64_md_do_align): Don't set align_frag here.
2643         (ia64_handle_align): Add a stop bit to the previous bundle if
2644         needed.
2645
2646         * config/tc-ia64.h (ia64_do_align): Removed.
2647
2648 2003-03-03  Andrew Stubbs  <andrew.stubbs@superh.com>
2649
2650         * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2651         -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2652         (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2653         the most general type or the user specifically requested it.
2654         (md_assemble): Add a new error message for when an instruction
2655         is understood, but is not allowed due to an -isa option.
2656
2657 2004-03-02  H.J. Lu  <hongjiu.lu@intel.com>
2658
2659         * config/tc-ia64.c (align_frag): New.
2660         (md_assemble): Set the tc_frag_data field in align_frag for
2661         IA64_OPCODE_FIRST instructions.
2662         (ia64_md_do_align): Set align_frag.
2663         (ia64_handle_align): Add a stop bit if needed.
2664
2665         * config/tc-ia64.h (TC_FRAG_TYPE): New.
2666         (TC_FRAG_INIT): New.
2667
2668 2004-03-01  Richard Sandiford  <rsandifo@redhat.com>
2669
2670         * config/tc-frv.c (fr400_audio): New variable.
2671         (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2672         (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2673         (target_implements_insn_p): New function.
2674         (md_assemble): Report an error if the processor doesn't implement
2675         the instruction.
2676
2677 2004-02-27  Kazuhiro Inaoka  <inaoka.kazuhiro@renesas.com>
2678
2679         * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2680         (md_parse_option): Ditto.
2681         (OPTION_NO_SPECIAL_M32R): Added.
2682         (md_show_usage): Document it.
2683         (enable_speial_m32r): Changed a default value from 0 to 1.
2684         * doc/c-m32r.texi: Document the -no-bitinst option.
2685
2686 2004-02-27  Nick Clifton  <nickc@redhat.com>
2687
2688         * config/tc-sh.c (get_operand): Revert previous delta.
2689         (tc_gen_reloc): Check for an unknown reloc type before processing
2690         the addend.
2691
2692 2004-02-27  Hannes Reinecke  <hare@suse.de>
2693
2694         * config/tc-s390.c (s390_insn): Correct range check for opcode in
2695         .insn pseudo operation.
2696
2697 2004-02-27  Anil Paranjpe  <anilp1@kpitcummins.com>
2698
2699         * config/tc-sh.c (get_operand):  In case of #Imm, check has been
2700         added for wrong syntax.
2701
2702 2004-02-26  Eric Christopher  <echristo@redhat.com>
2703
2704         * config/tc-mips.c (mips_dwarf2_addr_size): New.
2705         * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2706
2707 2004-02-26  Andrew Stubbs <andrew.stubbs@superh.com>
2708
2709         * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2710         nibble types to assembler.
2711
2712 2004-02-25  Fred Fish  <fnf@redhat.com>
2713
2714         * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2715
2716 2004-02-20  James E Wilson  <wilson@specifixinc.com>
2717
2718         * config/tc-ia64.c (slot_index): New arg before_relax.  Use instead of
2719         finalize_syms.
2720         (fixup_unw_records): New arg before_relax.  Pass to slot_index.
2721         (ia64_estimate_size_before_relax): New.
2722         (ia64_convert_frag): Pass 0 to fixup_unw_records.  Add comment.
2723         (generate_unwind_image): Pass 1 to fixup_unw_records.
2724         * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2725         (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2726
2727 2004-02-19  Jakub Jelinek  <jakub@redhat.com>
2728
2729         * stabs.c (generate_asm_file): Avoid warning about use of
2730         uninitialized variable.
2731
2732 2004-02-18  David Mosberger  <davidm@hpl.hp.com>
2733
2734         * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2735         body, and endp, allow unwind records which do not have a "t"
2736         (time/instruction) field.
2737
2738 2004-02-17  Petko Manolov  <petkan@nucleusys.com>
2739
2740         * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2741         (do_mav_dspsc_2): Likewise.
2742         Fix accumulator registers move opcodes.
2743
2744 2004-02-13  Hannes Reinecke  <hare@suse.de>
2745             Jakub Jelinek  <jakub@redhat.com>
2746
2747         * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2748
2749 2004-02-10  Steve Ellcey  <sje@cup.hp.com>
2750
2751         * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2752         (tc_frob_symbol): New macro definition.
2753         * config/tc-ia64.c (ia64_frob_symbol): New routine.
2754
2755 2004-02-09  Daniel Jacobowitz  <drow@mvista.com>
2756
2757         * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2758         read-only.
2759
2760 2004-02-09  Nathan Sidwell  <nathan@codesourcery.com>
2761
2762         * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2763         (TC_CASE_SENSITIVE): ... this.
2764         * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2765         * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2766
2767 2004-02-06  James E Wilson  <wilson@specifixinc.com>
2768
2769         * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2770         (generate_unwind_image): Re-add it here.
2771
2772 2004-02-06  Nathan Sidwell  <nathan@codesourcery.com>
2773
2774         * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2775         * read.h (SKIP_WHITESPACE): Turn into an expression.
2776         * read.c (read_a_source_file): A pseudo is removed by having a
2777         NULL handler.
2778
2779 2004-02-05  James E Wilson  <wilson@specifixinc.com>
2780
2781         * config/tc-ia64.c (output_endp): New.
2782         (count_bits): Delete.
2783         (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2784         endp unwind records.
2785         (fixup_unw_records): Handle endp unwind records.  Delete code for
2786         shortening prologue regions not followed by a body record.
2787         (dot_endp): Call add_unwind_entry to emit endp unwind record.
2788         * config/tc-ia64.h (unw_record_type): Add endp.
2789
2790 2004-02-03  James E Wilson  <wilson@specifixinc.com>
2791
2792         * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2793         fill padding bytes with zeroes.
2794         (emit_one_bundle): New locals last_ptr, end_ptr.  Rewrite code that
2795         sets unwind_record slot_number and slot_frag fields.
2796
2797 2004-02-02  Maciej W. Rozycki  <macro@ds2.pg.gda.pl>
2798
2799         * config/tc-mips.c (add_got_offset_hilo): New function.
2800         (macro): Use load_register() and add_got_offset_hilo() to load
2801         constants instead of hardcoding code sequences throughout.
2802
2803 2004-01-28  H.J. Lu  <hongjiu.lu@intel.com>
2804
2805         * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2806
2807 2004-01-26  Bernardo Innocenti  <bernie@develer.com>
2808
2809         * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2810         like m68k-elf.
2811         * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2812         of hard-coded test for TARGET_OS=elf.
2813
2814 2004-01-24  Chris Demetriou  <cgd@broadcom.com>
2815
2816         * config/tc-mips.c (hilo_interlocks): Change definition
2817         so that MIPS32, MIPS64 and later ISAs are included, along with
2818         the already-included machines.  Update comments.
2819
2820 2004-01-23  Daniel Jacobowitz  <drow@mvista.com>
2821
2822         * config/tc-arm.c (tc_gen_reloc): Improve error message for
2823         undefined local labels.
2824
2825 2004-01-23  Richard Sandiford  <rsandifo@redhat.com>
2826
2827         * config/tc-mips.c (load_address, macro): Update comments about
2828         NewABI GP relaxation.
2829
2830 2004-01-23  Richard Sandiford  <rsandifo@redhat.com>
2831
2832         * config/tc-mips.c (macro_build): Remove place and counter arguments.
2833         (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2834         (mips16_macro_build, macro_build_jalr): Remove counter argument.
2835         (set_at, load_register, load_address, move_register): Likewise.
2836         (load_got_offset, add_got_offset): Likewise.
2837         Update all calls and tidy accordingly.
2838
2839 2004-01-23  Richard Sandiford  <rsandifo@redhat.com>
2840
2841         * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2842         (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2843         (RELAX_USE_SECOND): Bump to 0x10000.
2844         (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2845         (mips_macro_warning): New variable.
2846         (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2847         (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2848         (relax_close_frag): Set mips_macro_warning.first_frag.  Adjust use
2849         of RELAX_ENCODE.
2850         (append_insn): Update mips_macro_warning.sizes.
2851         (macro_start, macro_warning, macro_end): New functions.
2852         (macro_build): Don't emit warnings here.
2853         (macro_build_lui, md_estimate_size_before_relax): ...or here.
2854         (md_convert_frag): Check for cases where one macro alternative
2855         needs a warning and the other doesn't.  Emit a warning if the
2856         longer sequence was chosen.
2857
2858 2004-01-23  Richard Sandiford  <rsandifo@redhat.com>
2859
2860         * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2861         * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2862         the first sequence, the size of the second sequence, and a flag
2863         that says whether we should warn.
2864         (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2865         (RELAX_FIRST, RELAX_SECOND): New.
2866         (mips_relax): New variable.
2867         (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2868         (append_insn): Remove "place" argument.  Use mips_relax.sequence
2869         rather than "place" to check whether we're expanding the second
2870         alternative of a relaxable macro.  Remove redundant check for
2871         branch relaxation.  If generating a normal insn, and there
2872         is not enough room in the current frag, call relax_close_frag()
2873         to close it.  Update mips_relax.sizes[].  Emit fixups for the
2874         second version of a relaxable macro.  Record the first relaxable
2875         fixup in mips_relax.  Remove tc_gen_reloc workaround.
2876         (macro_build): Remove all uses of "place".  Use mips_relax.sequence
2877         in the same way as in append_insn.
2878         (mips16_macro_build): Remove "place" argument.
2879         (macro_build_lui): As for macro_build.  Don't drop the add_symbol
2880         when generating the second version of a relaxable macro.
2881         (load_got_offset, add_got_offset): New functions.
2882         (load_address, macro): Use new relaxation machinery.  Remove
2883         tc_gen_reloc workarounds.
2884         (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2885         version of a relaxable macro is needed.  Return -RELAX_SECOND if the
2886         first version is needed.
2887         (tc_gen_reloc): Remove relaxation handling.
2888         (md_convert_frag): Go through the fixups for a relaxable macro and
2889         mark those that belong to the unneeded alternative as done.  If the
2890         second alternative is needed, adjust the fixup addresses to account
2891         for the deleted first alternative.
2892
2893 2004-01-23  Richard Sandiford  <rsandifo@redhat.com>
2894
2895         * frags.h (frag_room): Declare.
2896         * frags.c (frag_room): New function.
2897         * doc/internals.texi: Document it.
2898
2899 2004-01-22  Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2900
2901         * config/tc-mips.c (append_insn): Don't do r3900 interlock
2902         optimization for -mtune=r3900, as this will break on other CPUs.
2903
2904 2004-01-11  Tom Rix   <tcrix@worldnet.att.net>
2905
2906         * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2907         be relaxed, use fixup.
2908         (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2909
2910 2004-01-19  Jakub Jelinek  <jakub@redhat.com>
2911
2912         * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2913         precision operands.
2914
2915 2004-01-14  Maciej W. Rozycki  <macro@ds2.pg.gda.pl>
2916
2917         * config/tc-mips.c (append_insn): Properly detect variant frags
2918         that preclude swapping of relaxed branches.  Correctly swap
2919         instructions between frags when dealing with relaxed branches.
2920
2921 2004-01-14  Maciej W. Rozycki  <macro@ds2.pg.gda.pl>
2922
2923         * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2924         throughout.
2925         * aclocal.m4: Regenerate.
2926         * configure: Regenerate.
2927
2928 2004-01-12  Anil Paranjpe  <anilp1@KPITCummins.com>
2929
2930         * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2931         manipulation insns.
2932
2933 2004-01-12  Richard Sandiford  <rsandifo@redhat.com>
2934
2935         * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2936         reloc, reserve space for the delay slot as well as the jalr itself.
2937
2938 2004-01-09  Paul Brook  <paul@codesourcery.com>
2939
2940         * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2941         (do_vfp_sp2_from_reg2): New function.
2942         (insns): Use them.
2943         (do_vfp_dp_from_reg2): Check return values properly.
2944
2945 2004-01-08  Ian Lance Taylor  <ian@wasabisystems.com>
2946
2947         * config/tc-mips.c (warn_nops): Remove static variable.
2948         (macro): Remove test of warn_nops.
2949         (md_shortops): Remove 'n'.
2950         (md_parse_option): Remove 'n' case.
2951         (md_show_usage): Remove -n.
2952         * doc/as.texinfo (Overview): Remove MIPS -n option.
2953         * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2954         * NEWS: Mention removal of MIPS -n option.
2955
2956         * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2957         (cop_interlocks): Check ISA level.
2958         (cop_mem_interlocks): Define.
2959         (reg_needs_delay): Check cop_interlocks rather than
2960         ISA_HAS_COPROC_DELAYS.
2961         (append_insn): Likewise.  Use cop_mem_interlocks rather than
2962         directly checking mips_opts.isa.
2963         (mips_emit_delays): Likewise.
2964
2965 2004-01-07  H.J. Lu  <hongjiu.lu@intel.com>
2966
2967         * config/tc-ia64.c (unwind): Move next_slot_number and
2968         next_slot_frag to ...
2969         (unw_rec_list): Here.
2970         (free_list_records): Removed.
2971         (output_unw_records): Likewise.
2972         (generate_unwind_image): Make it void.
2973         (alloc_record): Initialize next_slot_number and next_slot_frag.
2974         (slot_index): Take .org, .space and .align into account.
2975         (fixup_unw_records): Don't set slot_number to 0. Use
2976         list->next_slot_number and list->next_slot_frag instead of
2977         unwind.next_slot_number and unwind.next_slot_frag.
2978         (ia64_convert_frag): New.
2979         (generate_unwind_image): Generate a rs_machine_dependent frag
2980         for unwind record.
2981         (emit_one_bundle): Use list->next_slot_number and
2982         list->next_slot_frag instead of unwind.next_slot_number and
2983         unwind.next_slot_frag.
2984
2985         * config/tc-ia64.h (md_convert_frag): Defined as
2986         ia64_convert_frag.
2987         (md_estimate_size_before_relax): Defined as (f)->fr_var.
2988
2989 2004-01-06  Alexandre Oliva  <aoliva@redhat.com>
2990
2991         2003-12-19  Alexandre Oliva  <aoliva@redhat.com>
2992         * config/tc-frv.h (md_apply_fix3): Don't define.
2993         * config/tc-frv.c (md_apply_fix3): New.  Shift/truncate %hi/%lo
2994         operands.
2995         * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2996         2003-10-07  Alexandre Oliva  <aoliva@redhat.com>
2997         * config/tc-frv.c (line_separator_chars): Add `!'.
2998         2003-09-19  Alexandre Oliva  <aoliva@redhat.com>
2999         * config/tc-frv.c (md_assemble): Clear insn upfront.
3000         2003-09-18  Alexandre Oliva  <aoliva@redhat.com>
3001         * config/tc-frv.c (OPTION_FDPIC): New macro.
3002         (md_longopts): Add mfdpic.
3003         (md_parse_option): Handle it.
3004         2003-08-04  Alexandre Oliva  <aoliva@redhat.com>
3005         * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
3006         FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
3007         (frv_pic_ptr): Parse funcdesc.
3008
3009 2004-01-05  Maciej W. Rozycki  <macro@ds2.pg.gda.pl>
3010
3011         * doc/as.texinfo: Let texi2pod parse asconfig.texi and
3012         gasver.texi.  Remove duplicate symbol definitions for texi2pod.
3013
3014 2004-01-05  Maciej W. Rozycki  <macro@ds2.pg.gda.pl>
3015
3016         * Makefile.am (Makefile): Move the dependency on
3017         $(BFDDIR)/configure.in to...
3018         (CONFIG_STATUS_DEPENDENCIES): ... here.
3019         (AUTOMAKE_OPTIONS): Require automake 1.8.
3020         * Makefile.in: Regenerate.
3021         * doc/Makefile.am (BASEDIR, BFDDIR): Define.
3022         (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
3023         $(BFDDIR)/configure.in here as well.
3024         * doc/Makefile.in: Regenerate.
3025
3026 2004-01-05  Maciej W. Rozycki  <macro@ds2.pg.gda.pl>
3027
3028         * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
3029         * Makefile.in: Regenerate.
3030         * aclocal.m4: Regenerate.
3031         * doc/Makefile.am (install, install-info): Remove.
3032         (install-data-local): A new hook for install-info.
3033         (AUTOMAKE_OPTIONS): Require automake 1.8.
3034         * doc/Makefile.in: Regenerate.
3035
3036 2004-01-02  Nutan Singh <nutan@kpitcummins.com>
3037
3038         * doc/c-sh.texi: Update description about floating point behavior
3039         of SH family.
3040
3041 2004-01-02  Bernardo Innocenti  <bernie@develer.com>
3042
3043         * configure.in: Add m68k-uClinux target.
3044         * configure: Regenerate.
3045
3046 For older changes see ChangeLog-0203
3047 \f
3048 Local Variables:
3049 mode: change-log
3050 left-margin: 8
3051 fill-column: 74
3052 version-control: never
3053 End: