gas/NEWS

   1 -*- text -*-
   2
   3 Changes in 2.26:
   4
   5 * Symbol and label names can now be enclosed in double quotes (") which allows
   6   them to contain characters that are not part of valid symbol names in high
   7   level languages.
   8
   9 * Added the correctly spelled -march=armv6kz, for ARMv6KZ support.  The
  10   previous spelling, -march=armv6zk, is still accepted.
  11
  12 * Support for the ARMv8.1 architecture has been added to the Aarch64 port.
  13   Support for the individual ARMv8.1 Adv.SIMD, LOR and PAN architecture
  14   extensions has also been added to the Aarch64 port.
  15
  16 * Support for the ARMv8.1 architecture has been added to the ARM port.  Support
  17   for the individual ARMv8.1 Adv.SIMD and PAN architecture extensions has also
  18   been added to the ARM port.
  19
  20 * Extend --compress-debug-sections option to support
  21   --compress-debug-sections=[none|zlib|zlib-gnu|zlib-gabi] for ELF
  22   targets.
  23
  24 * --compress-debug-sections is turned on for Linux/x86 by default.
  25
  26 Changes in 2.25:
  27
  28 * Add support for the AVR Tiny microcontrollers.
  29
  30 * Replace support for openrisc and or32 with support for or1k.
  31
  32 * Enhanced the ARM port to accept the assembler output from the CodeComposer
  33   Studio tool.  Support is enabled via the new command line option -mccs.
  34
  35 * Add support for the Andes NDS32.
  36
  37 Changes in 2.24:
  38
  39 * Add support for the Texas Instruments MSP430X processor.
  40
  41 * Add -gdwarf-sections command line option to enable per-code-section
  42   generation of DWARF .debug_line sections.
  43
  44 * Add support for Altera Nios II.
  45
  46 * Add support for the Imagination Technologies Meta processor.
  47
  48 * Add support for the v850e3v5.
  49
  50 * Remove assembler support for MIPS ECOFF targets.
  51
  52 Changes in 2.23:
  53
  54 * Add support for the 64-bit ARM architecture: AArch64.
  55
  56 * Add support for S12X processor.
  57
  58 * Add support for the VLE extension to the PowerPC architecture.
  59
  60 * Add support for the Freescale XGATE architecture.
  61
  62 * Add support for .bundle_align_mode, .bundle_lock, and .bundle_unlock
  63   directives.  These are currently available only for x86 and ARM targets.
  64
  65 * Add support for the Renesas RL78 architecture.
  66
  67 * Add support for the Adapteva EPIPHANY architecture.
  68
  69 * For x86, allow 'rep bsf', 'rep bsr', and 'rep ret' syntax.
  70
  71 Changes in 2.22:
  72
  73 * Add support for the Tilera TILEPro and TILE-Gx architectures.
  74
  75 Changes in 2.21:
  76
  77 * Gas no longer requires doubling of ampersands in macros.
  78
  79 * Add support for the TMS320C6000 (TI C6X) processor family.
  80
  81 * GAS now understands an extended syntax in the .section directive flags
  82   for COFF targets that allows the section's alignment to be specified.  This
  83   feature has also been backported to the 2.20 release series, starting with
  84   2.20.1.
  85
  86 * Add support for the Renesas RX processor.
  87
  88 * New command line option, --compress-debug-sections, which requests
  89   compression of DWARF debug information sections in the relocatable output
  90   file.  Compressed debug sections are supported by readelf, objdump, and
  91   gold, but not currently by Gnu ld.
  92
  93 Changes in 2.20:
  94
  95 * Added support for v850e2 and v850e2v3.
  96
  97 * GNU/Linux targets now supports "gnu_unique_object" as a value in the .type
  98   pseudo op.  It marks the symbol as being globally unique in the entire
  99   process.
 100
 101 * ARM assembler now supports .inst[.nw] pseudo-ops to insert opcodes specified
 102   in binary rather than text.
 103
 104 * Add support for common symbol alignment to PE formats.
 105
 106 * Add support for the new discriminator column in the DWARF line table,
 107   with a discriminator operand for the .loc directive.
 108
 109 * Add support for Sunplus score architecture.
 110
 111 * The .type pseudo-op now accepts a type of STT_GNU_IFUNC which can be used to
 112   indicate that if the symbol is the target of a relocation, its value should
 113   not be use.  Instead the function should be invoked and its result used as
 114   the value.
 115
 116 * Add support for Lattice Mico32 (lm32) architecture.
 117
 118 * Add support for Xilinx MicroBlaze architecture.
 119
 120 Changes in 2.19:
 121
 122 * New pseudo op .cfi_val_encoded_addr, to record constant addresses in unwind
 123   tables without runtime relocation.
 124
 125 * New command line option, -h-tick-hex, for sh, m32c, and h8/300 targets, which
 126   adds compatibility with H'00 style hex constants.
 127
 128 * New command line option, -msse-check=[none|error|warning], for x86
 129   targets.
 130
 131 * New sub-option added to the assembler's -a command line switch to
 132   generate a listing output.  The 'g' sub-option will insert into the listing
 133   various information about the assembly, such as assembler version, the
 134   command line options used, and a time stamp.
 135
 136 * New command line option -msse2avx for x86 target to encode SSE
 137   instructions with VEX prefix.
 138
 139 * Add Intel XSAVE, EPT, MOVBE, AES, PCLMUL, AVX/FMA support for x86 target.
 140
 141 * New command line options, -march=CPU[,+EXTENSION...], -mtune=CPU,
 142   -mmnemonic=[att|intel], -msyntax=[att|intel], -mindex-reg,
 143   -mnaked-reg and -mold-gcc, for x86 targets.
 144
 145 * Support for generating wide character strings has been added via the new
 146   pseudo ops: .string16, .string32 and .string64.
 147
 148 * Support for SSE5 has been added to the i386 port.
 149
 150 Changes in 2.18:
 151
 152 * The GAS sources are now released under the GPLv3.
 153
 154 * Support for the National Semiconductor CR16 target has been added.
 155
 156 * Added gas .reloc pseudo.  This is a low-level interface for creating
 157   relocations.
 158
 159 * Add support for x86_64 PE+ target.
 160
 161 * Add support for Score target.
 162
 163 Changes in 2.17:
 164
 165 * Support for the Infineon XC16X has been added by KPIT Cummins Infosystems.
 166
 167 * Support for ms2 architecture has been added.
 168
 169 * Support for the Z80 processor family has been added.
 170
 171 * Add support for the "@<file>" syntax to the command line, so that extra
 172   switches can be read from <file>.
 173
 174 * The SH target supports a new command line switch --enable-reg-prefix which,
 175   if enabled, will allow register names to be optionally prefixed with a $
 176   character.  This allows register names to be distinguished from label names.
 177
 178 * Macros with a variable number of arguments are now supported. See the
 179   documentation for how this works.
 180
 181 * Added --reduce-memory-overheads switch to reduce the size of the hash
 182   tables used, at the expense of longer assembly times, and
 183   --hash-size=<NUMBER> to set the size of the hash tables used by gas.
 184
 185 * Macro names and macro parameter names can now be any identifier that would
 186   also be legal as a symbol elsewhere. For macro parameter names, this is
 187   known to cause problems in certain sources when the respective target uses
 188   characters inconsistently, and thus macro parameter references may no longer
 189   be recognized as such (see the documentation for details).
 190
 191 * Support the .f_floating, .d_floating, .g_floating and .h_floating directives
 192   for the VAX target in order to be more compatible with the VAX MACRO
 193   assembler.
 194
 195 * New command line option -mtune=[itanium1|itanium2] for IA64 targets.
 196
 197 Changes in 2.16:
 198
 199 * Redefinition of macros now results in an error.
 200
 201 * New command line option -mhint.b=[ok|warning|error] for IA64 targets.
 202
 203 * New command line option -munwind-check=[warning|error] for IA64
 204   targets.
 205
 206 * The IA64 port now uses automatic dependency violation removal as its default
 207   mode.
 208
 209 * Port to MAXQ processor contributed by HCL Tech.
 210
 211 * Added support for generating unwind tables for ARM ELF targets.
 212
 213 * Add a -g command line option to generate debug information in the target's
 214   preferred debug format.
 215
 216 * Support for the crx-elf target added.
 217
 218 * Support for the sh-symbianelf target added.
 219
 220 * Added a pseudo-op (.secrel32) to generate 32 bit section relative relocations
 221   on pe[i]-i386; required for this target's DWARF 2 support.
 222
 223 * Support for Motorola MCF521x/5249/547x/548x added.
 224
 225 * Support for ColdFire EMAC instructions added and Motorola syntax for MAC/EMAC
 226   instrucitons.
 227
 228 * New command line option -mno-shared for MIPS ELF targets.
 229
 230 * New command line option --alternate and pseudo-ops .altmacro and .noaltmacro
 231   added to enter (and leave) alternate macro syntax mode.
 232
 233 Changes in 2.15:
 234
 235 * The MIPS -membedded-pic option (Embedded-PIC code generation) is
 236   deprecated and will be removed in a future release.
 237
 238 * Added PIC m32r Linux (ELF) and support to M32R assembler.
 239
 240 * Added support for ARM V6.
 241
 242 * Added support for sh4a and variants.
 243
 244 * Support for Renesas M32R2 added.
 245
 246 * Limited support for Mapping Symbols as specified in the ARM ELF
 247   specification has been added to the arm assembler.
 248
 249 * On ARM architectures, added a new gas directive ".unreq" that undoes
 250   definitions created by ".req".
 251
 252 * Support for Motorola ColdFire MCF528x added.
 253
 254 * Added --gstabs+ switch to enable the generation of STABS debug format
 255   information with GNU extensions.
 256
 257 * Added support for MIPS64 Release 2.
 258
 259 * Added support for v850e1.
 260
 261 * Added -n switch for x86 assembler.  By default, x86 GAS replaces
 262   multiple nop instructions used for alignment within code sections
 263   with multi-byte nop instructions such as leal 0(%esi,1),%esi.  This
 264   switch disables the optimization.
 265
 266 * Removed -n option from MIPS assembler.  It was not useful, and confused the
 267   existing -non_shared option.
 268
 269 Changes in 2.14:
 270
 271 * Added support for MIPS32 Release 2.
 272
 273 * Added support for Xtensa architecture.
 274
 275 * Support for Intel's iWMMXt processor (an ARM variant) added.
 276
 277 * An assembler test generator has been contributed and an example file that
 278   uses it (gas/testsuite/gas/all/test-gen.c and test-exmaple.c).
 279
 280 * Support for SH2E added.
 281
 282 * GASP has now been removed.
 283
 284 * Support for Texas Instruments TMS320C4x and TMS320C3x series of
 285   DSP's contributed by Michael Hayes and Svein E. Seldal.
 286
 287 * Support for the Ubicom IP2xxx microcontroller added.
 288
 289 Changes in 2.13:
 290
 291 * Support for the Fujitsu FRV architecture added by Red Hat. Models for FR400
 292   and FR500 included.
 293
 294 * Support for DLX processor added.
 295
 296 * GASP has now been deprecated and will be removed in a future release.  Use
 297   the macro facilities in GAS instead.
 298
 299 * GASP now correctly parses floating point numbers. Unless the base is
 300   explicitly specified, they are interpreted as decimal numbers regardless of
 301   the currently specified base.
 302
 303 Changes in 2.12:
 304
 305 * Support for Don Knuth's MMIX, by Hans-Peter Nilsson.
 306
 307 * Support for the OpenRISC 32-bit embedded processor by OpenCores.
 308
 309 * The ARM assembler now accepts -march=..., -mcpu=... and -mfpu=... for
 310   specifying the target instruction set.  The old method of specifying the
 311   target processor has been deprecated, but is still accepted for
 312   compatibility.
 313
 314 * Support for the VFP floating-point instruction set has been added to
 315   the ARM assembler.
 316
 317 * New psuedo op: .incbin to include a set of binary data at a given point
 318   in the assembly.  Contributed by Anders Norlander.
 319
 320 * The MIPS assembler now accepts -march/-mtune. -mcpu has been deprecated
 321   but still works for compatability.
 322
 323 * The MIPS assembler no longer issues a warning by default when it
 324   generates a nop instruction from a macro. The new command line option
 325   -n will turn on the warning.
 326
 327 Changes in 2.11:
 328
 329 * Support for PDP-11 and 2.11BSD a.out format, by Lars Brinkhoff.
 330
 331 * x86 gas now supports the full Pentium4 instruction set.
 332
 333 * Support for AMD x86-64 architecture, by Jan Hubicka, SuSE Labs.
 334
 335 * Support for Motorola 68HC11 and 68HC12.
 336
 337 * Support for Texas Instruments TMS320C54x (tic54x).
 338
 339 * Support for IA-64.
 340
 341 * Support for i860, by Jason Eckhardt.
 342
 343 * Support for CRIS (Axis Communications ETRAX series).
 344
 345 * x86 gas has a new .arch pseudo op to specify the target CPU architecture.
 346
 347 * x86 gas -q command line option quietens warnings about register size changes
 348   due to suffix, indirect jmp/call without `*', stand-alone prefixes, and
 349   translating various deprecated floating point instructions.
 350
 351 Changes in 2.10:
 352
 353 * Support for the ARM msr instruction was changed to only allow an immediate
 354   operand when altering the flags field.
 355
 356 * Support for ATMEL AVR.
 357
 358 * Support for IBM 370 ELF.  Somewhat experimental.
 359
 360 * Support for numbers with suffixes.
 361
 362 * Added support for breaking to the end of repeat loops.
 363
 364 * Added support for parallel instruction syntax (DOUBLEBAR_PARALLEL).
 365
 366 * New .elseif pseudo-op added.
 367
 368 * New --fatal-warnings option.
 369
 370 * picoJava architecture support added.
 371
 372 * Motorola MCore 210 processor support added.
 373
 374 * A new pseudo-op .intel_syntax has been implemented to allow gas to parse i386
 375   assembly programs with intel syntax.
 376
 377 * New pseudo-ops .func,.endfunc to aid in debugging user-written assembler code.
 378
 379 * Added -gdwarf2 option to generate DWARF 2 debugging information.
 380
 381 * Full 16-bit mode support for i386.
 382
 383 * Greatly improved instruction operand checking for i386.  This change will
 384   produce errors or warnings on incorrect assembly code that previous versions
 385   of gas accepted.  If you get unexpected messages from code that worked with
 386   older versions of gas, please double check the code before reporting a bug.
 387
 388 * Weak symbol support added for COFF targets.
 389
 390 * Mitsubishi D30V support added.
 391
 392 * Texas Instruments c80 (tms320c80) support added.
 393
 394 * i960 ELF support added.
 395
 396 * ARM ELF support added.
 397
 398 Changes in 2.9:
 399
 400 * Texas Instruments c30 (tms320c30) support added.
 401
 402 * The assembler now optimizes the exception frame information generated by egcs
 403   and gcc 2.8.  The new --traditional-format option disables this optimization.
 404
 405 * Added --gstabs option to generate stabs debugging information.
 406
 407 * The -a option takes a new suboption, m (e.g., -alm) to expand macros in a
 408   listing.
 409
 410 * Added -MD option to print dependencies.
 411
 412 Changes in 2.8:
 413
 414 * BeOS support added.
 415
 416 * MIPS16 support added.
 417
 418 * Motorola ColdFire 5200 support added (configure for m68k and use -m5200).
 419
 420 * Alpha/VMS support added.
 421
 422 * m68k options --base-size-default-16, --base-size-default-32,
 423   --disp-size-default-16, and --disp-size-default-32 added.
 424
 425 * The alignment directives now take an optional third argument, which is the
 426   maximum number of bytes to skip.  If doing the alignment would require
 427   skipping more than the given number of bytes, the alignment is not done at
 428   all.
 429
 430 * The ELF assembler has a new pseudo-op, .symver, used for symbol versioning.
 431
 432 * The -a option takes a new suboption, c (e.g., -alc), to skip false
 433   conditionals in listings.
 434
 435 * Added new pseudo-op, .equiv; it's like .equ, except that it is an error if
 436   the symbol is already defined.
 437
 438 Changes in 2.7:
 439
 440 * The PowerPC assembler now allows the use of symbolic register names (r0,
 441   etc.) if -mregnames is used.  Symbolic names preceded by a '%' (%r0, etc.)
 442   can be used any time.  PowerPC 860 move to/from SPR instructions have been
 443   added.
 444
 445 * Alpha Linux (ELF) support added.
 446
 447 * PowerPC ELF support added.
 448
 449 * m68k Linux (ELF) support added.
 450
 451 * i960 Hx/Jx support added.
 452
 453 * i386/PowerPC gnu-win32 support added.
 454
 455 * SCO ELF support added.  For OpenServer 5 targets (i386-unknown-sco3.2v5) the
 456   default is to build COFF-only support.  To get a set of tools that generate
 457   ELF (they'll understand both COFF and ELF), you must configure with
 458   target=i386-unknown-sco3.2v5elf.
 459
 460 * m88k-motorola-sysv3* support added.
 461
 462 Changes in 2.6:
 463
 464 * Gas now directly supports macros, without requiring GASP.
 465
 466 * Gas now has an MRI assembler compatibility mode.  Use -M or --mri to select
 467   MRI mode.  The pseudo-op ``.mri 1'' will switch into the MRI mode until the
 468   ``.mri 0'' is seen; this can be convenient for inline assembler code.
 469
 470 * Added --defsym SYM=VALUE option.
 471
 472 * Added -mips4 support to MIPS assembler.
 473
 474 * Added PIC support to Solaris and SPARC SunOS 4 assembler.
 475
 476 Changes in 2.4:
 477
 478 * Converted this directory to use an autoconf-generated configure script.
 479
 480 * ARM support, from Richard Earnshaw.
 481
 482 * Updated VMS support, from Pat Rankin, including considerably improved
 483   debugging support.
 484
 485 * Support for the control registers in the 68060.
 486
 487 * Handles (ignores) a new directive ".this_GCC_requires_the_GNU_assembler", to
 488   provide for possible future gcc changes, for targets where gas provides some
 489   features not available in the native assembler.  If the native assembler is
 490   used, it should become obvious pretty quickly what the problem is.
 491
 492 * Usage message is available with "--help".
 493
 494 * The GNU Assembler Preprocessor (gasp) is included.  (Actually, it was in 2.3
 495   also, but didn't get into the NEWS file.)
 496
 497 * Weak symbol support for a.out.
 498
 499 * A bug in the listing code which could cause an infinite loop has been fixed.
 500   Bugs in listings when generating a COFF object file have also been fixed.
 501
 502 * Initial i386-svr4 PIC implementation from Eric Youngdale, based on code by
 503   Paul Kranenburg.
 504
 505 * Improved Alpha support.  Immediate constants can have a much larger range
 506   now. Support for the 21164 has been contributed by Digital.
 507
 508 * Updated ns32k (pc532-mach, netbsd532) support from Ian Dall.
 509
 510 Changes in 2.3:
 511
 512 * Mach i386 support, by David Mackenzie and Ken Raeburn.
 513
 514 * RS/6000 and PowerPC support by Ian Taylor.
 515
 516 * VMS command scripts (make-gas.com, config-gas.com) have been worked on a bit,
 517   based on mail received from various people.  The `-h#' option should work
 518   again too.
 519
 520 * HP-PA work, by Jeff Law.  Note, for the PA, gas-2.3 has been designed to work
 521   with gdb-4.12 and gcc-2.6.  As gcc-2.6 has not been released yet, a special
 522   version of gcc-2.5.8 has been patched to work with gas-2.3.  You can retrieve
 523   this special version of gcc-2.5.8 via anonymous ftp from jaguar.cs.utah.edu
 524   in the "dist" directory.
 525
 526 * Vax support in gas fixed for BSD, so it builds and seems to run a couple
 527   simple tests okay.  I haven't put it through extensive testing.  (GNU make is
 528   currently required for BSD 4.3 builds.)
 529
 530 * Support for the DEC Alpha, running OSF/1 (ECOFF format).  The gas support is
 531   based on code donated by CMU, which used an a.out-based format.  I'm afraid
 532   the alpha-a.out support is pretty badly mangled, and much of it removed;
 533   making it work will require rewriting it as BFD support for the format anyways.
 534
 535 * Irix 5 support.
 536
 537 * The test suites have been fixed up a bit, so that they should work with a
 538   couple different versions of expect and dejagnu.
 539
 540 * Symbols' values are now handled internally as expressions, permitting more
 541   flexibility in evaluating them in some cases.  Some details of relocation
 542   handling have also changed, and simple constant pool management has been
 543   added, to make the Alpha port easier.
 544
 545 * New option "--statistics" for printing out program run times.  This is
 546   intended to be used with the gcc "-Q" option, which prints out times spent in
 547   various phases of compilation.  (You should be able to get all of them
 548   printed out with "gcc -Q -Wa,--statistics", I think.)
 549
 550 Changes in 2.2:
 551
 552 * RS/6000 AIX and MIPS SGI Irix 5 support has been added.
 553
 554 * Configurations that are still in development (and therefore are convenient to
 555   have listed in configure.in) still get rejected without a minor change to
 556   gas/Makefile.in, so people not doing development work shouldn't get the
 557   impression that support for such configurations is actually believed to be
 558   reliable.
 559
 560 * The program name (usually "as") is printed when a fatal error message is
 561   displayed.  This should prevent some confusion about the source of occasional
 562   messages about "internal errors".
 563
 564 * ELF support is falling into place.  Support for the 386 should be working.
 565   Support for SPARC Solaris is in.  HPPA support from Utah is being integrated.
 566
 567 * Symbol values are maintained as expressions instead of being immediately
 568   boiled down to add-symbol, sub-symbol, and constant.  This permits slightly
 569   more complex calculations involving symbols whose values are not alreadey
 570   known.
 571
 572 * DBX-style debugging info ("stabs") is now supported for COFF formats.
 573   If any stabs directives are seen in the source, GAS will create two new
 574   sections: a ".stab" and a ".stabstr" section.  The format of the .stab
 575   section is nearly identical to the a.out symbol format, and .stabstr is
 576   its string table.  For this to be useful, you must have configured GCC
 577   to generate stabs (by defining DBX_DEBUGGING_INFO), and must have a GDB
 578   that can use the stab sections (4.11 or later).
 579
 580 * LynxOS, on i386 and m68k platforms, is now supported.  SPARC LynxOS
 581   support is in progress.
 582
 583 Changes in 2.1:
 584
 585 * Several small fixes for i386-aix (PS/2) support from Minh Tran-Le have been
 586   incorporated, but not well tested yet.
 587
 588 * Altered the opcode table split for m68k; it should require less VM to compile
 589   with gcc now.
 590
 591 * Some minor adjustments to add (Convergent Technologies') Miniframe support,
 592   suggested by Ronald Cole.
 593
 594 * HPPA support (running OSF only, not HPUX) has been contributed by Utah.  This
 595   includes improved ELF support, which I've started adapting for SPARC Solaris
 596   2.x.  Integration isn't completely, so it probably won't work.
 597
 598 * HP9000/300 support, donated by HP, has been merged in.
 599
 600 * Ian Taylor has finished the MIPS ECOFF (Ultrix, Irix) support.
 601
 602 * Better error messages for unsupported configurations (e.g., hppa-hpux).
 603
 604 * Test suite framework is starting to become reasonable.
 605
 606 Changes in 2.0:
 607
 608 * Mostly bug fixes.
 609
 610 * Some more merging of BFD and ELF code, but ELF still doesn't work.
 611
 612 Changes in 1.94:
 613
 614 * BFD merge is partly done.  Adventurous souls may try giving configure the
 615   "--with-bfd-assembler" option.  Currently, ELF format requires it, a.out
 616   format accepts it; SPARC CPU accepts it.  It's the default only for OS "elf"
 617   or "solaris".  (ELF isn't really supported yet.  It needs work.  I've got
 618   some code from Utah for HP-PA ELF, and from DG for m88k ELF, but they're not
 619   fully merged yet.)
 620
 621 * The 68K opcode table has been split in half.  It should now compile under gcc
 622   without consuming ridiculous amounts of memory.
 623
 624 * A couple data structures have been reduced in size.  This should result in
 625   saving a little bit of space at runtime.
 626
 627 * Support for MIPS, from OSF and Ralph Campbell, has been merged in.  The OSF
 628   code provided ROSE format support, which I haven't merged in yet.  (I can
 629   make it available, if anyone wants to try it out.)  Ralph's code, for BSD
 630   4.4, supports a.out format.  We don't have ECOFF support in just yet; it's
 631   coming.
 632
 633 * Support for the Hitachi H8/500 has been added.
 634
 635 * VMS host and target support should be working now, thanks chiefly to Eric
 636   Youngdale.
 637
 638 Changes in 1.93.01:
 639
 640 * For m68k, support for more processors has been added: 68040, CPU32, 68851.
 641
 642 * For i386, .align is now power-of-two; was number-of-bytes.
 643
 644 * For m68k, "%" is now accepted before register names.  For COFF format, which
 645   doesn't use underscore prefixes for C labels, it is required, so variable "a0"
 646   can be distinguished from the register.
 647
 648 * Last public release was 1.38.  Lots of configuration changes since then, lots
 649   of new CPUs and formats, lots of bugs fixed.
 650
 651 \f
 652 Copyright (C) 2012-2015 Free Software Foundation, Inc.
 653
 654 Copying and distribution of this file, with or without modification,
 655 are permitted in any medium without royalty provided the copyright
 656 notice and this notice are preserved.
 657
 658 Local variables:
 659 fill-column: 79
 660 End: