gas/NEWS

   1 -*- text -*-
   2
   3 * Add support for the MIPS Loongson EXTensions (EXT) instructions.
   4
   5 * Add support for the MIPS Loongson Content Address Memory (CAM) ASE.
   6
   7 * Add support for the C-SKY processor series.
   8
   9 * Add support for the MIPS Loongson MultiMedia extensions Instructions (MMI)
  10   ASE.
  11
  12 Changes in 2.31:
  13
  14 * The ADR and ADRL pseudo-instructions supported by the ARM assembler
  15   now only set the bottom bit of the address of thumb function symbols
  16   if the -mthumb-interwork command line option is active.
  17
  18 * Add support for the MIPS Global INValidate (GINV) ASE.
  19
  20 * Add support for the MIPS Cyclic Redudancy Check (CRC) ASE.
  21
  22 * Add support for the Freescale S12Z architecture.
  23
  24 * Add --generate-missing-build-notes=[yes|no] option to create (or not) GNU
  25   Build Attribute notes if none are present in the input sources.  Add a
  26   --enable-generate-build-notes=[yes|no] configure time option to set the
  27   default behaviour.  Set the default if the configure option is not used
  28   to "no".
  29
  30 * Remove -mold-gcc command-line option for x86 targets.
  31
  32 * Add -O[2|s] command-line options to x86 assembler to enable alternate
  33   shorter instruction encoding.
  34
  35 * Add support for .nops directive.  It is currently supported only for
  36   x86 targets.
  37
  38 Changes in 2.30:
  39
  40 * Add support for loaction views in DWARF debug line information.
  41
  42 Changes in 2.29:
  43
  44 * Add support for ELF SHF_GNU_MBIND.
  45
  46 * Add support for the WebAssembly file format and wasm32 ELF conversion.
  47
  48 * PowerPC gas now checks that the correct register class is used in
  49   instructions.  For instance, "addi %f4,%cr3,%r31" warns three times
  50   that the registers are invalid.
  51
  52 * Add support for the Texas Instruments PRU processor.
  53
  54 * Support for the ARMv8-R architecture and Cortex-R52 processor has been
  55   added to the ARM port.
  56
  57 Changes in 2.28:
  58
  59 * Add support for the RISC-V architecture.
  60
  61 * Add support for the ARM Cortex-M23 and Cortex-M33 processors.
  62
  63 Changes in 2.27:
  64
  65 * Default to --enable-compressed-debug-sections=gas for Linux/x86 targets.
  66
  67 * Add --no-pad-sections to stop the assembler from padding the end of output
  68   sections up to their alignment boundary.
  69
  70 * Support for the ARMv8-M architecture has been added to the ARM port.  Support
  71   for the ARMv8-M Security and DSP Extensions has also been added to the ARM
  72   port.
  73
  74 * ARC backend accepts .extInstruction, .extCondCode, .extAuxRegister, and
  75   .extCoreRegister pseudo-ops that allow an user to define custom
  76   instructions, conditional codes, auxiliary and core registers.
  77
  78 * Add a configure option --enable-elf-stt-common to decide whether ELF
  79   assembler should generate common symbols with the STT_COMMON type by
  80   default.  Default to no.
  81
  82 * New command-line option --elf-stt-common= for ELF targets to control
  83   whether to generate common symbols with the STT_COMMON type.
  84
  85 * Add ability to set section flags and types via numeric values for ELF
  86   based targets.
  87
  88 * Add a configure option --enable-x86-relax-relocations to decide whether
  89   x86 assembler should generate relax relocations by default.  Default to
  90   yes, except for x86 Solaris targets older than Solaris 12.
  91
  92 * New command-line option -mrelax-relocations= for x86 target to control
  93   whether to generate relax relocations.
  94
  95 * New command-line option -mfence-as-lock-add=yes for x86 target to encode
  96   lfence, mfence and sfence as "lock addl $0x0, (%[re]sp)".
  97
  98 * Add assembly-time relaxation option for ARC cpus.
  99
 100 * Add --with-cpu=TYPE configure option for ARC gas.  This allows the default
 101   cpu type to be adjusted at configure time.
 102
 103 Changes in 2.26:
 104
 105 * Add a configure option --enable-compressed-debug-sections={all,gas} to
 106   decide whether DWARF debug sections should be compressed by default.
 107
 108 * Add support for the ARC EM/HS, and ARC600/700 architectures.  Remove
 109   assembler support for Argonaut RISC architectures.
 110
 111 * Symbol and label names can now be enclosed in double quotes (") which allows
 112   them to contain characters that are not part of valid symbol names in high
 113   level languages.
 114
 115 * Added the correctly spelled -march=armv6kz, for ARMv6KZ support.  The
 116   previous spelling, -march=armv6zk, is still accepted.
 117
 118 * Support for the ARMv8.1 architecture has been added to the Aarch64 port.
 119   Support for the individual ARMv8.1 Adv.SIMD, LOR and PAN architecture
 120   extensions has also been added to the Aarch64 port.
 121
 122 * Support for the ARMv8.1 architecture has been added to the ARM port.  Support
 123   for the individual ARMv8.1 Adv.SIMD and PAN architecture extensions has also
 124   been added to the ARM port.
 125
 126 * Extend --compress-debug-sections option to support
 127   --compress-debug-sections=[none|zlib|zlib-gnu|zlib-gabi] for ELF
 128   targets.
 129
 130 * --compress-debug-sections is turned on for Linux/x86 by default.
 131
 132 Changes in 2.25:
 133
 134 * Add support for the AVR Tiny microcontrollers.
 135
 136 * Replace support for openrisc and or32 with support for or1k.
 137
 138 * Enhanced the ARM port to accept the assembler output from the CodeComposer
 139   Studio tool.  Support is enabled via the new command-line option -mccs.
 140
 141 * Add support for the Andes NDS32.
 142
 143 Changes in 2.24:
 144
 145 * Add support for the Texas Instruments MSP430X processor.
 146
 147 * Add -gdwarf-sections command-line option to enable per-code-section
 148   generation of DWARF .debug_line sections.
 149
 150 * Add support for Altera Nios II.
 151
 152 * Add support for the Imagination Technologies Meta processor.
 153
 154 * Add support for the v850e3v5.
 155
 156 * Remove assembler support for MIPS ECOFF targets.
 157
 158 Changes in 2.23:
 159
 160 * Add support for the 64-bit ARM architecture: AArch64.
 161
 162 * Add support for S12X processor.
 163
 164 * Add support for the VLE extension to the PowerPC architecture.
 165
 166 * Add support for the Freescale XGATE architecture.
 167
 168 * Add support for .bundle_align_mode, .bundle_lock, and .bundle_unlock
 169   directives.  These are currently available only for x86 and ARM targets.
 170
 171 * Add support for the Renesas RL78 architecture.
 172
 173 * Add support for the Adapteva EPIPHANY architecture.
 174
 175 * For x86, allow 'rep bsf', 'rep bsr', and 'rep ret' syntax.
 176
 177 Changes in 2.22:
 178
 179 * Add support for the Tilera TILEPro and TILE-Gx architectures.
 180
 181 Changes in 2.21:
 182
 183 * Gas no longer requires doubling of ampersands in macros.
 184
 185 * Add support for the TMS320C6000 (TI C6X) processor family.
 186
 187 * GAS now understands an extended syntax in the .section directive flags
 188   for COFF targets that allows the section's alignment to be specified.  This
 189   feature has also been backported to the 2.20 release series, starting with
 190   2.20.1.
 191
 192 * Add support for the Renesas RX processor.
 193
 194 * New command-line option, --compress-debug-sections, which requests
 195   compression of DWARF debug information sections in the relocatable output
 196   file.  Compressed debug sections are supported by readelf, objdump, and
 197   gold, but not currently by Gnu ld.
 198
 199 Changes in 2.20:
 200
 201 * Added support for v850e2 and v850e2v3.
 202
 203 * GNU/Linux targets now supports "gnu_unique_object" as a value in the .type
 204   pseudo op.  It marks the symbol as being globally unique in the entire
 205   process.
 206
 207 * ARM assembler now supports .inst[.nw] pseudo-ops to insert opcodes specified
 208   in binary rather than text.
 209
 210 * Add support for common symbol alignment to PE formats.
 211
 212 * Add support for the new discriminator column in the DWARF line table,
 213   with a discriminator operand for the .loc directive.
 214
 215 * Add support for Sunplus score architecture.
 216
 217 * The .type pseudo-op now accepts a type of STT_GNU_IFUNC which can be used to
 218   indicate that if the symbol is the target of a relocation, its value should
 219   not be use.  Instead the function should be invoked and its result used as
 220   the value.
 221
 222 * Add support for Lattice Mico32 (lm32) architecture.
 223
 224 * Add support for Xilinx MicroBlaze architecture.
 225
 226 Changes in 2.19:
 227
 228 * New pseudo op .cfi_val_encoded_addr, to record constant addresses in unwind
 229   tables without runtime relocation.
 230
 231 * New command-line option, -h-tick-hex, for sh, m32c, and h8/300 targets, which
 232   adds compatibility with H'00 style hex constants.
 233
 234 * New command-line option, -msse-check=[none|error|warning], for x86
 235   targets.
 236
 237 * New sub-option added to the assembler's -a command-line switch to
 238   generate a listing output.  The 'g' sub-option will insert into the listing
 239   various information about the assembly, such as assembler version, the
 240   command-line options used, and a time stamp.
 241
 242 * New command-line option -msse2avx for x86 target to encode SSE
 243   instructions with VEX prefix.
 244
 245 * Add Intel XSAVE, EPT, MOVBE, AES, PCLMUL, AVX/FMA support for x86 target.
 246
 247 * New command-line options, -march=CPU[,+EXTENSION...], -mtune=CPU,
 248   -mmnemonic=[att|intel], -msyntax=[att|intel], -mindex-reg,
 249   -mnaked-reg and -mold-gcc, for x86 targets.
 250
 251 * Support for generating wide character strings has been added via the new
 252   pseudo ops: .string16, .string32 and .string64.
 253
 254 * Support for SSE5 has been added to the i386 port.
 255
 256 Changes in 2.18:
 257
 258 * The GAS sources are now released under the GPLv3.
 259
 260 * Support for the National Semiconductor CR16 target has been added.
 261
 262 * Added gas .reloc pseudo.  This is a low-level interface for creating
 263   relocations.
 264
 265 * Add support for x86_64 PE+ target.
 266
 267 * Add support for Score target.
 268
 269 Changes in 2.17:
 270
 271 * Support for the Infineon XC16X has been added by KPIT Cummins Infosystems.
 272
 273 * Support for ms2 architecture has been added.
 274
 275 * Support for the Z80 processor family has been added.
 276
 277 * Add support for the "@<file>" syntax to the command line, so that extra
 278   switches can be read from <file>.
 279
 280 * The SH target supports a new command-line switch --enable-reg-prefix which,
 281   if enabled, will allow register names to be optionally prefixed with a $
 282   character.  This allows register names to be distinguished from label names.
 283
 284 * Macros with a variable number of arguments are now supported. See the
 285   documentation for how this works.
 286
 287 * Added --reduce-memory-overheads switch to reduce the size of the hash
 288   tables used, at the expense of longer assembly times, and
 289   --hash-size=<NUMBER> to set the size of the hash tables used by gas.
 290
 291 * Macro names and macro parameter names can now be any identifier that would
 292   also be legal as a symbol elsewhere. For macro parameter names, this is
 293   known to cause problems in certain sources when the respective target uses
 294   characters inconsistently, and thus macro parameter references may no longer
 295   be recognized as such (see the documentation for details).
 296
 297 * Support the .f_floating, .d_floating, .g_floating and .h_floating directives
 298   for the VAX target in order to be more compatible with the VAX MACRO
 299   assembler.
 300
 301 * New command-line option -mtune=[itanium1|itanium2] for IA64 targets.
 302
 303 Changes in 2.16:
 304
 305 * Redefinition of macros now results in an error.
 306
 307 * New command-line option -mhint.b=[ok|warning|error] for IA64 targets.
 308
 309 * New command-line option -munwind-check=[warning|error] for IA64
 310   targets.
 311
 312 * The IA64 port now uses automatic dependency violation removal as its default
 313   mode.
 314
 315 * Port to MAXQ processor contributed by HCL Tech.
 316
 317 * Added support for generating unwind tables for ARM ELF targets.
 318
 319 * Add a -g command-line option to generate debug information in the target's
 320   preferred debug format.
 321
 322 * Support for the crx-elf target added.
 323
 324 * Support for the sh-symbianelf target added.
 325
 326 * Added a pseudo-op (.secrel32) to generate 32 bit section relative relocations
 327   on pe[i]-i386; required for this target's DWARF 2 support.
 328
 329 * Support for Motorola MCF521x/5249/547x/548x added.
 330
 331 * Support for ColdFire EMAC instructions added and Motorola syntax for MAC/EMAC
 332   instrucitons.
 333
 334 * New command-line option -mno-shared for MIPS ELF targets.
 335
 336 * New command-line option --alternate and pseudo-ops .altmacro and .noaltmacro
 337   added to enter (and leave) alternate macro syntax mode.
 338
 339 Changes in 2.15:
 340
 341 * The MIPS -membedded-pic option (Embedded-PIC code generation) is
 342   deprecated and will be removed in a future release.
 343
 344 * Added PIC m32r Linux (ELF) and support to M32R assembler.
 345
 346 * Added support for ARM V6.
 347
 348 * Added support for sh4a and variants.
 349
 350 * Support for Renesas M32R2 added.
 351
 352 * Limited support for Mapping Symbols as specified in the ARM ELF
 353   specification has been added to the arm assembler.
 354
 355 * On ARM architectures, added a new gas directive ".unreq" that undoes
 356   definitions created by ".req".
 357
 358 * Support for Motorola ColdFire MCF528x added.
 359
 360 * Added --gstabs+ switch to enable the generation of STABS debug format
 361   information with GNU extensions.
 362
 363 * Added support for MIPS64 Release 2.
 364
 365 * Added support for v850e1.
 366
 367 * Added -n switch for x86 assembler.  By default, x86 GAS replaces
 368   multiple nop instructions used for alignment within code sections
 369   with multi-byte nop instructions such as leal 0(%esi,1),%esi.  This
 370   switch disables the optimization.
 371
 372 * Removed -n option from MIPS assembler.  It was not useful, and confused the
 373   existing -non_shared option.
 374
 375 Changes in 2.14:
 376
 377 * Added support for MIPS32 Release 2.
 378
 379 * Added support for Xtensa architecture.
 380
 381 * Support for Intel's iWMMXt processor (an ARM variant) added.
 382
 383 * An assembler test generator has been contributed and an example file that
 384   uses it (gas/testsuite/gas/all/test-gen.c and test-exmaple.c).
 385
 386 * Support for SH2E added.
 387
 388 * GASP has now been removed.
 389
 390 * Support for Texas Instruments TMS320C4x and TMS320C3x series of
 391   DSP's contributed by Michael Hayes and Svein E. Seldal.
 392
 393 * Support for the Ubicom IP2xxx microcontroller added.
 394
 395 Changes in 2.13:
 396
 397 * Support for the Fujitsu FRV architecture added by Red Hat. Models for FR400
 398   and FR500 included.
 399
 400 * Support for DLX processor added.
 401
 402 * GASP has now been deprecated and will be removed in a future release.  Use
 403   the macro facilities in GAS instead.
 404
 405 * GASP now correctly parses floating point numbers. Unless the base is
 406   explicitly specified, they are interpreted as decimal numbers regardless of
 407   the currently specified base.
 408
 409 Changes in 2.12:
 410
 411 * Support for Don Knuth's MMIX, by Hans-Peter Nilsson.
 412
 413 * Support for the OpenRISC 32-bit embedded processor by OpenCores.
 414
 415 * The ARM assembler now accepts -march=..., -mcpu=... and -mfpu=... for
 416   specifying the target instruction set.  The old method of specifying the
 417   target processor has been deprecated, but is still accepted for
 418   compatibility.
 419
 420 * Support for the VFP floating-point instruction set has been added to
 421   the ARM assembler.
 422
 423 * New psuedo op: .incbin to include a set of binary data at a given point
 424   in the assembly.  Contributed by Anders Norlander.
 425
 426 * The MIPS assembler now accepts -march/-mtune. -mcpu has been deprecated
 427   but still works for compatability.
 428
 429 * The MIPS assembler no longer issues a warning by default when it
 430   generates a nop instruction from a macro. The new command-line option
 431   -n will turn on the warning.
 432
 433 Changes in 2.11:
 434
 435 * Support for PDP-11 and 2.11BSD a.out format, by Lars Brinkhoff.
 436
 437 * x86 gas now supports the full Pentium4 instruction set.
 438
 439 * Support for AMD x86-64 architecture, by Jan Hubicka, SuSE Labs.
 440
 441 * Support for Motorola 68HC11 and 68HC12.
 442
 443 * Support for Texas Instruments TMS320C54x (tic54x).
 444
 445 * Support for IA-64.
 446
 447 * Support for i860, by Jason Eckhardt.
 448
 449 * Support for CRIS (Axis Communications ETRAX series).
 450
 451 * x86 gas has a new .arch pseudo op to specify the target CPU architecture.
 452
 453 * x86 gas -q command-line option quietens warnings about register size changes
 454   due to suffix, indirect jmp/call without `*', stand-alone prefixes, and
 455   translating various deprecated floating point instructions.
 456
 457 Changes in 2.10:
 458
 459 * Support for the ARM msr instruction was changed to only allow an immediate
 460   operand when altering the flags field.
 461
 462 * Support for ATMEL AVR.
 463
 464 * Support for IBM 370 ELF.  Somewhat experimental.
 465
 466 * Support for numbers with suffixes.
 467
 468 * Added support for breaking to the end of repeat loops.
 469
 470 * Added support for parallel instruction syntax (DOUBLEBAR_PARALLEL).
 471
 472 * New .elseif pseudo-op added.
 473
 474 * New --fatal-warnings option.
 475
 476 * picoJava architecture support added.
 477
 478 * Motorola MCore 210 processor support added.
 479
 480 * A new pseudo-op .intel_syntax has been implemented to allow gas to parse i386
 481   assembly programs with intel syntax.
 482
 483 * New pseudo-ops .func,.endfunc to aid in debugging user-written assembler code.
 484
 485 * Added -gdwarf2 option to generate DWARF 2 debugging information.
 486
 487 * Full 16-bit mode support for i386.
 488
 489 * Greatly improved instruction operand checking for i386.  This change will
 490   produce errors or warnings on incorrect assembly code that previous versions
 491   of gas accepted.  If you get unexpected messages from code that worked with
 492   older versions of gas, please double check the code before reporting a bug.
 493
 494 * Weak symbol support added for COFF targets.
 495
 496 * Mitsubishi D30V support added.
 497
 498 * Texas Instruments c80 (tms320c80) support added.
 499
 500 * i960 ELF support added.
 501
 502 * ARM ELF support added.
 503
 504 Changes in 2.9:
 505
 506 * Texas Instruments c30 (tms320c30) support added.
 507
 508 * The assembler now optimizes the exception frame information generated by egcs
 509   and gcc 2.8.  The new --traditional-format option disables this optimization.
 510
 511 * Added --gstabs option to generate stabs debugging information.
 512
 513 * The -a option takes a new suboption, m (e.g., -alm) to expand macros in a
 514   listing.
 515
 516 * Added -MD option to print dependencies.
 517
 518 Changes in 2.8:
 519
 520 * BeOS support added.
 521
 522 * MIPS16 support added.
 523
 524 * Motorola ColdFire 5200 support added (configure for m68k and use -m5200).
 525
 526 * Alpha/VMS support added.
 527
 528 * m68k options --base-size-default-16, --base-size-default-32,
 529   --disp-size-default-16, and --disp-size-default-32 added.
 530
 531 * The alignment directives now take an optional third argument, which is the
 532   maximum number of bytes to skip.  If doing the alignment would require
 533   skipping more than the given number of bytes, the alignment is not done at
 534   all.
 535
 536 * The ELF assembler has a new pseudo-op, .symver, used for symbol versioning.
 537
 538 * The -a option takes a new suboption, c (e.g., -alc), to skip false
 539   conditionals in listings.
 540
 541 * Added new pseudo-op, .equiv; it's like .equ, except that it is an error if
 542   the symbol is already defined.
 543
 544 Changes in 2.7:
 545
 546 * The PowerPC assembler now allows the use of symbolic register names (r0,
 547   etc.) if -mregnames is used.  Symbolic names preceded by a '%' (%r0, etc.)
 548   can be used any time.  PowerPC 860 move to/from SPR instructions have been
 549   added.
 550
 551 * Alpha Linux (ELF) support added.
 552
 553 * PowerPC ELF support added.
 554
 555 * m68k Linux (ELF) support added.
 556
 557 * i960 Hx/Jx support added.
 558
 559 * i386/PowerPC gnu-win32 support added.
 560
 561 * SCO ELF support added.  For OpenServer 5 targets (i386-unknown-sco3.2v5) the
 562   default is to build COFF-only support.  To get a set of tools that generate
 563   ELF (they'll understand both COFF and ELF), you must configure with
 564   target=i386-unknown-sco3.2v5elf.
 565
 566 * m88k-motorola-sysv3* support added.
 567
 568 Changes in 2.6:
 569
 570 * Gas now directly supports macros, without requiring GASP.
 571
 572 * Gas now has an MRI assembler compatibility mode.  Use -M or --mri to select
 573   MRI mode.  The pseudo-op ``.mri 1'' will switch into the MRI mode until the
 574   ``.mri 0'' is seen; this can be convenient for inline assembler code.
 575
 576 * Added --defsym SYM=VALUE option.
 577
 578 * Added -mips4 support to MIPS assembler.
 579
 580 * Added PIC support to Solaris and SPARC SunOS 4 assembler.
 581
 582 Changes in 2.4:
 583
 584 * Converted this directory to use an autoconf-generated configure script.
 585
 586 * ARM support, from Richard Earnshaw.
 587
 588 * Updated VMS support, from Pat Rankin, including considerably improved
 589   debugging support.
 590
 591 * Support for the control registers in the 68060.
 592
 593 * Handles (ignores) a new directive ".this_GCC_requires_the_GNU_assembler", to
 594   provide for possible future gcc changes, for targets where gas provides some
 595   features not available in the native assembler.  If the native assembler is
 596   used, it should become obvious pretty quickly what the problem is.
 597
 598 * Usage message is available with "--help".
 599
 600 * The GNU Assembler Preprocessor (gasp) is included.  (Actually, it was in 2.3
 601   also, but didn't get into the NEWS file.)
 602
 603 * Weak symbol support for a.out.
 604
 605 * A bug in the listing code which could cause an infinite loop has been fixed.
 606   Bugs in listings when generating a COFF object file have also been fixed.
 607
 608 * Initial i386-svr4 PIC implementation from Eric Youngdale, based on code by
 609   Paul Kranenburg.
 610
 611 * Improved Alpha support.  Immediate constants can have a much larger range
 612   now. Support for the 21164 has been contributed by Digital.
 613
 614 * Updated ns32k (pc532-mach, netbsd532) support from Ian Dall.
 615
 616 Changes in 2.3:
 617
 618 * Mach i386 support, by David Mackenzie and Ken Raeburn.
 619
 620 * RS/6000 and PowerPC support by Ian Taylor.
 621
 622 * VMS command scripts (make-gas.com, config-gas.com) have been worked on a bit,
 623   based on mail received from various people.  The `-h#' option should work
 624   again too.
 625
 626 * HP-PA work, by Jeff Law.  Note, for the PA, gas-2.3 has been designed to work
 627   with gdb-4.12 and gcc-2.6.  As gcc-2.6 has not been released yet, a special
 628   version of gcc-2.5.8 has been patched to work with gas-2.3.  You can retrieve
 629   this special version of gcc-2.5.8 via anonymous ftp from jaguar.cs.utah.edu
 630   in the "dist" directory.
 631
 632 * Vax support in gas fixed for BSD, so it builds and seems to run a couple
 633   simple tests okay.  I haven't put it through extensive testing.  (GNU make is
 634   currently required for BSD 4.3 builds.)
 635
 636 * Support for the DEC Alpha, running OSF/1 (ECOFF format).  The gas support is
 637   based on code donated by CMU, which used an a.out-based format.  I'm afraid
 638   the alpha-a.out support is pretty badly mangled, and much of it removed;
 639   making it work will require rewriting it as BFD support for the format anyways.
 640
 641 * Irix 5 support.
 642
 643 * The test suites have been fixed up a bit, so that they should work with a
 644   couple different versions of expect and dejagnu.
 645
 646 * Symbols' values are now handled internally as expressions, permitting more
 647   flexibility in evaluating them in some cases.  Some details of relocation
 648   handling have also changed, and simple constant pool management has been
 649   added, to make the Alpha port easier.
 650
 651 * New option "--statistics" for printing out program run times.  This is
 652   intended to be used with the gcc "-Q" option, which prints out times spent in
 653   various phases of compilation.  (You should be able to get all of them
 654   printed out with "gcc -Q -Wa,--statistics", I think.)
 655
 656 Changes in 2.2:
 657
 658 * RS/6000 AIX and MIPS SGI Irix 5 support has been added.
 659
 660 * Configurations that are still in development (and therefore are convenient to
 661   have listed in configure.in) still get rejected without a minor change to
 662   gas/Makefile.in, so people not doing development work shouldn't get the
 663   impression that support for such configurations is actually believed to be
 664   reliable.
 665
 666 * The program name (usually "as") is printed when a fatal error message is
 667   displayed.  This should prevent some confusion about the source of occasional
 668   messages about "internal errors".
 669
 670 * ELF support is falling into place.  Support for the 386 should be working.
 671   Support for SPARC Solaris is in.  HPPA support from Utah is being integrated.
 672
 673 * Symbol values are maintained as expressions instead of being immediately
 674   boiled down to add-symbol, sub-symbol, and constant.  This permits slightly
 675   more complex calculations involving symbols whose values are not alreadey
 676   known.
 677
 678 * DBX-style debugging info ("stabs") is now supported for COFF formats.
 679   If any stabs directives are seen in the source, GAS will create two new
 680   sections: a ".stab" and a ".stabstr" section.  The format of the .stab
 681   section is nearly identical to the a.out symbol format, and .stabstr is
 682   its string table.  For this to be useful, you must have configured GCC
 683   to generate stabs (by defining DBX_DEBUGGING_INFO), and must have a GDB
 684   that can use the stab sections (4.11 or later).
 685
 686 * LynxOS, on i386 and m68k platforms, is now supported.  SPARC LynxOS
 687   support is in progress.
 688
 689 Changes in 2.1:
 690
 691 * Several small fixes for i386-aix (PS/2) support from Minh Tran-Le have been
 692   incorporated, but not well tested yet.
 693
 694 * Altered the opcode table split for m68k; it should require less VM to compile
 695   with gcc now.
 696
 697 * Some minor adjustments to add (Convergent Technologies') Miniframe support,
 698   suggested by Ronald Cole.
 699
 700 * HPPA support (running OSF only, not HPUX) has been contributed by Utah.  This
 701   includes improved ELF support, which I've started adapting for SPARC Solaris
 702   2.x.  Integration isn't completely, so it probably won't work.
 703
 704 * HP9000/300 support, donated by HP, has been merged in.
 705
 706 * Ian Taylor has finished the MIPS ECOFF (Ultrix, Irix) support.
 707
 708 * Better error messages for unsupported configurations (e.g., hppa-hpux).
 709
 710 * Test suite framework is starting to become reasonable.
 711
 712 Changes in 2.0:
 713
 714 * Mostly bug fixes.
 715
 716 * Some more merging of BFD and ELF code, but ELF still doesn't work.
 717
 718 Changes in 1.94:
 719
 720 * BFD merge is partly done.  Adventurous souls may try giving configure the
 721   "--with-bfd-assembler" option.  Currently, ELF format requires it, a.out
 722   format accepts it; SPARC CPU accepts it.  It's the default only for OS "elf"
 723   or "solaris".  (ELF isn't really supported yet.  It needs work.  I've got
 724   some code from Utah for HP-PA ELF, and from DG for m88k ELF, but they're not
 725   fully merged yet.)
 726
 727 * The 68K opcode table has been split in half.  It should now compile under gcc
 728   without consuming ridiculous amounts of memory.
 729
 730 * A couple data structures have been reduced in size.  This should result in
 731   saving a little bit of space at runtime.
 732
 733 * Support for MIPS, from OSF and Ralph Campbell, has been merged in.  The OSF
 734   code provided ROSE format support, which I haven't merged in yet.  (I can
 735   make it available, if anyone wants to try it out.)  Ralph's code, for BSD
 736   4.4, supports a.out format.  We don't have ECOFF support in just yet; it's
 737   coming.
 738
 739 * Support for the Hitachi H8/500 has been added.
 740
 741 * VMS host and target support should be working now, thanks chiefly to Eric
 742   Youngdale.
 743
 744 Changes in 1.93.01:
 745
 746 * For m68k, support for more processors has been added: 68040, CPU32, 68851.
 747
 748 * For i386, .align is now power-of-two; was number-of-bytes.
 749
 750 * For m68k, "%" is now accepted before register names.  For COFF format, which
 751   doesn't use underscore prefixes for C labels, it is required, so variable "a0"
 752   can be distinguished from the register.
 753
 754 * Last public release was 1.38.  Lots of configuration changes since then, lots
 755   of new CPUs and formats, lots of bugs fixed.
 756
 757 \f
 758 Copyright (C) 2012-2018 Free Software Foundation, Inc.
 759
 760 Copying and distribution of this file, with or without modification,
 761 are permitted in any medium without royalty provided the copyright
 762 notice and this notice are preserved.
 763
 764 Local variables:
 765 fill-column: 79
 766 End: