/usr/share/doc/gcc-4.9-aarch64-linux-gnu-base/NEWS.html is in gcc-4.9-aarch64-linux-gnu 4.9.3-13ubuntu2cross1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 | <?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<link rev="made" href="mailto:gcc@gcc.gnu.org" />
<link rel="stylesheet" type="text/css" href="http://gcc.gnu.org/gcc.css" />
<title>
GCC 4.9 Release Series — Changes, New Features, and Fixes
- GNU Project - Free Software Foundation (FSF)</title>
</head>
<!-- GCC maintainers, please do not hesitate to update/contribute entries
concerning those part of GCC you maintain! 2002-03-23, Gerald.
-->
<body>
<h1>GCC 4.9 Release Series<br />Changes, New Features, and Fixes</h1>
<h2>Caveats</h2>
<ul>
<li><p>The mudflap run time checker has been removed. The mudflap
options remain, but do nothing.</p></li>
<li><p>Support for a number of older systems and recently
unmaintained or untested target ports of GCC has been declared
obsolete in GCC 4.9. Unless there is activity to revive them, the
next release of GCC will have their sources permanently
<strong>removed</strong>.</p>
<p>The following ports for individual systems on
particular architectures have been obsoleted:</p>
<ul>
<li>Solaris 9 (*-*-solaris2.9). Details can be found in the
<a href="http://gcc.gnu.org/ml/gcc-patches/2013-05/msg00728.html">
announcement</a>.</li>
</ul>
</li>
</ul>
<p>
More information on porting to GCC 4.9 from previous versions
of GCC can be found in
the <a href="http://gcc.gnu.org/gcc-4.9/porting_to.html">porting
guide</a> for this release.
</p>
<h2>General Optimizer Improvements</h2>
<ul>
<li>AddressSanitizer, a fast memory error detector, is now available on ARM.
</li>
<li>UndefinedBehaviorSanitizer (ubsan), a fast undefined behavior detector,
has been added and can be enabled via <code>-fsanitize=undefined</code>.
Various computations will be instrumented to detect undefined behavior
at runtime. UndefinedBehaviorSanitizer is currently available for the C
and C++ languages.
</li>
<li>Link-time optimization (LTO) improvements:
<ul>
<li>Type merging was rewritten. The new implementation is significantly faster
and uses less memory.</li>
<li>Better partitioning algorithm resulting in less streaming during
link time.</li>
<li>Early removal of virtual methods reduces the size of object files and
improves link-time memory usage and compile time.</li>
<li>Function bodies are now loaded on-demand and released early improving
overall memory usage at link time.</li>
<li>C++ hidden keyed methods can now be optimized out.</li>
<li>When using a linker plugin, compiling with the <code>-flto</code>
option now generates slim objects files (<code>.o</code>) which only
contain intermediate language representation for LTO. Use
<code>-ffat-lto-objects</code> to create files which contain
additionally the object code. To generate static libraries suitable
for LTO processing, use <code>gcc-ar</code> and
<code>gcc-ranlib</code>; to list symbols from a slim object file use
<code>gcc-nm</code>. (Requires that <code>ar</code>,
<code>ranlib</code> and <code>nm</code> have been compiled with
plugin support.)</li>
</ul>
Memory usage building Firefox with debug enabled was reduced from 15GB to
3.5GB; link time from 1700 seconds to 350 seconds.
</li>
<li>Inter-procedural optimization improvements:
<ul>
<li>New type inheritance analysis module improving devirtualization.
Devirtualization now takes into account anonymous name-spaces and the
C++11 <code>final</code> keyword.</li>
<li>New speculative devirtualization pass (controlled by
<code>-fdevirtualize-speculatively</code>.</li>
<li>Calls that were speculatively made direct are turned back to indirect
where direct call is not cheaper.</li>
<li>Local aliases are introduced for symbols that are known to be
semantically equivalent across shared libraries improving dynamic
linking times.</li>
</ul></li>
<li>Feedback directed optimization improvements:
<ul>
<li>Profiling of programs using C++ inline functions is now more reliable.</li>
<li>New time profiling determines typical order in which functions are
executed.</li>
<li>A new function reordering pass (controlled by
<code>-freorder-functions</code>) significantly reduces
startup time of large applications. Until binutils support is
completed, it is effective only with link-time optimization.</li>
<li>Feedback driven indirect call removal and devirtualization now handle
cross-module calls when link-time optimization is enabled.</li>
</ul></li>
</ul>
<h2 id="languages">New Languages and Language specific improvements</h2>
<ul>
<li>Version 4.0 of the <a href="http://openmp.org/wp/openmp-specifications/"
>OpenMP specification</a> is now supported for the C and C++ compilers.
The new <code>-fopenmp-simd</code> option can be used to enable OpenMP's
SIMD directives, while ignoring other OpenMP directives. The new <a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Optimize-Options.html#index-fsimd-cost-model-908"
><code>-fsimd-cost-model=</code></a> option permits to tune the
vectorization cost model for loops annotated with OpenMP and Cilk
Plus <code>simd</code> directives; <code>-Wopenmp-simd</code> warns when
the current costmodel overrides simd directives set by the user.</li>
<li>The <code>-Wdate-time</code> option has been added for the C, C++ and
Fortran compilers, which warns when the <code>__DATE__</code>,
<code>__TIME__</code> or <code>__TIMESTAMP__</code> macros are used.
Those macros might prevent bit-wise-identical reproducible
compilations.</li>
</ul>
<h3 id="ada">Ada</h3>
<ul>
<li>GNAT switched to Ada 2012 instead of Ada 2005 by default.</li>
</ul>
<h3 id="c-family">C family</h3>
<ul>
<li>Support for colorizing diagnostics emitted by GCC has been added.
The <code><a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Language-Independent-Options.html#index-fdiagnostics-color-252"
>-fdiagnostics-color=auto</a></code> will enable it when
outputting to terminals, <code>-fdiagnostics-color=always</code>
unconditionally. The <code>GCC_COLORS</code> environment variable
can be used to customize the colors or disable coloring.
If <code>GCC_COLORS</code> variable is present in the environment,
the default is <code>-fdiagnostics-color=auto</code>, otherwise
<code>-fdiagnostics-color=never</code>.<br/>
Sample diagnostics output:<br/>
<pre>
$ g++ -fdiagnostics-color=always -S -Wall test.C
<b>test.C:</b> In function ‘<b>int foo()</b>’:
<b>test.C:1:14:</b> <b style='color:magenta'>warning:</b> no return statement in function returning non-void [-Wreturn-type]
int foo () { }
<b style='color:lime'>^</b>
<b>test.C:2:46:</b> <b style='color:red'>error:</b> template instantiation depth exceeds maximum of 900 (use -ftemplate-depth= to increase the maximum) instantiating ‘<b>struct X<100></b>’
template <int N> struct X { static const int value = X<N-1>::value; }; template struct X<1000>;
<b style='color:lime'>^</b>
<b>test.C:2:46:</b> recursively required from ‘<b>const int X<999>::value</b>’
<b>test.C:2:46:</b> required from ‘<b>const int X<1000>::value</b>’
<b>test.C:2:88:</b> required from here
<b>test.C:2:46:</b> <b style='color:red'>error:</b> incomplete type ‘<b>X<100></b>’ used in nested name specifier
</pre></li>
<li>With the new <a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Loop-Specific-Pragmas.html"
><code>#pragma GCC ivdep</code></a>, the user can assert that there are no
loop-carried dependencies which would prevent concurrent execution of
consecutive iterations using SIMD (single instruction multiple data)
instructions.</li>
<li>Support for <a href="https://www.cilkplus.org/">Cilk Plus</a> has been
added and can be enabled with the <code>-fcilkplus</code> option. Cilk Plus
is an extension to the C and C++ languages to support data and task
parallelism. The present implementation follows ABI version 1.2; all
features but <code>_Cilk_for</code> have been implemented.</li>
</ul>
<h3>C</h3>
<ul>
<li>ISO C11 atomics (the <code>_Atomic</code> type specifier and
qualifier and the <code><stdatomic.h></code> header) are now
supported.</li>
<li>ISO C11 generic selections (<code>_Generic</code> keyword) are
now supported.</li>
<li>ISO C11 thread-local storage (<code>_Thread_local</code>,
similar to GNU C <code>__thread</code>) is now supported.</li>
<li>ISO C11 support is now at a similar level of completeness to ISO
C99 support: substantially complete modulo bugs, extended
identifiers (supported except for corner cases
when <code>-fextended-identifiers</code> is used), floating-point
issues (mainly but not entirely relating to optional C99 features
from Annexes F and G) and the optional Annexes K (Bounds-checking
interfaces) and L (Analyzability).</li>
<li>A new C extension <code>__auto_type</code> provides a subset of
the functionality of C++11 <code>auto</code> in GNU C.</li>
</ul>
<h3 id="cxx">C++</h3>
<ul>
<li>
The G++ implementation of <a href="../projects/cxx1y.html">C++1y</a> return type deduction for normal
functions has been updated to conform to
<a href="http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2013/n3638.html">N3638</a>,
the proposal accepted into the working paper. Most notably, it adds <tt>decltype(auto)</tt> for
getting <tt>decltype</tt> semantics rather than the template argument deduction semantics of plain <tt>auto</tt>:
<blockquote><pre>
int& f();
auto i1 = f(); // int
decltype(auto) i2 = f(); // int&
</pre></blockquote>
</li>
<li>
G++ supports <a href="../projects/cxx1y.html">C++1y</a> lambda capture initializers:
<blockquote><pre>
[x = 42]{ ... };
</pre></blockquote>
Actually, they have been accepted since GCC 4.5, but now the compiler doesn't
warn about them with <tt>-std=c++1y</tt>, and supports parenthesized and
brace-enclosed initializers as well.
</li>
<li>
G++ supports <a href="../projects/cxx1y.html">C++1y</a> variable length
arrays. G++ has supported GNU/C99-style VLAs for a long time, but now
additionally supports initializers and lambda capture by reference. In
C++1y mode G++ will complain about VLA uses that are not permitted by
the draft standard, such as forming a pointer to VLA type or
applying <tt>sizeof</tt> to a VLA variable. Note that it now appears
that VLAs will not be part of C++14, but will be part of a separate
document and then perhaps C++17.
<blockquote><pre>
void f(int n) {
int a[n] = { 1, 2, 3 }; // throws std::bad_array_length if n < 3
[&a]{ for (int i : a) { cout << i << endl; } }();
&a; // error, taking address of VLA
}
</pre></blockquote>
</li>
<li>
G++ supports the <a href="../projects/cxx1y.html">C++1y</a> [[deprecated]]
attribute modulo bugs in the underlying [[gnu::deprecated]] attribute. Classes
and functions can be marked deprecated and a diagnostic message added:
<blockquote><pre>
class A;
int bar(int n);
#if __cplusplus > 201103
class [[deprecated("A is deprecated in C++14; Use B instead")]] A;
[[deprecated("bar is unsafe; use foo() instead")]]
int bar(int n);
int foo(int n);
class B;
#endif
A aa; // warning: 'A' is deprecated : A is deprecated in C++14; Use B instead
int j = bar(2); // warning: 'int bar(int)' is deprecated : bar is unsafe; use foo() instead
</pre></blockquote>
</li>
<li>
G++ supports <a href="../projects/cxx1y.html">C++1y</a> digit separators.
Long numeric literals can be subdivided with a single quote ' to enhance readability:
<blockquote><pre>
int i = 1048576;
int j = 1'048'576;
int k = 0x10'0000;
int m = 0'004'000'000;
int n = 0b0001'0000'0000'0000'0000'0000;
double x = 1.602'176'565e-19;
double y = 1.602'176'565e-1'9;
</pre></blockquote>
</li>
<li>
G++ supports <a href="../projects/cxx1y.html">C++1y</a> polymorphic lambdas.
<blockquote><pre>
// a functional object that will increment any type
auto incr = [](auto x) { return x++; };
</pre></blockquote>
</li>
</ul>
<h4>Runtime Library (libstdc++)</h4>
<ul>
<li><a href="http://gcc.gnu.org/onlinedocs/libstdc++/manual/status.html#status.iso.2011">
Improved support for C++11</a>, including:
<ul>
<li> support for <code><regex></code>;</li>
<li> The associative containers in <code><map></code> and
<code><set></code> and the unordered associative containers
in <code><unordered_map></code> and <code><unordered_set></code>
meet the allocator-aware container requirements; </li>
</ul>
</li>
<li><a href="http://gcc.gnu.org/onlinedocs/libstdc++/manual/status.html#status.iso.2014">
Improved experimental support for the upcoming ISO C++ standard, C++14</a>,
including:
<ul>
<li> fixing <code>constexpr</code> member functions without <code>const</code>; </li>
<li> implementation of the <code>std::exchange()</code> utility function; </li>
<li> addressing tuples by type; </li>
<li> implemention of <code>std::make_unique</code>; </li>
<li> implemention of <code>std::shared_lock</code>; </li>
<li> making <code>std::result_of</code> SFINAE-friendly; </li>
<li> adding <code>operator()</code> to <code>integral_constant</code>; </li>
<li> adding user-defined literals for standard library types
<code>std::basic_string</code>, <code>std::chrono::duration</code>,
and <code>std::complex</code>; </li>
<li> adding two range overloads to non-modifying sequence oprations
<code>std::equal</code> and <code>std::mismatch</code>; </li>
<li> adding IO manipulators for quoted strings; </li>
<li> adding <code>constexpr</code> members to <code><utility></code>,
<code><complex></code>, <code><chrono></code>, and some containers; </li>
<li> adding compile-time <code>std::integer_sequence</code>; </li>
<li> adding cleaner transformation traits; </li>
<li> making <code><functional></code>s operator functors easier to use
and more generic; </li>
</ul>
</li>
<li>An implementation of <code>std::experimental::optional</code>.</li>
<li>An implementation of <code>std::experimental::string_view</code>.</li>
<li>The non-standard function <code>std::copy_exception</code> has been deprecated
and will be removed in a future version. <code>std::make_exception_ptr</code>
should be used instead.
</li>
</ul>
<h3 id="fortran">Fortran</h3>
<ul>
<li>Compatibility notice:
<ul>
<li>Module files: The version of the module files (<code>.mod</code>)
has been incremented; additionally, module files are now compressed.
Fortran <code>MODULE</code>s compiled by earlier GCC versions have
to be recompiled, when they are <code>USE</code>d by files compiled
with GCC 4.9, because GCC 4.9 is not able to read <code>.mod</code>
files of earlier GCC versions; attempting to do so gives an error
message. Note: The ABI of the produced assembler data itself has not
changed: object files and libraries are fully compatible to older
versions. (Except for the next items.)</li>
<li>ABI changes:
<ul>
<li>Note that the <a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gfortran/Argument-passing-conventions.html"
>argument passing ABI</a> has changed for scalar dummy
arguments of type <code>INTEGER</code>, <code>REAL</code>,
<code>COMPLEX</code> and <code>LOGICAL</code>, which have
<em>both</em> the <code>VALUE</code> and the <code>OPTIONAL</code>
attribute.</li>
<li>Due to the support of finalization, the virtual table associated
with polymorphic variables has changed. Therefore, code containing
<code>CLASS</code> should be recompiled, including all files which
define derived types involved in the type definition used by
polymorphic variables. (Note: Due to the incremented module version,
trying to mix old code with new code will usually give an error
message.)</li>
</ul></li>
<li>GNU Fortran no longer deallocates allocatable variables or
allocatable components of variables declared in the main program. Since
Fortran 2008, the standard explicitly states that variables declared
in the Fortran main program automatically have the <code>SAVE</code>
attribute.</li>
<li>When opening files, the close-on-exec flag is set if the system
supports such a feature. This is generally considered good practice
these days, but if there is a need to pass file descriptors to child
processes the parent process must now remember to clear the
close-on-exec flag by calling <code>fcntl()</code>, e.g. via
<code>ISO_C_BINDING</code>, before executing the child process.</li>
</ul></li>
<li>The deprecated command-line option <code>-fno-whole-file</code>
has been removed. (<code>-fwhole-file</code> is the default since
GCC 4.6.) <code>-fwhole-file</code>/<code>-fno-whole-file</code>
continue to be accepted but do not influence the code generation.</li>
<li>The compiler no longer unconditionally warns
about <code>DO</code> loops with zero iterations. This warning is now
controlled by the <code>-Wzerotrips</code> option, which is implied by
<code>-Wall</code>.</li>
<li>The new <code>NO_ARG_CHECK</code> attribute of the <a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gfortran/GNU-Fortran-Compiler-Directives.html"
><code>!GCC$</code> directive</a> can be used to disable the
type-kind-rank (TKR) argument check for a dummy argument. The feature
is similar to ISO/IEC TS 29133:2012's <code>TYPE(*)</code>, except that
it additionally also disables the rank check. Variables with
<code>NO_ARG_CHECK</code> have to be dummy arguments and may only be
used as argument to <code>ISO_C_BINDING</code>'s <code>C_LOC</code>
and as actual argument to another <code>NO_ARG_CHECK</code> dummy
argument; also the other constraints of <code>TYPE(*)</code> apply.
The dummy arguments should be declared as scalar or assumed-size
variable of type <code>type(*)</code> (recommended) – or of
type <code>integer</code>, <code>real</code>, <code>complex</code>
or <code>logical</code>. With <code>NO_ARG_CHECK</code>, a pointer
to the data without further type or shape information is passed,
similar to C's <code>void*</code>. Note that also TS 29113's
<code>type(*),dimension(..)</code> accepts arguments of any type and
rank; contrary to <code>NO_ARG_CHECK</code> assumed-rank arguments
pass an array descriptor which contains the array shape and stride
of the argument.</li>
<li><a href="http://gcc.gnu.org/wiki/Fortran2003Status">Fortran 2003</a>:
<ul>
<li>Finalization is now supported. Note that finalization is currently
only done for a subset of the situations in which it should occur.</li>
<li>Experimental support for <em>scalar</em> character components with
deferred length (i.e. allocatable string length) in derived types has
been added. (Deferred-length character variables are supported since
GCC 4.6.)</li>
</ul></li>
<li><a href="http://gcc.gnu.org/wiki/Fortran2008Status">Fortran 2008</a>:
<ul>
<li>When <code>STOP</code> or <code>ERROR STOP</code> is used to terminate
the execution and any exception (but inexact) is signaling, a warning is
printed to <code>ERROR_UNIT</code>, indicating which exceptions are
signaling. The <code><a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gfortran/Debugging-Options.html"
>-ffpe-summary=</a></code> command-line option can be used to fine-tune
for which exception the warning should be shown.</li>
<li>Rounding on input (<code>READ</code>) is now handled on systems where
<code>strtod</code> honours the rounding mode. (For output, rounding is
supported since GCC 4.5.) Note that for input, the
<code>compatible</code> rounding mode is handled as <code>nearest</code>
(i.e., for a tie, rounding to an even last significant
[cf. IEC 60559:1989] – while <code>compatible</code> rounds away
from zero for a tie).</li>
</ul></li>
</ul>
<h3 id="go">Go</h3>
<ul>
<li>GCC 4.9 provides a complete implementation of the Go 1.2.1
release.</li>
</ul>
<!--
<h3>Java (GCJ)</h3>
-->
<h2 id="targets">New Targets and Target Specific Improvements</h2>
<h3 id="aarch64">AArch64</h3>
<ul>
<li> The ARMv8-A crypto and CRC instructions are now supported through
intrinsics. These are enabled when the architecture supports these
and are available through the <code>-march=armv8-a+crc</code>
and <code>-march=armv8-a+crypto</code> options.
</li>
<li> Initial support for ILP32 has now been added to the
compiler. This is now available through the command line option
<code>-mabi=ilp32</code>. Support for ILP32 is
considered experimental as the ABI specification is still beta.
</li>
<li> Coverage of more of the ISA including the SIMD extensions has
been added. The Advanced SIMD intrinsics have also been improved.
</li>
<li> The new local register allocator (LRA) is now on by default
for the AArch64 backend.
</li>
<li> The REE (Redundant extension elimination) pass has now been enabled
by default for the AArch64 backend.
</li>
<li> Tuning for the Cortex-A53 and Cortex-A57 has been improved.
</li>
<li> Initial big.LITTLE tuning support for the combination of Cortex-A57
and Cortex-A53 was added through the <code>-mcpu=cortex-a57.cortex-a53
</code> option.
</li>
<li> A number of structural changes have been made to both the ARM
and AArch64 backends to facilitate improved code-generation.
</li>
</ul>
<h3 id="arm">ARM</h3>
<ul>
<li> Use of Advanced SIMD (Neon) for 64-bit scalar computations has been
disabled by default. This was found to generate better code in only
a small number of cases. It can be turned back on with the
<code>-mneon-for-64bits</code> option.
</li>
<li> Further support for the ARMv8-A architecture, notably implementing
the restriction around IT blocks in the Thumb32 instruction set has
been added. The <code>-mrestrict-it</code> option can be used with
<code>-march=armv7-a</code> or the <code>-march=armv7ve</code> options
to make code generation fully compatible with the deprecated instructions
in ARMv8-A.
</li>
<li> Support has now been added for the ARMv7ve variant of the
architecture. This can be used by the <code>-march=armv7ve</code> option.
</li>
<li> The ARMv8-A crypto and CRC instructions are now supported through
intrinsics and are available through the <code>-march=armv8-a+crc</code>
and <code>mfpu=crypto-neon-fp-armv8</code> options.
</li>
<li> LRA is now on by default for the ARM target. This can be turned off
using the <code>-mno-lra</code> option. This option is purely
transitionary command line option and will be removed in a future
release. We are interested in any bug reports regarding functional and
performance regressions with LRA.
</li>
<li> A new option <code>-mslow-flash-data</code> to improve performance
of programs fetching data on slow flash memory has now been
introduced for the ARMv7-M profile cores.
</li>
<li> A new option <code>-mpic-data-is-text-relative</code> for targets
that allows data segments to be relative to text segments has
been added. This is on by default for all targets except VxWorks RTP.
</li>
<li> A number of infrastructural changes have been made to both the ARM
and AArch64 backends to facilitate improved code-generation.
</li>
<li> GCC now supports Cortex-A12 and the Cortex-R7 through the
<code>-mcpu=cortex-a12</code> and <code>-mcpu=cortex-r7</code> options.
</li>
<li> GCC now has tuning for the Cortex-A57 and Cortex-A53
through the <code>-mcpu=cortex-a57</code> and <code>-mcpu=cortex-a53
</code> options.
</li>
<li> Initial big.LITTLE tuning support for the combination of Cortex-A57
and Cortex-A53 was added through the <code>-mcpu=cortex-a57.cortex-a53
</code> option. Similar support was added for the combination of
Cortex-A15 and Cortex-A7 through the <code>-mcpu=cortex-a15.cortex-a7
</code> option.
</li>
<li> Further performance optimizations for the Cortex-A15 and the
Cortex-M4 have been added.
</li>
<li>A number of code generation improvements for Thumb2 to reduce code
size when compiling for the M-profile processors.
</li>
</ul>
<h3 id="x86">IA-32/x86-64</h3>
<ul>
<li><code>-mfpmath=sse</code> is now implied by <code>-ffast-math</code>
on all targets where SSE2 is supported.</li>
<li>Intel AVX-512 support was added to GCC. That includes inline
assembly support, new registers and extending existing ones,
new intrinsics (covered by corresponding testsuite), and basic
autovectorization. AVX-512 instructions are available via
the following GCC switches: AVX-512 foundation instructions:
<code>-mavx512f</code>, AVX-512 prefetch instructions: <code>-mavx512pf</code>,
AVX-512 exponential and reciprocal instructions: <code>-mavx512er</code>,
AVX-512 conflict detection instructions: <code>-mavx512cd</code>.
</li>
<li> It is now possible to call x86 intrinsics from select functions in
a file that are tagged with the corresponding target attribute without
having to compile the entire file with the <code>-mxxx</code> option.
This improves the usability of x86 intrinsics and is particularly useful
when doing <a
href="http://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Function-Multiversioning.html"
>Function Multiversioning</a>.
</li>
<li>GCC now supports the new Intel microarchitecture named Silvermont
through <code>-march=silvermont</code>.
</li>
<li>GCC now supports the new Intel microarchitecture named Broadwell
through <code>-march=broadwell</code>.
</li>
<li>Optimizing for other Intel microarchitectures have been renamed
to <code>-march=nehalem</code>, <code>westmere</code>,
<code>sandybridge</code>, <code>ivybridge</code>,
<code>haswell</code>, <code>bonnell</code>.
</li>
<li><code>-march=generic</code> has been retuned for better support of
Intel core and AMD Bulldozer architectures. Performance of AMD K7, K8,
Intel Pentium-M, and Pentium4 based CPUs is no longer considered important
for generic.
</li>
<li><code>-mtune=intel</code> can now be used to generate code running
well on the most current Intel processors, which are Haswell
and Silvermont for GCC 4.9.
</li>
<li>Support to encode 32-bit assembly instructions in 16-bit format
is now available through the <code>-m16</code> command-line option.
</li>
<li>Better inlining of <code>memcpy</code> and <code>memset</code>
that is aware of value ranges and produces shorter alignment prologues.
</li>
<li><code>-mno-accumulate-outgoing-args</code> is now honored when unwind
information is output. Argument accumulation is also now turned off
for portions of programs optimized for size.</li>
<li>Support for new AMD family 15h processors (Excavator core)
is now available through the <code>-march=bdver4</code> and
<code>-mtune=bdver4</code> options.</li>
</ul>
<h3 id="msp430">MSP430</h3>
<ul>
<li>A new command-line option <code>-mcpu=</code> has been added to the MSP430 backend.
This option is used to specify the ISA to be used. Accepted values are
<code>msp430</code> (the default), <code>msp430x</code> and <code>msp430xv2</code>. The ISA is no longer deduced
from the <code>-mmcu=</code> option as there are far too many different MCU names. The
<code>-mmcu=</code> option is still supported, and this is still used to select linker
scripts and generate a C preprocessor symbol that will be recognised by the
<code>msp430.h</code> header file.</li>
</ul>
<h3 id="nds32">NDS32</h3>
<ul>
<li> A new nds32 port supports the 32-bit architecture from Andes
Technology Corporation.</li>
<li> The port provides initial support for the V2, V3, V3m
instruction set architectures.</li>
</ul>
<h3 id="nios2">Nios II</h3>
<ul>
<li> A port for the Altera Nios II has been contributed by
Mentor Graphics.</li>
</ul>
<h3 id="powerpc">PowerPC / PowerPC64 / RS6000</h3>
<ul>
<li>GCC now supports Power ISA 2.07, which includes support for Hardware
Transactional Memory (HTM), Quadword atomics and several VMX and VSX
additions, including Crypto, 64-bit integer, 128-bit integer and
decimal integer operations.</li>
<li>Support for the POWER8 processor is now available through the
<code>-mcpu=power8</code> and <code>-mtune=power8</code> options.</li>
<li>The libitm library has been modified to add a HTM fastpath that
automatically uses POWER's HTM hardware instructions when it is
executing on a HTM enabled processor.</li>
<li>Support for the new powerpc64le-linux platform has been added.
It defaults to generating code that conforms to the ELFV2 ABI.</li>
</ul>
<h3>S/390, System z</h3>
<ul>
<li>Support for the Transactional Execution Facility included with
the IBM zEnterprise zEC12 processor has been added. A set of
GCC style builtins as well as XLC style builtins are provided.
The builtins are enabled by default when using
the <code>-march=zEC12</code> option but can explicitly be
disabled with <code>-mno-htm</code>.
Using the GCC builtins also libitm supports hardware
transactions on S/390.</li>
<li>The hotpatch features allows to prepare functions for
hotpatching. A certain amount of bytes is reserved before the
function entry label plus a NOP is inserted at its very
beginning to implement a backward jump when applying a patch.
The feature can either be enabled via command line
option <code>-mhotpatch</code> for a compilation unit or can be
enabled per function using the <code>hotpatch</code>
attribute.</li>
<li>The shrink wrap optimization is now supported on S/390 and
enabled by default.</li>
<li>A major rework of the routines to determine which registers
need to be saved and restored in function prologue/epilogue now
allow to use floating point registers as save slots. This will
happen for certain leaf function with <code>-march=z10</code>
or higher.</li>
<li>The LRA rtl pass replaces reload by default on S/390.</li>
</ul>
<h3 id="rx">RX</h3>
<ul>
<li> The port now allows to specify the RX100, RX200, and RX600 processors
with the command line options -mcpu=rx100, -mcpu=rx200 and -mcpu=rx600.
</li>
</ul>
<h3 id="sh">SH</h3>
<ul>
<li>Minor improvements to code generated for integer arithmetic and code
that involves the T bit.</li>
<li>Added support for the SH2A <code>clips</code> and <code>clipu</code>
instructions. The compiler will now try to utilize them for min/max
expressions such as <code>max (-128, min (127, x))</code>.</li>
<li>Added support for the <code>cmp/str</code> instruction through built-in
functions such as <code>__builtin_strlen</code>. When not optimizing for
size, the compiler will now expand calls to e.g. <code>strlen</code> as an
inlined sequences which utilize the <code>cmp/str</code> instruction.</li>
<li>Improved code generated around volatile memory loads and stores.</li>
<li>The option <code>-mcbranchdi</code> has been deprecated. Specifying it
will result in a warning and will not influence code generation.</li>
<li>The option <code>-mcmpeqdi</code> has been deprecated. Specifying it
will result in a warning and will not influence code generation.</li>
</ul>
<!--
<h2>Documentation improvements</h2>
-->
<!--
<h2>Other significant improvements</h2>
-->
<!-- ==================================================================== -->
<div class="copyright">
<address style="margin-top:0;">For questions related to the use of GCC,
please consult these web pages and the
<a href="http://gcc.gnu.org/onlinedocs/">GCC manuals</a>. If that fails,
the <a href="mailto:gcc-help@gcc.gnu.org">gcc-help@gcc.gnu.org</a>
mailing list might help.
Comments on these web pages and the development of GCC are welcome on our
developer list at <a href="mailto:gcc@gcc.gnu.org">gcc@gcc.gnu.org</a>.
All of <a href="http://gcc.gnu.org/lists.html">our lists</a>
have public archives.
</address>
<p>Copyright (C)
<a href="http://www.fsf.org">Free Software Foundation, Inc.</a>
Verbatim copying and distribution of this entire article is
permitted in any medium, provided this notice is preserved.</p>
<p style="margin-bottom:0;">These pages are
<a href="http://gcc.gnu.org/about.html">maintained by the GCC team</a>.
Last modified 2014-04-22<!-- IGNORE DIFF
--><a href="http://validator.w3.org/check/referer">.</a></p>
</div>
<!-- ==================================================================== -->
</body>
</html>
|