llvm-svn: 327319

https://reviews.llvm.org/D42582

Patch from Leandro Lupori.

llvm-svn: 327318

Patch by David Carlier!

Differential Revision: https://reviews.llvm.org/D44069

llvm-svn: 327317

Differential Revision: https://reviews.llvm.org/D44355

llvm-svn: 327316

getNumUses is a linear time operation. It traverses the user linked list to the end and counts as it goes. Since we are only interested in small constant counts, we should use hasNUses or hasNUsesMore more that terminate the traversal as soon as it can provide the answer.

There are still two other locations in InstCombine, but changing those would force a rebase of D44266 which if accepted would remove them.

Differential Revision: https://reviews.llvm.org/D44398

llvm-svn: 327315

[CallSiteSplitting] Use !Instruction::use_empty instead of checking for a non-zero return from getNumUses

getNumUses is a linear operation. It walks a linked list to get a count. So in this case its better to just ask if there are any users rather than how many.

llvm-svn: 327314

llvm-svn: 327313

llvm-svn: 327312

llvm-svn: 327311

llvm-svn: 327310

rdar://38383753

Differential Revision: https://reviews.llvm.org/D44228

llvm-svn: 327309

llvm-svn: 327308

This is the FP equivalent of D42818. Use it for the few cases in InstSimplify 
with -0.0 folds (that's the only current use of m_NegZero()).

Differential Revision: https://reviews.llvm.org/D43792

llvm-svn: 327307

I love you llvm-mca.

llvm-svn: 327306

llvm-svn: 327305

After two failed attempts last week to make this work I am
going back to a known good method of making this test pass on
macOS...adding the current apple-clang version to the
UNSUPPORTED list.

During a previous patch review (https://reviews.llvm.org/D44103)
it was suggested to just XFAIL libcpp-no-deduction-guides
as was done to iter_alloc_deduction.pass.cpp. However
this caused a an unexpected pass on:
http://lab.llvm.org:8011/builders/libcxx-libcxxabi-x86_64-linux-ubuntu-gcc-tot-latest-std/builds/214

I then attempted to just mark libcpp-no-deduction-guides
as UNSUPPORTED, however this caused an additional bot
failure.  So I reverted everything (https://reviews.llvm.org/rCXX327191).

To solve this and get work unblocked I am adding
apple-clang-9 to the original UNSUPPORTED list.

llvm-svn: 327304

Summary:
1) Make sure to discard dangling debug info if the variable (or
variable fragment) is mapped to something new before we had a
chance to resolve the dangling debug info.

2) When resolving debug info, make sure to bump the associated
SDNodeOrder to ensure that the DBG_VALUE is emitted after the
instruction that defines the value used in the DBG_VALUE.
This will avoid a debug-use before def scenario as seen in
https://bugs.llvm.org/show_bug.cgi?id=36417.

The new test case, test/DebugInfo/X86/sdag-dangling-dbgvalue.ll,
show some other limitations in how dangling debug info is
handled in the SelectionDAG. Since we currently only support
having one dangling dbg.value per Value, we will end up dropping
debug info when there are more than one variable that is described
by the same "dangling value".

Reviewers: aprantl

Reviewed By: aprantl

Subscribers: aprantl, eraman, llvm-commits, JDevlieghere

Tags: #debug-info

Differential Revision: https://reviews.llvm.org/D44369

llvm-svn: 327303

This adds two features: "packets", and "nvj".

Enabling "packets" allows the compiler to generate instruction packets,
while disabling it will prevent it and disable all optimizations that
generate them. This feature is enabled by default on all subtargets.
The feature "nvj" allows the compiler to generate new-value jumps and it
implies "packets". It is enabled on all subtargets.

The exception is made for packets with endloop instructions, since they
require a certain minimum number of instructions in the packets to which
they apply. Disabling "packets" will not prevent hardware loops from
being generated.

llvm-svn: 327302

Summary:
This pattern came up in PR36682:
https://bugs.llvm.org/show_bug.cgi?id=36682
https://godbolt.org/g/LhuD9A

Tests for proposed fix in D44367.

Looking at the IR pattern in question, as per [[ https://github.com/rutgers-apl/alive-nj | alive-nj ]], for all the type combinations i checked
(input: `i16`, `i32`, `i64`; intermediate: `half`/`i16`, `float`/`i32`, `double`/`i64`)
for the following `icmp` comparisons the `sitofp`+`bitcast` can be dropped:
* `eq 0`
* `ne 0`
* `slt 0`
* `sle 0`
* `sge 0`
* `sgt 0`
* `slt 1`
* `sge 1`
* `sle -1`
* `sgt -1`
I did not check vectors, but i'm guessing it's the same there.
{F5887419}

Thus all these cases are in the testcase (along with the vector variant with additional `undef` element in the middle).
There are no negative patterns here (unless alive-nj lied/is broken), all of these should be optimized.

Generated with {F5887551}

Reviewers: spatel, majnemer, efriedma, arsenm

Reviewed By: spatel

Subscribers: nlopes, wdng, llvm-commits

Differential Revision: https://reviews.llvm.org/D44390

llvm-svn: 327301

MMX buildvectors were improved at rL327247 - new MMX bugs should be raised on bugzilla

llvm-svn: 327300

See bug 36558: https://bugs.llvm.org/show_bug.cgi?id=36558

Differential Revision: https://reviews.llvm.org/D43950

Reviewers: artem.tamazov, arsenm
llvm-svn: 327299

[X86] Add all of the MRM_C0-MRM_FF forms to the switch in RecognizableInstr::emitInstructionSpecifier. NFC

Remove the special casing for MRM_F8 by using HANDLE_OPTIONAL.

This should be NFC as the forms that were missing aren't used by any instructions today. They exist in the enum so that we didn't have to put them in one at a time when instructions are added. But looks like we failed here.

llvm-svn: 327298

Summary:
This is a new version of D44261, which broke some builds with older gcc, as
they can't align on a constexpr, but rather require an integer (see
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56859) among others.

We introduce `SANITIZER_CACHE_LINE_SIZE` in `sanitizer_platform.h` to be
used in `ALIGNED` attributes instead of using directly `kCacheLineSize`.

Reviewers: alekseyshl, thakis

Reviewed By: alekseyshl

Subscribers: kubamracek, delcypher, #sanitizers, llvm-commits

Differential Revision: https://reviews.llvm.org/D44326

llvm-svn: 327297

These are all global, so prefix with 'J' to help prevent accidental name clashes with other models.

llvm-svn: 327296

There was a missing newline in the docs, and a static_assert that needed
to be a normal assert.

llvm-svn: 327295

ARC."

This reverts commit r327206 as there were test failures caused by this
patch.

http://lists.llvm.org/pipermail/cfe-commits/Week-of-Mon-20180312/221427.html

llvm-svn: 327294

Reviewers: ilya-biryukov

Reviewed By: ilya-biryukov

Subscribers: klimek, jkorous-apple, cfe-commits, ioeric

Differential Revision: https://reviews.llvm.org/D44294

llvm-svn: 327293

MVT belongs to the CodeGen layer, but ShuffleDecode is used by the X86 InstPrinter which is part of the MC layer. This only worked because MVT is completely implemented in a header file with no other library dependencies.

Differential Revision: https://reviews.llvm.org/D44353

llvm-svn: 327292

Since the enqueued kernels have internal linkage, their names may be dropped.
In this case, give them unique names __amdgpu_enqueued_kernel or
__amdgpu_enqueued_kernel.n where n is a sequential number starting from 1.

Differential Revision: https://reviews.llvm.org/D44322

llvm-svn: 327291

This bug was found by accident while trying to expand out testcases
for imported symbols, and is covered by the additional test case.

Differential Revision: https://reviews.llvm.org/D44331

llvm-svn: 327290

This allows the single resource classes (VarBlend, MPSAD, VarVecShift) to use the JWriteResFpuPair macro.

llvm-svn: 327289

See bug 36572: https://bugs.llvm.org/show_bug.cgi?id=36572

Differential Revision: https://reviews.llvm.org/D44020

Reviewers: artem.tamazov, vpykhtin
llvm-svn: 327288

llvm-svn: 327287

This matches the existing ordering that's been there for globals
for a while (__stack_pointer coming first).

Differential Revision: https://reviews.llvm.org/D44333

llvm-svn: 327286

Summary:
Previously, clang-format would detect the following as an
Objective-C block type:

  FOO(^);

when it actually must be a C or C++ macro dealing with an XOR
statement or an XOR operator overload.

According to the Clang Block Language Spec:

https://clang.llvm.org/docs/BlockLanguageSpec.html

block types are of the form:

  int (^)(char, float)

and block variables of block type are of the form:

  void (^blockReturningVoidWithVoidArgument)(void);
  int (^blockReturningIntWithIntAndCharArguments)(int, char);
  void (^arrayOfTenBlocksReturningVoidWithIntArgument[10])(int);

This tightens up the detection so we don't unnecessarily detect
C macros which pass in the XOR operator.

Depends On D43904

Test Plan: New tests added. Ran tests with:
  make -j12 FormatTests &&
  ./tools/clang/unittests/Format/FormatTests

Reviewers: krasimir, jolesiak, djasper

Reviewed By: djasper

Subscribers: djasper, cfe-commits, klimek

Differential Revision: https://reviews.llvm.org/D43906

llvm-svn: 327285

Summary:
Previously, clang-format would detect C++11 and C++17 attribute
specifiers like the following as Objective-C method invocations:

  [[noreturn]];
  [[clang::fallthrough]];
  [[noreturn, deprecated("so sorry")]];
  [[using gsl: suppress("type")]];

To fix this, I ported part of the logic from
tools/clang/lib/Parse/ParseTentative.cpp into TokenAnnotator.cpp so we
can explicitly parse and identify C++11 attribute specifiers.

This allows the guessLanguage() and getStyle() APIs to correctly
guess files containing the C++11 attribute specifiers as C++,
not Objective-C.

Test Plan: New tests added. Ran tests with:
  make -j12 FormatTests && ./tools/clang/unittests/Format/FormatTests

Reviewers: krasimir, jolesiak, djasper

Reviewed By: djasper

Subscribers: aaron.ballman, cfe-commits, klimek

Differential Revision: https://reviews.llvm.org/D43902

llvm-svn: 327284

These are single pipe and have the default resource/uop counts like JWriteResFpuPair so there's no need to handle them separately.

llvm-svn: 327283

Summary:
The new implementation attaches notes to diagnostic message and shows
the original diagnostics in the message of the note.

Reviewers: hokein, ioeric, sammccall

Reviewed By: sammccall

Subscribers: klimek, mgorny, cfe-commits, jkorous-apple

Differential Revision: https://reviews.llvm.org/D44142

llvm-svn: 327282

llvm-svn: 327281

llvm-svn: 327280