llvm-svn: 223691

Summary:
Store the result matcher after the first call and reuse it later on.
Recreating the matchers just to use them once incurs in a lot of
unnecessary temporary memory allocations.
This change speeds up our clang-tidy benchmarks by ~2%.

Reviewers: klimek

Subscribers: klimek, cfe-commits

Differential Revision: http://reviews.llvm.org/D6564

llvm-svn: 223690

llvm-svn: 223689

Before:
  @Mock DataLoader loooooooooooooooooooooooader =
      aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa;

After:
  @Mock
  DataLoader loooooooooooooooooooooooader =
      aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa;

llvm-svn: 223688

Differential Revision: http://reviews.llvm.org/D6567

llvm-svn: 223687

There's nothing else to bother trying to shrink these.

llvm-svn: 223686

llvm-svn: 223685

A zero sized array is zero sized and might share its address with
another global.

llvm-svn: 223684

The logic for lowering profiling counters has been moved to an LLVM
pass. Emit the intrinsics rather than duplicating the whole pass in
clang.

llvm-svn: 223683

llvm-svn: 223682

We were already lazily linking functions, but all GlobalValues can be treated
uniformly for this.

The test updates are to ensure that a given GlobalValue is still linked in.

This fixes pr21494.

llvm-svn: 223681

llvm-svn: 223680

This reverts r223624 with a small tweak, hopefully this will make stage3
equivalent.

llvm-svn: 223679

Thanks to Rafael Espindola for testing assistance.

llvm-svn: 223678

dependencies for the KS tutorials

llvm-svn: 223677

Fix a compact unwind encoding logic bug which would try to encode
more callee saved registers than it should, leading to early bail out
in the encoding logic and abusive use of DWARF frame mode unnecessarily.

Also remove no-compact-unwind.ll which was testing the wrong thing
based on this bug and move it to valid 'compact unwind' tests. Added
other few more tests too.

llvm-svn: 223676

make all doesn't build the examples and it was uniquified since
last build.

llvm-svn: 223675

There was an error in ORing mask which is used for getting a list of variables.
Previously, these constants were unnamed, and possible it become the reason of this
bug. Also added test case for -stack-list-local and -stack-list_arguments.

Patch from Ilia K <ki.stfu@gmail.com>.

llvm-svn: 223674

llvm-svn: 223673

Introduce the ``llvm.instrprof_increment`` intrinsic and the
``-instrprof`` pass. These provide the infrastructure for writing
counters for profiling, as in clang's ``-fprofile-instr-generate``.

The implementation of the instrprof pass is ported directly out of the
CodeGenPGO classes in clang, and with the followup in clang that rips
that code out to use these new intrinsics this ends up being NFC.

Doing the instrumentation this way opens some doors in terms of
improving the counter performance. For example, this will make it
simple to experiment with alternate lowering strategies, and allows us
to try handling profiling specially in some optimizations if we want
to.

Finally, this drastically simplifies the frontend and puts all of the
lowering logic in one place.

llvm-svn: 223672

a description of how to add debug information using DWARF and
DIBuilder to the language.

Thanks to David Blaikie for his assistance with this tutorial.

llvm-svn: 223671

llvm-svn: 223670

llvm-svn: 223669

Teach ISel how to match a TZCNT/LZCNT from a conditional move if the
condition code is X86_COND_NE.
Existing tablegen patterns only allowed to match TZCNT/LZCNT from a
X86cond with condition code equal to X86_COND_E. To avoid introducing
extra rules, I added an 'ImmLeaf' definition that checks if the
condition code is COND_E or COND_NE.

llvm-svn: 223668

llvm-svn: 223667

Since the main file was empty, we can just copy the content of the Input file
into it.

llvm-svn: 223666

This is just testing the largest merge mode for comdats. No need to use
hard to read names and fancy types.

llvm-svn: 223665

llvm-svn: 223664

llvm-svn: 223663

Reviewers: vmedic

Reviewed By: vmedic

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D6213

llvm-svn: 223662

It might be implicitly included by <pthread.h> (and that's why it worked
so far), but it's not guaranteed (for example, this is not the case with
newlib).

llvm-svn: 223661

Before this patch, the backend sub-optimally expanded the non-constant shift
count of a v8i16 shift into a sequence of two 'movd' plus 'movzwl'.

With this patch the backend checks if the target features sse4.1. If so, then
it lets the shuffle legalizer deal with the expansion of the shift amount.

Example:
;;
define <8 x i16> @test(<8 x i16> %A, <8 x i16> %B) {
  %shamt = shufflevector <8 x i16> %B, <8 x i16> undef, <8 x i32> zeroinitializer
  %shl = shl <8 x i16> %A, %shamt
  ret <8 x i16> %shl
}
;;

Before (with -mattr=+avx):
  vmovd  %xmm1, %eax
  movzwl  %ax, %eax
  vmovd  %eax, %xmm1
  vpsllw  %xmm1, %xmm0, %xmm0
  retq

Now:
  vpxor  %xmm2, %xmm2, %xmm2
  vpblendw  $1, %xmm1, %xmm2, %xmm1
  vpsllw  %xmm1, %xmm0, %xmm0
  retq

llvm-svn: 223660

llvm-svn: 223659

llvm-svn: 223658

llvm-svn: 223657

It would crash when the function was lazy linked.

llvm-svn: 223656

Introduce install_symlink.cmake from clang/tools/driver/clang_symlink.cmake.

FIXME: Would it be generalized?
llvm-svn: 223655

* Use a range loop.
* Move simple continue checks earlier.
* clang-format.

llvm-svn: 223654

Just less code duplication.

llvm-svn: 223653

Before:
  auto next_pair = [](A * a) -> pair<A*, A*>{};

After:
  auto next_pair = [](A* a) -> pair<A*, A*>{};

llvm-svn: 223652