This removes the following decorators:

* skipIfI386
* expectedFailureI386
* expectedFailurex86_64
* skipIfArch
* skipUnlessArch
* skipUnlessI386

And other related decorators.  All code using those decorators
is updated to use expectedFailureAll and skipIf

llvm-svn: 260178

llvm-svn: 260177

Convert everything over to using skipIf.

llvm-svn: 260176

llvm-svn: 260175

IndVarSimplify assumes scAddRecExpr to be expanded in literal form instead of
canonical form by calling disableCanonicalMode after it creates SCEVExpander.
When CanonicalMode is disabled, SCEVExpander::expand should always return PHI
node for scAddRecExpr. r259736 broke the assumption.

The fix is to let SCEVExpander::expand skip the reuse Value logic if
CanonicalMode is false.

In addition, Besides IndVarSimplify, LSR pass also calls disableCanonicalMode
before doing rewrite. We can remove the original check of LSRMode in reuse
Value logic and use CanonicalMode instead.

llvm-svn: 260174

llvm-svn: 260173

llvm-svn: 260172

r260146 | xur | 2016-02-08 13:07:46 -0800 (Mon, 08 Feb 2016) | 13 lines
[PGO] Differentiate Clang instrumentation and IR level instrumentation profiles

llvm-svn: 260170

Summary:
Unrolling Analyzer is already pretty complicated, and it becomes harder and harder to exercise it with usual IR tests, as with them we can only check the final decision: whether the loop is unrolled or not. This change factors this framework out from LoopUnrollPass to analyses, which allows to use unit tests.
The change itself is supposed to be NFC, except adding a couple of tests.

I plan to add more tests as I add new functionality and find/fix bugs.

Reviewers: chandlerc, hfinkel, sanjoy

Subscribers: zzheng, sanjoy, llvm-commits

Differential Revision: http://reviews.llvm.org/D16623

llvm-svn: 260169

As discussed on PR26491, this patch adds support for lowering v4f32 shuffles to the MOVLHPS/MOVHLPS instructions. It also adds support for memory folding with their MOVLPS/MOVHPS load equivalents.

This first patch only really helps SSE1 targets as SSE2+ targets will widen the shuffle mask and use v2f64 equivalents (although they still combine to MOVLHPS/MOVHLPS for v2f64 splats). This will have to be addressed in a future patch, most likely when we add support for binary target shuffle combines.

Differential Revision: http://reviews.llvm.org/D16956

llvm-svn: 260168

llvm-svn: 260167

llvm-svn: 260166

llvm-svn: 260165

Differential Revision: http://reviews.llvm.org/D16831

llvm-svn: 260164

llvm-svn: 260163

In order for recent gcov versions to read the coverage data, you have
to use UseCfgChecksum=true and FunctionNamesInData=false options for
coverage profiling pass. This is because gcov is expecting the
function section in .gcda to be exactly 3 words in size, containing
ident and two checksums.

While llvm-cov is compatible with UseCfgChecksum=true, it always
expects a function name in .gcda function sections (it's not
compatible with FunctionNamesInData=false). Thus it's currently
impossible to generate one set of coverage files that works with both
gcov and llvm-cov.

This change fixes the reading of coverage information to only read the
function name if it's present.

Patch by Arseny Kapoulkine. Thanks!

llvm-svn: 260162

llvm-svn: 260161

This patch adds doxygen comments for all the intrinsincs in the header file pmmintrin.h. The doxygen comments are automatically generated based on Sony's intrinsics document.

Differential Revision: http://reviews.llvm.org/D16913

llvm-svn: 260160

For an explicit specialization, we first build a FunctionDecl, and then
we call SubstDecl() on it to build a second FunctionDecl, which has the
first FunctionDecl as canonical decl.

The address of an explicit specialization of function template used to be the
canonical decl of the FunctionDecl.  This is different from all the other
DeduceTemplateArguments() calls in SemaOverload, and since the canonical decl
isn't visited by ParentMap while the redecl is, it also made ParentMap assert
when computing the parent of a address-of-explicit-specialization-fun-template.

To fix, remove the getCanonicalDecl() call.  No behavior difference for clang,
but it fixes an assert in ParentMap (which is e.g. used by libTooling).

llvm-svn: 260159

because the information is now stored in TemplateDecl.

llvm-svn: 260155

Apparently option is for bools and cmake-gui will display this
strangely with option.

Pointed out by edward-san - thanks!

llvm-svn: 260154

These includes were before the #ifndef/#define guards.  Moving them to
where they should be.

llvm-svn: 260153

llvm-svn: 260152

llvm-svn: 260151

Sync InstrProfData.inc with the one in llvm.

llvm-svn: 260148

Patch by Michael Stapelberg.

llvm-svn: 260147

This patch uses one bit in profile version to differentiate Clang
instrumentation and IR level instrumentation profiles.

PGOInstrumenation generates a COMDAT variable __llvm_profile_raw_version so
that the compiler runtime can set the right profile kind.
PGOInstrumenation now checks this bit to make sure it's an IR level
instrumentation profile.

Differential Revision: http://reviews.llvm.org/D15540

llvm-svn: 260146

Another opportunity to reduce masked stores: in D16691, we decided not to attempt the 'one mask element is set'
transform in InstCombine, but this should be a win for any AVX machine.

Code comments note that this transform could be extended for other targets / cases.

Differential Revision: http://reviews.llvm.org/D16828

llvm-svn: 260145

Mehdi suggested in a review of r259766 that it's also useful to easily
set the type of LTO. Augment the cmake variable to support that.

llvm-svn: 260143

llvm-svn: 260142

llvm-svn: 260138

Summary:
We will hit this once we have enabled uniform branches.  The
smrd-vccz-bug.ll test will be added with the uniform branch commit.

Reviewers: mareko, arsenm

Subscribers: arsenm, llvm-commits

Differential Revision: http://reviews.llvm.org/D16725

llvm-svn: 260137

llvm-svn: 260136

* Change the `not_in` function to be called `no_match`.  This makes
  it clear that keyword arguments can be more than just lists.
* Change the name of `_check_list_or_lambda` to
  `_match_decorator_property`.  Again clarifying that decorator params
  are not always lists.
* Always use a regex match when matching strings.  This allows automatic
  support for regex matching on all decorator properties.  Also support
  compiled regex values.
* Fix a bug in the compiler check used by _decorateTest.  The two
  arguments were reversed, the condition was always wrong.
* Change one test that uses skipUnlessArch to use skipIf, to
  demonstrate that skipIf can now handle more scenarios.

Differential Revision: http://reviews.llvm.org/D16938

llvm-svn: 260135

expectedFailureWindows is equivalent to using the general
expectedFailureAll decorator with oslist="windows".  Additionally,
by moving towards these common decorators we can solve the issue
of having to support decorators that can be called with or without
arguments.  Once all decorators are always called with arguments,
and this is enforced by design (because you can't specify the condition
you're decorating for without passing an argument) the implementation
of the decorators can become much simpler

Differential Revision: http://reviews.llvm.org/D16936

llvm-svn: 260134

This matches GCC and MSVC's behaviour, and saves on code size.

We were already not extending i1 return values on x86_64 after r127766. This
takes that patch further by applying it to x86 target as well, and also for i8
and i16.

The ABI docs have been unclear about the required behaviour here. The new i386
psABI [1] clearly states (Table 2.4, page 14) that i1, i8, and i16 return
vales do not need to be extended beyond 8 bits. The x86_64 ABI doc is being
updated to say the same [2].

Differential Revision: http://reviews.llvm.org/D16907

 [1]. https://01.org/sites/default/files/file_attach/intel386-psabi-1.0.pdf
 [2]. https://groups.google.com/d/msg/x86-64-abi/E8O33onbnGQ/_RFWw_ixDQAJ

llvm-svn: 260133

llvm-svn: 260132

The accumulator in multiply-and-subtract instructions is actually subtracted
*from* so these patterns were computing the wrong value.

llvm-svn: 260131

llvm-svn: 260130

When handling 'if' statements, we crash if the condition and the consequent
branch are spanned by a single macro expansion.

The crash occurs because of a sanity 'reset' in popRegions(): if an expansion
exactly spans an entire region, we set MostRecentLocation to the start of the
expansion (its 'include location'). This ensures we don't handleFileExit()
ourselves out of the expansion before we're done processing all of the regions
within it. This is tested in test/CoverageMapping/macro-expressions.c.

This causes a problem when an expansion spans both the condition and the
consequent branch of an 'if' statement. MostRecentLocation is updated to the
start of the 'if' statement in popRegions(), so the file for the expansion
isn't exited by the time we're done handling the statement. We then crash with
'fatal: File exit not handled before popRegions'.

The fix for this is to detect these kinds of expansions, and conservatively
update MostRecentLocation to the end of expansion region containing the
conditional. I've added tests to make sure we don't have the same problem with
other kinds of statements.

rdar://problem/23630316

Differential Revision: http://reviews.llvm.org/D16934

llvm-svn: 260129