__has_attribute now understands target-specific attributes. So when you ask whether an ARM target has the "interrupt" attribute, it will return true for ARM and MSP430 targets, and false for others.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198897 91177308-0d34-0410-b5e6-96231b3b80d8

Removing the notion of TargetAttributesSema and replacing it with one where the parsed attributes are responsible for knowing their target-specific nature, instead of letting Sema figure it out. This is necessary so that __has_attribute can eventually determine whether a parsed attribute applies to the given target or not.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198896 91177308-0d34-0410-b5e6-96231b3b80d8

The commit added a source location that used to be missing from the AST.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198892 91177308-0d34-0410-b5e6-96231b3b80d8

I introduced this bug in 198815. Thanks for Mark Lacey for noticing.
Unfortunately, I have no idea how to test this code.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198891 91177308-0d34-0410-b5e6-96231b3b80d8

Preserves the setting of -fretain-comments-from-system-headers when
building/saving/loading module files. This allows code completion to pick up
documentation comments from system modules.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198890 91177308-0d34-0410-b5e6-96231b3b80d8

Fixes <rdar://problem/15596883>

In ARC, __attribute__((objc_precise_lifetime)) guarantees that the
object stored in it will survive to the end of the variable's formal
lifetime.  It is therefore useful even if it completely unused.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198888 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198887 91177308-0d34-0410-b5e6-96231b3b80d8

To declare or define reserved identifers is undefined behaviour in standard
C++. This needs to be addressed in compiler-rt before it can be used in LLVM.

See the list discussion for details.

This reverts commit r198858.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198885 91177308-0d34-0410-b5e6-96231b3b80d8

Removing a bit of custom parsing functionality used by the thread safety analysis APIs. Now using tablegen to determine whether an attribute's arguments should be parsed in an unevaluated context instead of relying on a separate, hard-coded list of attributes.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198883 91177308-0d34-0410-b5e6-96231b3b80d8

While it is allowed to not have an @ on subsequent lines, it seems
general practice to add them. If undesired, the code author can easily
remove them again and clang-format won't re-add them.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198871 91177308-0d34-0410-b5e6-96231b3b80d8

Before:
  #pragma mark Any non - hyphenated or hyphenated string(including parentheses).
After:
  #pragma mark Any non-hyphenated or hyphenated string (including parentheses).

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198870 91177308-0d34-0410-b5e6-96231b3b80d8

- Format a braced list with one element per line if it has nested
  braced lists.
- Use a column layout only when the list has 6+ elements (instead of the
  current 4+ elements).

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198869 91177308-0d34-0410-b5e6-96231b3b80d8

from the global address space (6.5.1 of the OpenCL 1.2 specification).
This makes clang construct the image arguments in the global address 
space and generate the argument metadata with the correct address space 
descriptor. 

Patch by Pedro Ferreira!



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198868 91177308-0d34-0410-b5e6-96231b3b80d8

check-clang: Add dependencies to PrintFunctionNames and SampleAnalyzerPlugin, for r198747 and r198820.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198867 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198858 91177308-0d34-0410-b5e6-96231b3b80d8

Summary: This fixes the leak described in http://llvm.org/bugs/show_bug.cgi?id=18318

Reviewers: chandlerc, dblaikie

Reviewed By: chandlerc

CC: cfe-commits

Differential Revision: http://llvm-reviews.chandlerc.com/D2474

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198857 91177308-0d34-0410-b5e6-96231b3b80d8

rdar://15779837.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198856 91177308-0d34-0410-b5e6-96231b3b80d8

It broke building.
This reverts commit r198845.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198849 91177308-0d34-0410-b5e6-96231b3b80d8

SPARC, where uint64_t apparently requires higher alignment than void*.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198845 91177308-0d34-0410-b5e6-96231b3b80d8

LLVM's Value interface which is used in LLVM's DominatorTree analysis
and which changed in LLVM r198836.

The DominatorTree analysis is actually a generic graph analysis and
should be moved to LLVM's support library to clarify that Clang and
others are using it with arbitrary graphs. Further, it seems likely that
it should be using something other than printAsOperand, but this is
a simpler build fix. I'll clean this up later.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198840 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198837 91177308-0d34-0410-b5e6-96231b3b80d8

issue 1430. Don't allow a pack expansion to be used as an argument to an alias
template unless the corresponding parameter is a parameter pack.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198833 91177308-0d34-0410-b5e6-96231b3b80d8

This test adjustment was missing from the previous patch.



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198822 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198820 91177308-0d34-0410-b5e6-96231b3b80d8

This patch refactors microsoft record layout to be more "natural".  The 
most dominant change is that vbptrs and vfptrs are injected after the 
fact.  This simplifies the implementation and the math for the offest 
for the first base/field after the vbptr.



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198818 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198815 91177308-0d34-0410-b5e6-96231b3b80d8

It's not worth keeping two copies of the identifier init and comparison code
just to save a pointer coparison.

This should reduce further once we get proper contextual keywords in the token
stream, so having the identifier checks in one place is a step towards that.

Cleanup only.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198814 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198805 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198804 91177308-0d34-0410-b5e6-96231b3b80d8

We (perhaps over-cautiously) disable the new plugin tests on static Windows
builds right now, matching what LLVM core does. This change was needed for the
lit check to work.

Thanks to Warren Hunt for spotting this.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198782 91177308-0d34-0410-b5e6-96231b3b80d8

[Serialization] In ASTReader::getInputFile record it when we didn't find the file to avoid looking it up again.

Hopefully addresses rdar://14514222.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198781 91177308-0d34-0410-b5e6-96231b3b80d8

...rather somewhere in the destructor when we try to access something and
realize the object has already been deleted. This is necessary because
the destructor is processed before the 'delete' itself.

Patch by Karthik Bhat!

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198779 91177308-0d34-0410-b5e6-96231b3b80d8

PR18406.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198770 91177308-0d34-0410-b5e6-96231b3b80d8

Before:
  vector<int> foo{ ::SomeFunction()};

After:
  vector<int> foo{::SomeFunction()};

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198769 91177308-0d34-0410-b5e6-96231b3b80d8

Patch reviewed by Rafael Espindola.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198765 91177308-0d34-0410-b5e6-96231b3b80d8

This is a further build fix attempt for r198747 on some Makefile builders where
the value wasn't set at all.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198764 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198751 91177308-0d34-0410-b5e6-96231b3b80d8

Convert CMake CLANG_BUILD_EXAMPLES to a boolean value for consumption by
Python, otherwise the raw config string gets passed through which may not be
0/1.

The if/else matches the technique used to solve the problem in LLVM's own
CMakeLists.txt.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198750 91177308-0d34-0410-b5e6-96231b3b80d8

With the old use of -std=c89 off_t is not defined and the build fails.

This seems to be another variation of
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40278.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198748 91177308-0d34-0410-b5e6-96231b3b80d8

Somehow the entire plugin infrastructure went wholly untested until now.

The only plugins available for use in testing are the examples, so plugin tests
will only be run if CLANG_BUILD_EXAMPLES is enabled in the build.

(The examples should really be enabled by default, not just to aid testing but
also to prevent bitrot in some key user-facing code. I'll propose that
shortly.)

Requires supporting changes in LLVM r198746.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@198747 91177308-0d34-0410-b5e6-96231b3b80d8