This fixes llvm.org/PR15033.

Also: Always break before a parameter, if the previous parameter was
split over multiple lines. This was necessary to make the right
decisions in for-loops, almost always makes the code more readable and
also fixes llvm.org/PR14873.

Before:
for (llvm::ArrayRef<NamedDecl *>::iterator I = FD->getDeclsInPrototypeScope()
         .begin(), E = FD->getDeclsInPrototypeScope().end();
     I != E; ++I) {
}
foo(bar(bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb,
        ccccccccccccccccccccccccccccc), d, bar(e, f));

After:
for (llvm::ArrayRef<NamedDecl *>::iterator
         I = FD->getDeclsInPrototypeScope().begin(),
         E = FD->getDeclsInPrototypeScope().end();
     I != E; ++I) {
}
foo(bar(bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb,
        ccccccccccccccccccccccccccccc),
    d, bar(e, f));

llvm-svn: 175741

that had been used on OS X only.
The INTERCEPTOR() macro on OS X is now responsible for declaring the wrapped function, the wrapper and the
pair of pointers to them in __DATA,__interposition section. Thus adding an interceptor requires editing a single file now.

llvm-svn: 175740

llvm-svn: 175739

llvm-svn: 175738

llvm-svn: 175737

llvm-svn: 175736

__global, __constant or __local qualifier.

llvm-svn: 175735

Add support to Sema and CodeGen for floating point vector types in OpenCL.

llvm-svn: 175734

NOTE: This is a candidate for the Mesa stable branch.

Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 175733

llvm-svn: 175732

llvm-svn: 175731

llvm-svn: 175730

Code review feedback for r175580 by Jordan Rose.

llvm-svn: 175729

llvm-svn: 175728

This brings the number of remaining failures in 'make check' without
LiveVariables down to 39, with 1 unexpectedly passing test.

llvm-svn: 175727

llvm-svn: 175726

Post commit code review feedback to r175705 from Jordan Rose.

llvm-svn: 175725

Post commit code review feedback to r175705 from Jordan Rose.

llvm-svn: 175724

Provides a general way to add 'explicit' for conversion operators (a no-op when
compiling as C++98).

llvm-svn: 175723

This is a precursor to making Optional<T>'s operator bool 'explicit' when
building Clang & LLVM as C++11.

llvm-svn: 175722

llvm-svn: 175721

llvm-svn: 175720

available.

With this commit there are no longer any assertion or verifier failures when
running 'make check' without LiveVariables. There are still 56 failing tests
with codegen differences and 1 unexpectedly passing test.

llvm-svn: 175719

there were inline br .+4 instructions. Soon everything can enjoy the
full instruction scheduling experience.

llvm-svn: 175718

llvm-svn: 175717

This allows MemRegion and MemRegionManager to avoid asking over and over
again whether an class is a virtual base or a non-virtual base.

Minor optimization/cleanup; no functionality change.

llvm-svn: 175716

The test is similar to <rdar://problem/13239840> but doesn't actually test
the case that fails there. It's still a good test, though.

llvm-svn: 175715

Some that I just added needed conversion to use 'None', others looked
better using Optional<SVal>::create.

No functionality change.

llvm-svn: 175714

CIndex.cpp: Appease g++-4.4. "if (Optional<unsigned> Minor = In.getMinor())" did not work as expected.

llvm-svn: 175711

llvm-svn: 175710

llvm-svn: 175709

being included in C++. Don't define alignof or alignas in this case. Note that
the C++11 standard is broken in various ways here (it refers to the contents
of <stdalign.h> in C99, where that header did not exist, and doesn't mention
the alignas macro at all), but we do our best to do what it intended.

llvm-svn: 175708

The constructs %hi() and %lo() represent the high and low 16 
bits of the address. 
Because the 16 bit offset field of an LW instruction is 
interpreted as signed, if bit 15 of the low part is 1 then the 
low part will act as a negative and 1 needs to be added to the 
high part.

Contributer: Vladimir Medic
llvm-svn: 175707

files.

llvm-svn: 175706

llvm-svn: 175705

Matches changes made to SVal's similar functions based on Jordan Rose's review
feedback to r175594.

Also change isKind to take a reference rather than a non-null pointer, while I'm
at it. (& make TypeLoc::isKind private)

llvm-svn: 175704

- When deciding if we can reuse a lazy binding, make sure to check if there
  are additional bindings in the sub-region.
- When reading from a lazy binding, don't accidentally strip off casts or
  base object regions. This slows down lazy binding reading a bit but is
  necessary for type sanity when treating one class as another.

A bit of minor refactoring allowed these two checks to be unified in a nice
early-return-using helper function.

<rdar://problem/13239840>

llvm-svn: 175703

(lldb) frame variable
without first launching the inferior, you get:
error: invalid frame

this is misleading and should probably hint that there is no process. Adding this flag makes sure that we get:
error: invalid process

The difference between eFlagRequiresProcess and eFlagProcessMustBeLaunched is an open question.

llvm-svn: 175702

llvm-svn: 175701

work on i386.  Now we let the JIT emit SSE/SSE2
instructions on i386.

<rdar://problem/13240476>

llvm-svn: 175700