optimize a call to a llvm intrinsic to something that invovles a call to a C
library call, make sure it sets the right calling convention on the call.

e.g.
extern double pow(double, double);
double t(double x) {
  return pow(10, x);
}

Compiles to something like this for AAPCS-VFP:
define arm_aapcs_vfpcc double @t(double %x) #0 {
entry:
  %0 = call double @llvm.pow.f64(double 1.000000e+01, double %x)
  ret double %0
}

declare double @llvm.pow.f64(double, double) #1

Simplify libcall (part of instcombine) will turn the above into:
define arm_aapcs_vfpcc double @t(double %x) #0 {
entry:
  %__exp10 = call double @__exp10(double %x) #1
  ret double %__exp10
}

declare double @__exp10(double)

The pre-instcombine code works because calls to LLVM builtins are special.
Instruction selection will chose the right calling convention for the call.
However, the code after instcombine is wrong. The call to __exp10 will use
the C calling convention.

I can think of 3 options to fix this.

1. Make "C" calling convention just work since the target should know what CC
   is being used.

   This doesn't work because each function can use different CC with the "pcs"
   attribute.

2. Have Clang add the right CC keyword on the calls to LLVM builtin.

   This will work but it doesn't match the LLVM IR specification which states
   these are "Standard C Library Intrinsics".

3. Fix simplify libcall so the resulting calls to the C routines will have the
   proper CC keyword. e.g.
   %__exp10 = call arm_aapcs_vfpcc double @__exp10(double %x) #1

   This works and is the solution I implemented here.

Both solutions #2 and #3 would work. After carefully considering the pros and
cons, I decided to implement #3 for the following reasons.

1. It doesn't change the "spec" of the intrinsics.
2. It's a self-contained fix.

There are a couple of potential downsides.
1. There could be other places in the optimizer that is broken in the same way
   that's not addressed by this.
2. There could be other calling conventions that need to be propagated by
   simplify-libcall that's not handled.

But for now, this is the fix that I'm most comfortable with.

llvm-svn: 203488

llvm-svn: 203487

llvm-svn: 203486

[forgot to 'svn add' before committing r203483]

llvm-svn: 203485

  * Add masking instructions before loads and stores (in MC layer).
  * Add masking instructions after SP changes (in MC layer).
  * Forbid loads, stores and SP changes in delay slots (in MI layer).

Differential Revision: http://llvm-reviews.chandlerc.com/D2904

llvm-svn: 203484

NVPTX, like the other backends, relies on generic symbol name sanitizing done by
MCSymbol. However, the ptxas assembler is more stringent and disallows some
additional characters in symbol names.

See PR19099 for more details.

llvm-svn: 203483

Patch by Manuel Jacob.

llvm-svn: 203482

There's now a normal UI for that, apparently.

Patch by Manuel Jacob.

llvm-svn: 203481

More tests for LWG Issue #2263; this time to the associative and unordered containers. Still no changes to libc++

llvm-svn: 203480

Added tests to the sequence containers for for LWG Issue #2263. Comparing iterators and allocator pointers with different const-character. No changes to libc++

llvm-svn: 203479

Mark issues #2357 (wording changes in the standard, no functionality change) and #2132 (libc++ already does this) as complete.

llvm-svn: 203478

[C++11] Avoid implicit conversion of ArrayRef to std::vector and use move semantics where appropriate.

llvm-svn: 203477

This is necessary to avoid test failures in the CLooG test suite due to the
recent isl update.

We also need to update two polly test cases which rely on a certain order in the
textual description that isl chooses for its sets and maps. Changes here are not
often, but we should probably switch to a check that verifies such maps are
semantically equivalent instead of represented identically.

llvm-svn: 203476

What was use_iterator is now user_iterator.  Also switch to range-based
APIs, as in Clang r203365.

(This part of the change was missed in r203463)

Differential Revision: http://llvm-reviews.chandlerc.com/D3030

llvm-svn: 203475

[C++11] Replacing DeclBase iterators specific_attr_begin() and specific_attr_end() with iterator_range specific_attrs(). Updating all of the usages of the iterators with range-based for loops.

llvm-svn: 203474

llvm-svn: 203473

llvm-svn: 203472

This will replace the now badly named CLANG_IS_PRODUCTION.

llvm-svn: 203471

A fix for platform-dependent types in sanitizers' profiling support lib on x64 FreeBSD in 32-bit mode

llvm-svn: 203470

llvm-svn: 203469

llvm-svn: 203468

Summary:
Add loc() to the dynamic registry.
Other fixes:
 - Fix the polymorphic variant value to accept an exact match, even if
   there are other possible conversions.
 - Fix specifiesTypeLoc() to not crash on an empty
   NestedNameSpecifierLoc.

Reviewers: klimek

CC: cfe-commits, klimek

Differential Revision: http://llvm-reviews.chandlerc.com/D2928

llvm-svn: 203467

Before:
  void f() {
    bar([]() {}// Does not respect SpacesBeforeTrailingComments
        );
  }

After:
  void f() {
    bar([]() {} // Does not respect SpacesBeforeTrailingComments
        );
  }

This fixes llvm.org/PR19017.

llvm-svn: 203466

No functionality change.

llvm-svn: 203465

Summary:
This is a white lie to workaround a widespread bug in the -mfp64
implementation.

The problem is that none of the 32-bit fpu ops mention the fact that they
clobber the upper 32-bits of the 64-bit FPR. This allows MFHC1 to be
scheduled on the wrong side of most 32-bit FPU ops. Fixing that requires a
major overhaul of the FPU implementation which can't be done right now due to
time constraints.

MFHC1 is one of two affected instructions. These instructions are the only
FPU instructions that don't read or write the lower 32-bits. We therefore
pretend that it reads the bottom 32-bits to artificially create a dependency and
prevent the scheduler changing the behaviour of the code.
The other instruction is MTHC1 which will be fixed once I've have found a failing
test case for it. 

The testcase is test-suite/SingleSource/UnitTests/Vector/simple.c when
given TARGET_CFLAGS="-mips32r2 -mfp64 -mmsa".

Reviewers: jacksprat, matheusalmeida

Reviewed By: jacksprat

Differential Revision: http://llvm-reviews.chandlerc.com/D2966

llvm-svn: 203464

What was use_iterator is now user_iterator.  Also switch to range-based
APIs, as in Clang r203365.

llvm-svn: 203463

Older isl versions did not properly guard all function declarations for the
use in C++. Specifically, the isl/val_gmp.h header was not properly guarded.
This patch ensures we have the proper guards in place and do not accidentally
link to name-mangled C++ versions of those functions that are not available in
libisl.so.

llvm-svn: 203462

Reverting llvm::distance changes to use std::distance with iterators instead, per post-commit review feedback.

Replacing llvm::copy changes with SmallVector range-based construction which is a considerably cleaner approach.

llvm-svn: 203461

Removing llvm::distance and llvm::copy for iterator_range based on post-commit review feedback. Adding an explicit range-based constructor to SmallVector, which supersedes the llvm::copy functionality.

llvm-svn: 203460

llvm-svn: 203459

Summary:
If we need to break the second line here:
// something: aaaaa aaaaa aaaaaa aaaaa aaaaa
//            aaaaa aaaaa aaaaaa aaaaa aaaaa aaaaa

with the patch it will be turned to

// something: aaaaa aaaaa aaaaaa aaaaa aaaaa
//            aaaaa aaaaa aaaaaa aaaaa aaaaa
//            aaaaa

instead of

// something: aaaaa aaaaa aaaaaa aaaaa aaaaa
//            aaaaa aaaaa aaaaaa aaaaa aaaaa
// aaaaa

Reviewers: djasper, klimek

Reviewed By: klimek

CC: cfe-commits, klimek

Differential Revision: http://llvm-reviews.chandlerc.com/D2988

llvm-svn: 203458

Summary:
This is immediately useful for generating macro expansion notes, and
may be useful for other things later on.

Reviewers: klimek, djasper

Reviewed By: klimek

CC: cfe-commits

Differential Revision: http://llvm-reviews.chandlerc.com/D2950

llvm-svn: 203457

llvm-svn: 203456

llvm-svn: 203455

Before:
  int c = []()->int { return 2; }();

After:
  int c = []() -> int { return 2; }();

llvm-svn: 203452

Summary: Peter, I guess, this can help you in testing your check.

Reviewers: djasper, pcc, klimek

Reviewed By: klimek

CC: cfe-commits

Differential Revision: http://llvm-reviews.chandlerc.com/D2989

llvm-svn: 203451

The function was making too many assumptions about its input:

1. The NEON_VDUP optimisation was far too aggressive, assuming (I
think) that the input would always be BUILD_VECTOR.

2. We were treating most unknown concats as legal (by returning Op
rather than SDValue()). I think only concats of pairs of vectors are
actually legal.

http://llvm.org/PR19094

llvm-svn: 203450

Seed the QueueItem objects with the item_refs and addresses when they are fetched
in one batch.  If additional information is needed from the QueueItem, fetch it
lazily one pending item per function call.
<rdar://problem/16270007>, <rdar://problem/16032150> 

llvm-svn: 203449

as well. I don't see any particular need but it imposes no cost to
support it and it makes the API cleaner.

llvm-svn: 203448

now that there is essentially no cost to doing so. Yay C++11.

llvm-svn: 203447