git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215670 91177308-0d34-0410-b5e6-96231b3b80d8

Part of <rdar://problem/17688758>

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215666 91177308-0d34-0410-b5e6-96231b3b80d8

to recover the performance after r214064.

Also sorts out the naming for PostOrderCFGView, ReversePostOrderCFGView,
BackwardDataflowWorklist and ForwardDataflowWorklist, to match the accepted
terminology.

Also unifies BackwardDataflowWorklist and ForwardDataflowWorklist to share
the "worklist for prioritization, post-order traversal for fallback" logic,
and to avoid repetitive sorting.

Also cleans up comments in the affected area.




git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215650 91177308-0d34-0410-b5e6-96231b3b80d8

It fits better with LLVM's memory model to try to do this in the
backend. Specifically, narrowing wide loads in the backends should be
relatively straightforward and is generally valuable, whereas widening
loads tends to be very constrained.

Discussion here:

  http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20140811/112581.html

This reverts commit r215614.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215648 91177308-0d34-0410-b5e6-96231b3b80d8

auroraux.org is not resolving.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215644 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215639 91177308-0d34-0410-b5e6-96231b3b80d8

Before:
  template <class T>
  T *f(T &c)  // Problem here: no line break before f
  {
    return NULL;
  }

After:
  template <class T>
  T *
  f(T &c)
  {
    return NULL;
  }

Patch by Marek Kurdej, thank you!

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215633 91177308-0d34-0410-b5e6-96231b3b80d8

Before:
  x = * a(x) = *a(y);

After:
  x = *a(x) = *a(y);

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215632 91177308-0d34-0410-b5e6-96231b3b80d8

Before:
  typedef size_t (*aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa)(
      const aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa *aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa);

After:
  typedef size_t (*aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa)(
      const aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa *
          aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa);

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215631 91177308-0d34-0410-b5e6-96231b3b80d8

address spaces.



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215629 91177308-0d34-0410-b5e6-96231b3b80d8

It doesn't really make sense to try and do stuff with #pragma init_seg
when targeting non-Microsoft platforms; notions like library vs user
initializers don't exist for other targets.

This fixes PR20639.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215618 91177308-0d34-0410-b5e6-96231b3b80d8

avoids users of AllDiagnostics.h from needing to pregenerate *all* generated
headers. Hopefully this will make my modules buildbot happier...


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215617 91177308-0d34-0410-b5e6-96231b3b80d8

declaration has its definition instantiated in two sibling modules and they use
a partial specialization.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215616 91177308-0d34-0410-b5e6-96231b3b80d8

Currently when laying out bitfields that don't need any padding, we
represent them as a wide enough int to contain all of the bits. This
can be hard on the backend since we'll do things like represent stores
to a few bits as loading an i144, masking it with a large constant,
and storing it back.

This turns up in less pathological cases where we load and mask 64 bit
word on a 32 bit platform when we actually only need to access 32 bits.
This leads to bad code being generated in most of our 32 bit backends.

In practice, there are often natural breaks in bitfields, and it's a
fairly simple and effective heuristic to split these fields into legal
integer sized chunks when it will be equivalent (ie, it won't force us
to add any extra padding).

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215614 91177308-0d34-0410-b5e6-96231b3b80d8

definitions (because some other declaration declares a special member that
isn't present in the canonical definition), we need to search *all* of them; we
can't just stop when we find the requested name in any of the definitions,
because that can fail to find things (and in particular, it can fail to find
the member of the canonical declaration and return a bogus ODR failure).


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215612 91177308-0d34-0410-b5e6-96231b3b80d8

This fixes a regression I caused back in r211766.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215609 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215607 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215606 91177308-0d34-0410-b5e6-96231b3b80d8

found in global pool as well. rdar://16808765


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215603 91177308-0d34-0410-b5e6-96231b3b80d8

Sorry about the noise.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215601 91177308-0d34-0410-b5e6-96231b3b80d8

Implement __pld, __pldx, __pli and __plix builtin intrinsics as specified in
ARM ACLE 2.0.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215599 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215592 91177308-0d34-0410-b5e6-96231b3b80d8

The tests in r215568 hard code a value as %0 in their checks. This
isn't correct in asserts builds.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215585 91177308-0d34-0410-b5e6-96231b3b80d8

// rdar://16808765


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215581 91177308-0d34-0410-b5e6-96231b3b80d8

recursively within the emission of another inline function. This ultimately
led to us emitting the same inline function definition twice, which we then
rejected because we believed we had a mangled name conflict.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215579 91177308-0d34-0410-b5e6-96231b3b80d8

expression to the best method found in global method pools. 
This is wip.  // rdar://16808765


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215577 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215571 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215569 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215568 91177308-0d34-0410-b5e6-96231b3b80d8

Should fix test ulibc driver tests.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215561 91177308-0d34-0410-b5e6-96231b3b80d8

std::unique_ptr is null initialized and reset default to null.

Thanks to David Blaikie for noticing.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215560 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215559 91177308-0d34-0410-b5e6-96231b3b80d8

Modifications made by clang-tidy with minor tweaks.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215557 91177308-0d34-0410-b5e6-96231b3b80d8

Yet more problems due to the missing CXXBindTemporaryExpr in the CFG for
default arguments.

Unfortunately we cannot just switch off inserting temporaries for the
corresponding default arguments, as that breaks existing tests
(test/SemaCXX/return-noreturn.cpp:245).

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215554 91177308-0d34-0410-b5e6-96231b3b80d8

of MIPS toolchains.

The uCLibc implemented for multiple architectures. A couple of MIPS toolchains
contains both uCLibc and glibc implementation so these options allow to select
used C library.

Initially -muclibc / -mglibc (as well as -mbionic) have been implemented in gcc
for various architectures so they are not MIPS specific.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215552 91177308-0d34-0410-b5e6-96231b3b80d8

integrated assembler, libc++ and libgcc. Set emulation for ld for both
platforms for correct -m32 handling.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215551 91177308-0d34-0410-b5e6-96231b3b80d8

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215549 91177308-0d34-0410-b5e6-96231b3b80d8

With modules we start accessing headers for the first time while reading
the module map, which often has very different paths from the include
scanning logic.

Using the name by which the file was accessed gets us one step closer to
the right solution, which is using a FileName abstraction that decouples
the name by which a file was accessed from the FileEntry.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215541 91177308-0d34-0410-b5e6-96231b3b80d8

Rather than silently disabling unaligned accesses for v6m targets as
in the previous patch to llvm, instead produce a warning saying that
this architecture doesn't support unaligned accesses.

Patch by Ben Foster



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215531 91177308-0d34-0410-b5e6-96231b3b80d8

After (even with BinPacking = false):
  const Aaaaaa aaaaa = {
      aaaaa,  bbbbb,  ccccc,  ddddd,  eeeee,  ffffff, ggggg, hhhhhh,
      iiiiii, jjjjjj, kkkkkk, aaaaa,  bbbbb,  ccccc,  ddddd, eeeee,
      ffffff, ggggg,  hhhhhh, iiiiii, jjjjjj, kkkkkk,
  };

Before:
  <each element on its own line>

This fixes http://llvm.org/PR20623.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@215529 91177308-0d34-0410-b5e6-96231b3b80d8