llvm-svn: 179928

llvm-svn: 179927

The logic that actually compares the types considers pointers and integers the
same if they are of the same size. This created a strange mismatch between hash
and reality and made the test case for this fail on some platforms (yay,
test cases).

llvm-svn: 179905

Also make some static function class functions to avoid having to mention the
class namespace for enums all the time.

No functionality change intended.

llvm-svn: 179886

llvm-svn: 179826

If the return type is a pointer and the call returns an integer, then do the
inttoptr convertions. And vice versa.

llvm-svn: 179817

llvm-svn: 179789

In the function InstCombiner::visitExtractElementInst() removed the limitation that extract is promoted over a cast only if the cast has only one use.

llvm-svn: 179786

Added a function scalarizePHI() that sclarizes a vector phi instruction if it has only 2 uses: one to promote the vector phi in a loop and the other use is an extract operation of one element at a constant location.

llvm-svn: 179783

llvm-svn: 179775

A min/max operation is represented by a select(cmp(lt/le/gt/ge, X, Y), X, Y)
sequence in LLVM. If we see such a sequence we can treat it just as any other
commutative binary instruction and reduce it.

This appears to help bzip2 by about 1.5% on an imac12,2.

radar://12960601

llvm-svn: 179773

Fixes PR15748.

llvm-svn: 179757

It is causing stage2 builds to fail, let's get them running again.

llvm-svn: 179750

Simplify:
(select (icmp eq (and X, C1), 0), Y, (or Y, C2))

Into:
(or (shl (and X, C1), C3), y)

Where:
C3 = Log(C2) - Log(C1)

If:
C1 and C2 are both powers of two

llvm-svn: 179748

[objc-arc] Do not mismatch up retains inside a for loop with releases outside said for loop in the presense of differing provenance caused by escaping blocks.

This occurs due to an alloca representing a separate ownership from the
original pointer. Thus consider the following pseudo-IR:

  objc_retain(%a)
  for (...) {
    objc_retain(%a)
    %block <- %a
    F(%block)
    objc_release(%block)
  }
  objc_release(%a)

From the perspective of the optimizer, the %block is a separate
provenance from the original %a. Thus the optimizer pairs up the inner
retain for %a and the outer release from %a, resulting in segfaults.

This is fixed by noting that the signature of a mismatch of
retain/releases inside the for loop is a Use/CanRelease top down with an
None bottom up (since bottom up the Retain-CanRelease-Use-Release
sequence is completed by the inner objc_retain, but top down due to the
differing provenance from the objc_release said sequence is not
completed). In said case in CheckForCFGHazards, we now clear the state
of %a implying that no pairing will occur.

Additionally a test case is included.

rdar://12969722

llvm-svn: 179747

llvm-svn: 179746

llvm-svn: 179729

llvm-svn: 179721

[objc-arc] Added descriptions for EnableARCAnnotations, EnableCheckForCFGHazards, EnableARCOptimizations.

llvm-svn: 179718

llvm-svn: 179717

Differential Revision: http://llvm-reviews.chandlerc.com/D620

llvm-svn: 179661

If a switch instruction has a case for every possible value of its type,
with the same successor, SimplifyCFG would replace it with an icmp ult,
but the computation of the bound overflows in that case, which inverts
the test.

Patch by Jed Davis!

llvm-svn: 179587

Two return types are not equivalent if one is a pointer and the other is an
integral. This is because we cannot bitcast a pointer to an integral value.
PR15185

llvm-svn: 179569

SLPVectorizer: Make it a function pass and add code for hoisting the vector-gather sequence out of loops.

llvm-svn: 179562

llvm-svn: 179542

Add an option -vectorize-slp-aggressive for running the BB vectorizer. Make -fslp-vectorize run the slp-vectorizer.

llvm-svn: 179508

llvm-svn: 179505

llvm-svn: 179504

One performs: (X == 13 | X == 14) -> X-13 <u 2
The other: (A == C1 || A == C2) -> (A & ~(C1 ^ C2)) == C1

The problem is that there are certain values of C1 and C2 that
trigger both transforms but the first one blocks out the second,
this generates suboptimal code.

Reordering the transforms should be better in every case and
allows us to do interesting stuff like turn:
  %shr = lshr i32 %X, 4
  %and = and i32 %shr, 15
  %add = add i32 %and, -14
  %tobool = icmp ne i32 %add, 0

into:
  %and = and i32 %X, 240
  %tobool = icmp ne i32 %and, 224

llvm-svn: 179493

llvm-svn: 179483

llvm-svn: 179479

SLPVectorizer: Add support for trees that don't start at binary operators, and add the cost of extracting values from the roots of the tree.

llvm-svn: 179475

llvm-svn: 179470

There is a Constant with non-constant operands: blockaddress.

llvm-svn: 179460

This is basically the same fix in three different places. We use a set to avoid
walking the whole tree of a big ConstantExprs multiple times.

For example: (select cmp, (add big_expr 1), (add big_expr 2))
We don't want to visit big_expr twice here, it may consist of thousands of
nodes.

The testcase exercises this by creating an insanely large ConstantExprs out of
a loop. It's questionable if the optimizer should ever create those, but this
can be triggered with real C code. Fixes PR15714.

llvm-svn: 179458

Fixes PR15737.

llvm-svn: 179417

SLPVectorizer: add support for vectorization of diamond shaped trees. We now perform a preliminary traversal of the graph to collect values with multiple users and check where the users came from. 

llvm-svn: 179414

llvm-svn: 179412

The transform will execute like so:
(A & ~B) == 0 --> (A & B) != 0
(A & ~B) != 0 --> (A & B) == 0

llvm-svn: 179386

Don't classify idiv/udiv as a reduction operation. Integer division is lossy.
For example : (1 / 2) * 4 != 4/2.

Example:

int a[] = { 2, 5, 2, 2}
int x = 80;

for()
  x /= a[i];

Scalar:
  x /= 2 // = 40
  x /= 5 // = 8
  x /= 2 // = 4
  x /= 2 // = 2

Vectorized:

 <80, 1> / <2,5> //= <40,0>
 <40, 0> / <2,2> //= <20,0>

 20*0 = 0

radar://13640654

llvm-svn: 179381