clobbers to forward pieces of large stores to small loads, we need to consider
the properly phi translated pointer in the store block.

llvm-svn: 90978

implicitly constant folds.

llvm-svn: 90977

llvm-svn: 90975

add, there is no need to scan the world to find the same add again.
This invalidates the previous testcase, which wasn't wonderful anyway,
because it needed a run of instcombine to permute the use-lists in 
just the right way to before GVN was run (so it was really fragile).
Not a big loss.

llvm-svn: 90973

binary operator that wasn't an add.  In this case, a xor.  Whoops.

llvm-svn: 90971

llvm-svn: 90970

to require the load ty/ptr to be passed in, no functionality change.

llvm-svn: 90960

and pointer instead of the load.  No functionality change.

llvm-svn: 90959

of the query.

llvm-svn: 90958

llvm-svn: 90957

own small class.  No functionality change.

llvm-svn: 90956

When a call is placed to spill an interval this spiller will first try to
break the interval up into its component values. Single value intervals and
intervals which have already been split (or are the result of previous splits)
are spilled by the default spiller.

Splitting intervals as described above may improve the performance of generated
code in some circumstances. This work is experimental however, and it still
miscompiles many benchmarks. It's not recommended for general use yet.

llvm-svn: 90951

llvm-svn: 90937

llvm-svn: 90929

phi translation of complex expressions like &A[i+1].  This has the
following benefits:

1. The phi translation logic is all contained in its own class with
   a strong interface and verification that it is self consistent.

2. The logic is more correct than before.  Previously, if intermediate
   expressions got PHI translated, we'd miss the update and scan for
   the wrong pointers in predecessor blocks.  @phi_trans2 is a testcase
   for this.

3. We have a lot less code in memdep.

We can handle phi translation across blocks of things like @phi_trans3,
which is pretty insane :).

This patch should fix the miscompiles of 255.vortex, and I tested it 
with a bootstrap of llvm-gcc, llvm-test and dejagnu of course.

llvm-svn: 90926

llvm-svn: 90925

Move isConsecutiveLoad to SelectionDAG. It's not target dependent and it's primary used by selectdag passes.

llvm-svn: 90922

llvm-svn: 90920

llvm-svn: 90919

llvm-svn: 90918

llvm-svn: 90917

llvm-svn: 90915

that Chris noticed.

llvm-svn: 90910

the translated value back to it if an instruction.

llvm-svn: 90909

llvm-svn: 90908

llvm-svn: 90906

examines; fall back to a conservative answer if there are
more. This works around some several compile time problems
resulting from BasicAliasAnalysis calling PointerMayBeCaptured.

The value has been chosen arbitrarily.

This fixes rdar://7438917 and may partially address PR5708.

llvm-svn: 90905

date when instsimplify kicks in.

llvm-svn: 90901

llvm-svn: 90898

- Also support the 'q' NEON registers asm code.

llvm-svn: 90894

llvm-svn: 90866

I'm not aware that this does anything significant on its own, but it's
needed for another patch that I'm working on.

llvm-svn: 90864

There is no need to supply ModuleCU to addType() as a parameter.

llvm-svn: 90858

llvm-svn: 90857

llvm-svn: 90855

forwarding any uses).  GlobalDCE can also do this, but is only
run at -O3.

llvm-svn: 90850

side-effect the current object.

llvm-svn: 90837

llvm-svn: 90836

llvm-svn: 90834

Reduce (cmp 0, and_su (foo, bar)) into (bit foo, bar). This saves extra instruction. Patch inspired by Brian Lucas!

llvm-svn: 90819