inserted instructions.  No functionality change until someone starts using it.

llvm-svn: 90039

to inform GVN about the newly inserted values.  This fixes 
PR5631.

llvm-svn: 90022

fixes PR5630 and sets the stage for the next phase of goodness (testcase
pending).

llvm-svn: 90019

llvm-svn: 90011

array indexes.  The "complex" case of SRoA still handles them, and correctly.

This fixes a weirdness where we'd correctly avoid transforming A[0][42] if
the 42 was too large, but we'd only do it if it was one gep, not two separate
ones.

llvm-svn: 90007

where it is not available.  It's unclear how to get this inserted 
computation into GVN's scalar availability sets, Owen, help? :)

llvm-svn: 89997

translation done by memdep, and reenable gep translation 
again.

llvm-svn: 89992

SimplifyGEPInst method in InstructionSimplify.h.  No functionality change.

llvm-svn: 89980

llvm-svn: 89972

generates store to undef and some generates store to null as the idiom
for undefined behavior.  Since simplifycfg zaps both, don't remove the
undefined behavior in instcombine.

llvm-svn: 89971

by Alastair Lynn.

llvm-svn: 89970

llvm-svn: 89844

llvm-svn: 89758

ConstantExpr, not just the top-level operator. This allows it to
fold many more constants.

Also, make GlobalOpt call ConstantFoldConstantExpression on
GlobalVariable initializers.

llvm-svn: 89659

adjacent uses of a dead basic block from the same user. This fixes PR5596.

llvm-svn: 89658

llvm-svn: 89645

llvm-svn: 89644

llvm-svn: 89642

fixes part of PR5438.

llvm-svn: 89639

object size intrinsic and verify return type is correct. Collect various
code in one place.

llvm-svn: 89523

it may be used in contexts where preheader insertion may have failed due
to an indirectbr.

Make LoopSimplify's LoopSimplify::SeparateNestedLoop properly fail in
the case that it would require splitting an indirectbr edge.

These fix PR5502.

llvm-svn: 89484

blockaddress users. This fixes PR5569.

llvm-svn: 89483

llvm-svn: 89482

llvm-svn: 89479

tests/Transforms/InstCombine/shufflemask-undef.ll.  If
anyone cares, the use of 2*e here (and the equivalent
all over the place in instcombine) seems wrong, though
harmless: it should really be twice the length of the
input vector.  I think shufflevector used to require
that the mask have the same length as the input, but I
don't think that's true any more.  I don't care enough
about vectors to do anything about this...

llvm-svn: 89456

if it is not ultimately captured. Teach BasicAliasAnalysis that a 
local object address which does not escape and is never stored does
not alias with a value resulting from a load.

llvm-svn: 89398

they are lowered to instruction sequences more complex than a simple
load, such that CodeGen cannot rematerialize them, a reload from a
spill slot is likely to be cheaper than the complex sequence.

llvm-svn: 89374

Eliminate duplicate phi nodes in loops. Loop rotation, for example, can introduce these, and it's beneficial to later passes to clean them up.

llvm-svn: 89298

llvm-svn: 89297

llvm-svn: 89145

llvm-svn: 89123

llvm-svn: 89116

llvm-svn: 89110

llvm-svn: 89016

Fix an expensive-checks error.

The Mask and LHSMask may not be of the same size, so don't do the
transformation if they're different.

llvm-svn: 88972

a name.

llvm-svn: 88908

llvm-svn: 88865

llvm-svn: 88864

grief. I suspect this patch merely exposed a bug else.

llvm-svn: 88841

zext(icmp). It may be able to optimize that away. This fixes one of the cases
in PR5438.

llvm-svn: 88830