llvm-svn: 108137

llvm-svn: 108135

llvm-svn: 108131

llvm-svn: 108130

the LHS and RHS of an and/or instruction, don't multiply add
known predecessor values.  This fixes the crash on testcase
from PR7498

llvm-svn: 108114

operation, but the way it's implemented requires the operation to also be
commutative.  So add a check for commutativity (and tweak the corresponding
comments).  This makes no difference in practice since every associative
LLVM instruction is also commutative!  Here's an example to show the need
for commutativity: the accum_recursion.ll testcase calculates the factorial
function.  Before the transformation the result of a call is
  ((((1*1)*2)*3)...)*x
while afterwards it is
  (((1*x)*(x-1))...*2)*1
which clearly requires both associativity and commutativity of * to be equal
to the original.

llvm-svn: 108056

llvm-svn: 107976

llvm-svn: 107971

llvm-svn: 107969

address spaces when SRoA'ing memcpy's.

llvm-svn: 107846

llvm-svn: 107637

have any effect, and second, deleting stores can potentially invalidate
an AliasAnalysis, and there's currently no notification for this.

llvm-svn: 107496

llvm-svn: 107278

llvm-svn: 107273

llvm-svn: 107272

llvm-svn: 107271

llvm-svn: 107270

llvm-svn: 107015

llvm-svn: 107000

large integers, the first inserted value would always create
an 'or X, 0'.  Even though this is trivially zapped by
instcombine, don't bother creating this pointless instruction.

llvm-svn: 106979

the returned value after the tail call if it differs from other return
values.  The optimal thing to do would be to introduce a phi node for
the return value, but for the moment just fix the miscompile.

llvm-svn: 106947

SCEVUnknown values which are loop-variant, as LSR can't do anything
interesting with these values in any case. This fixes very slow compile
times on loops which have large numbers of such values.

llvm-svn: 106897

for an "i" constraint should get lowered; PR 6309.  While
this argument was passed around a lot, this is the only
place it was used, so it goes away from a lot of other
places.

llvm-svn: 106893

use ArgOperand API; tighten type of handleFreeWithNonTrivialDependency to be able to use isFreeCall whithout a cast or new overload

llvm-svn: 106823

llvm-svn: 106764

enough special case, and it theoretically allows more folding because
it works even when x is unanalyzable.

llvm-svn: 106763

llvm-svn: 106759

llvm-svn: 106734

llvm-svn: 106731

llvm-svn: 106730

llvm-svn: 106729

llvm-svn: 106709

llvm-svn: 106707

The ValueMapper used by various cloning utility maps MDNodes also.

llvm-svn: 106706

is another max which folds. This fixes PR7454.

llvm-svn: 106594

llvm-svn: 106542

SmallVector, and other SmallVector simplifications.

llvm-svn: 106452

llvm-svn: 106397

use sharing map. The reconcileNewOffset logic already forces a
separate use if the kinds differ, so incorporating the kind in the
key means we can track more sharing opportunities.

More sharing means fewer total uses to track, which means smaller
problem sizes, which means the conservative throttles don't kick
in as often.

llvm-svn: 106396

llvm-svn: 106395