llvm-svn: 26895

instructions

llvm-svn: 26894

llvm-svn: 26893

llvm-svn: 26891

llvm-svn: 26889

llvm-svn: 26888

TODO: leave specific ones as VECTOR_SHUFFLE's and turn them into specialized
operations like vsplt*

llvm-svn: 26887

llvm-svn: 26886

llvm-svn: 26885

llvm-svn: 26884

llvm-svn: 26883

llvm-svn: 26880

llvm-svn: 26877

llvm-svn: 26874

Transforms/CorrelatedExprs/switch.ll

Patch contributed by Eric Kidd!

llvm-svn: 26872

llvm-svn: 26870

llvm-svn: 26868

llvm-svn: 26867

llvm-svn: 26866

llvm-svn: 26865

llvm-svn: 26863

llvm-svn: 26862

llvm-svn: 26861

llvm-svn: 26860

llvm-svn: 26858

llvm-svn: 26857

llvm-svn: 26856

llvm-svn: 26853

This implements CodeGen/Generic/vector.ll:test_variable_buildvector

llvm-svn: 26852

a constant.  This implements test_constant_insert in CodeGen/Generic/vector.ll

llvm-svn: 26851

llvm-svn: 26849

llvm-svn: 26848

Rename ConstantVec -> BUILD_VECTOR and VConstant -> VBUILD_VECTOR.  Allow*BUILD_VECTOR to take variable inputs.

llvm-svn: 26847

llvm-svn: 26845

This fixes CodeGen/Generic/vector.ll

llvm-svn: 26843

- Added more debugging info.
- Allow reuse of IV of negative stride. e.g. -4 stride == 2 * iv of -2 stride.

llvm-svn: 26841

things.

llvm-svn: 26840

V8I16, etc.

llvm-svn: 26838

for more IV reuses.

llvm-svn: 26837

cost-modeled as producing a flag.  This fixes the test I just added for neg

llvm-svn: 26835