llvm-svn: 27443

llvm-svn: 27442

llvm-svn: 27441

llvm-svn: 27440

llvm-svn: 27439

llvm-svn: 27438

vector_shuffle v1, v1, <0, 4, 1, 5, 2, 6, 3, 7>

This is turned into
vector_shuffle v1, <undef>, <0, 0, 1, 1, 2, 2, 3, 3>
by dag combiner.

It would match a {p}unpckl on x86.

llvm-svn: 27437

CodeGen/X86/2006-04-04-CrossBlockCrash.ll

llvm-svn: 27436

llvm-svn: 27435

llvm-svn: 27434

llvm-svn: 27433

or custom lowering fails.

llvm-svn: 27432

llvm-svn: 27431

llvm-svn: 27430

llvm-svn: 27429

llvm-svn: 27428

llvm-svn: 27427

llvm-svn: 27426

llvm-svn: 27425

llvm-svn: 27424

llvm-svn: 27423

llvm-svn: 27422

llvm-svn: 27421

llvm-svn: 27420

llvm-svn: 27419

llvm-svn: 27418

llvm-svn: 27417

llvm-svn: 27416

llvm-svn: 27415

llvm-svn: 27414

Fix some broken logic that would cause us to codegen {2147483647,2147483647,2147483647,2147483647} as 'vspltisb v0, -1'.

llvm-svn: 27413

llvm-svn: 27412

__builtin_ia32_psrldqi128.

llvm-svn: 27411

llvm-svn: 27410

llvm-svn: 27409

llvm-svn: 27405

This, along with the previous dag combiner fix, fixes
CodeGen/Alpha/2006-04-04-zextload.ll

llvm-svn: 27403

legal.

llvm-svn: 27402

llvm-svn: 27401

handle all 4 PPC vector types.   This simplifies the matching code and allows
us to eliminate a bunch of patterns.  This also adds cases we were missing,
such as CodeGen/PowerPC/vec_splat.ll:splat_h.

llvm-svn: 27400