for SSE1, SSE2, SSE3, and SSSE3.

llvm-svn: 36656

llvm-svn: 36653

Due to darwin gcc bug, one version of darwin linker coalesces
static const int, which defauts PassID based pass identification.

llvm-svn: 36652

Emit correct register move information in eh frames for X86. This allows Shootout-C++/except to pass on x86/linux
with non-llvm-compiled (e.g. "native") unwind runtime.

llvm-svn: 36647

llvm-svn: 36646

1. Correct output offsets on Linux
2. Fix "style" of personality function. It shouldn't be indirect.

llvm-svn: 36633

llvm-svn: 36632

llvm-svn: 36628

llvm-svn: 36627

llvm-svn: 36626

should be taught to deal with protected symbols.

llvm-svn: 36565

llvm-svn: 36495

llvm-svn: 36494

llvm-svn: 36456

llvm-svn: 36452

clobbered by a call.

llvm-svn: 36448

when promoted to 64-bits

llvm-svn: 36442

will follow.

llvm-svn: 36435

llvm-svn: 36430

llvm-svn: 36425

llvm-svn: 36423

llvm-svn: 36407

llvm-svn: 36406

llvm-svn: 36405

llvm-svn: 36404

        vector_shuffle v1, v2, <2, 6, 3, 7>

I.e.

         vector_shuffle v, undef, <2, 2, 3, 3>

MMX only has a shuffle for v4i16 vectors. It needs to use the unpackh for
this type of operation.

llvm-svn: 36403

llvm-svn: 36359

llvm-svn: 36355

llvm-svn: 36318

llvm-svn: 36287

X86 32 bits.

llvm-svn: 36283

llvm-svn: 36279

llvm-svn: 36265

llvm-svn: 36256

llvm-svn: 36213

it is defined in.

llvm-svn: 36196

Also, fixed static case in presence of eax livin. This fixes PR331

PS: Why don't we still have push/pop instructions? :)
llvm-svn: 36195

llvm-svn: 36185

llvm-svn: 36177

replaced it with a FIXME should have determined what did work.  Then he would have
realized that the code was in fact correct, and would have avoided breaking it.

llvm-svn: 36173