llvm-svn: 26604

and variable value.
Similarly for memcpy.

llvm-svn: 26603

1. Don't emit debug info, or other llvm.metadata to the .cbe.c file.
2. Mark static ctors/dtors as such, so that bugpoint works on C++ code
   compiled with the new CFE.

llvm-svn: 26602

asm printer.

llvm-svn: 26599

llvm-svn: 26598

hypotheticalized about, where we would incorrectly merge two globals in
different sections.

llvm-svn: 26597

llvm-svn: 26595

llvm-svn: 26594

llvm-svn: 26593

llvm-svn: 26591

a nop into Freebench/neural, which speeds it up from 136->129s (~5.4%).

llvm-svn: 26590

llvm-svn: 26588

flushes

llvm-svn: 26587

things like copyfromreg.

llvm-svn: 26586

llvm-svn: 26585

llvm-svn: 26583

- Conditionalize Dwarf debugging output (Darwin only for now).

llvm-svn: 26582

llvm-svn: 26581

llvm-svn: 26580

the pointer is known to come from either a global variable, alloca or
malloc.  This allows us to compile this:

  P = malloc(28);
  memset(P, 0, 28);

into explicit stores on PPC instead of a memset call.

llvm-svn: 26577

llvm-svn: 26573

llvm-svn: 26572

range to be generated.

llvm-svn: 26570

targets can implement them.  Make the top-down scheduler non-g5-specific.

Remove the old testing hazard recognizer.

llvm-svn: 26569

llvm-svn: 26567

llvm-svn: 26566

noops as null pointers in the instruction sequence.

llvm-svn: 26564

llvm-svn: 26562

copyfromreg nodes.  Clearly useful!

llvm-svn: 26559

llvm-svn: 26558

llvm-svn: 26554

function of the top-down scheduler are completely bogus currently, and
having (future) PPC specific in this file is also wrong, but this is a
small incremental step.

llvm-svn: 26552

bounds its lifetime.

llvm-svn: 26550

llvm-svn: 26549

llvm-svn: 26548

we want to canonicalize the other way.

llvm-svn: 26547

llvm-svn: 26545

It's causing an infinite loop compiling ldecod on x86 / Darwin.

llvm-svn: 26544

llvm-svn: 26543

llvm-svn: 26542