llvm-svn: 43210

llvm-svn: 43209

 - enable phi instructions demotion to stack
 - create alloca instructions in the entry block

llvm-svn: 43208

llvm-svn: 43207

llvm-svn: 43206

Fixes 5550319.

llvm-svn: 43205

llvm-svn: 43204

with the new legalizer.

llvm-svn: 43199

llvm-svn: 43198

This fixes CodeGen/X86/mem*.ll.

llvm-svn: 43197

llvm-svn: 43196

the operand's type from the right place.

llvm-svn: 43195

llvm-svn: 43194

Turn a store folding instruction into a load folding instruction. e.g.
     xorl  %edi, %eax
     movl  %eax, -32(%ebp)
     movl  -36(%ebp), %eax
     orl   %eax, -32(%ebp)
=>
     xorl  %edi, %eax
     orl   -36(%ebp), %eax
     mov   %eax, -32(%ebp)
This enables the unfolding optimization for a subsequent instruction which will
also eliminate the newly introduced store instruction.

llvm-svn: 43192

llvm-svn: 43191

llvm-svn: 43190

llvm-svn: 43189

llvm-svn: 43181

by Duncan

llvm-svn: 43177

llvm-svn: 43176

llvm-svn: 43175

asserts in later checks rather than producing
the ordinary load it is supposed to.  Avoid all
such hassles by directly returning an ordinary
load in this case.

llvm-svn: 43174

To do this it is necessary to add a "always inline" argument to the
memcpy node. For completeness I have also added this node to memmove
and memset.  I have also added getMem* functions, because the extra
argument makes it cumbersome to use getNode and because I get confused
by it :-)

llvm-svn: 43172

llvm-svn: 43171

llvm-svn: 43170

llvm-svn: 43169

llvm-svn: 43168

llvm-svn: 43167

llvm-svn: 43166

llvm-svn: 43165

llvm-svn: 43164

llvm-svn: 43161

llvm-svn: 43160

- Added getOpcodeAfterMemoryUnfold(). It doesn't unfold an instruction, but only returns the opcode of the instruction post unfolding.
- Fix some copy+paste bugs.

llvm-svn: 43153

llvm-svn: 43150

Instead of loading small global string from memory, use
integer constant.

llvm-svn: 43148

llvm-svn: 43147

llvm-svn: 43145

llvm-svn: 43144

fixing some obviously broken code :(

llvm-svn: 43141