llvm-svn: 66359

llvm-svn: 64719

llvm-svn: 64620

llvm-svn: 64093

llvm-svn: 60632

llvm-svn: 60286

llvm-svn: 59052

llvm-svn: 58104

llvm-svn: 57733

llvm-svn: 55348

llvm-svn: 53722

llvm-svn: 52663

llvm-svn: 50681

llvm-svn: 50029

llvm-svn: 48912

llvm-svn: 48714

Fix X86's isTruncateFree to not claim that truncate to i1 is free. This fixes Bill's testcase that failed for r48491.

llvm-svn: 48542

llvm-svn: 48356

llvm-svn: 48054

llvm-svn: 47996

llvm-svn: 47828

llvm-svn: 47704

llvm-svn: 47702

llvm-svn: 47688

new things.

llvm-svn: 47458

llvm-svn: 47431

llvm-svn: 47278

llvm-svn: 47237

llvm-svn: 47204

llvm-svn: 47109

llvm-svn: 45869

commit all arguments where moved to the stack slot where they would
reside on a normal function call before the lowering to the tail call
stack slot. This was done to prevent arguments overwriting each other.
Now only arguments sourcing from a FORMAL_ARGUMENTS node or a
CopyFromReg node with virtual register (could also be a caller's
argument) are lowered indirectly.

 --This line, and those below, will be ignored--

M    X86/X86ISelLowering.cpp
M    X86/README.txt

llvm-svn: 45867

llvm-svn: 45768

llvm-svn: 45723

comparisons with a constant.  This allows us to compile isnan to:

_foo:
	fcmpu cr7, f1, f1
	mfcr r2
	rlwinm r3, r2, 0, 31, 31
	blr 

instead of:

LCPI1_0:					;  float
	.space	4
_foo:
	lis r2, ha16(LCPI1_0)
	lfs f0, lo16(LCPI1_0)(r2)
	fcmpu cr7, f1, f0
	mfcr r2
	rlwinm r3, r2, 0, 31, 31
	blr 

llvm-svn: 45405

llvm-svn: 45404

if we are just going to store it back anyway.  This improves things 
like:
double foo();
void bar(double *P) { *P = foo(); }

llvm-svn: 45399

llvm-svn: 45397

llvm-svn: 45387

llvm-svn: 45343