##DEBUG_VALUE: runOnMachineFunction:this <- RDI+0
	##DEBUG_VALUE: runOnMachineFunction:fn <- RSI+0
	##DEBUG_VALUE: DeadDefs <- undef ## SimpleRegisterCoalescing.cpp:2706
	##DEBUG_VALUE: getRegInfo:this <- [%rsp+$56]+$0
	##DEBUG_VALUE: getTarget:this <- [%rsp+$56]+$0

llvm-svn: 102655

llvm-svn: 102654

llvm-svn: 102653

llvm-svn: 102652

llvm-svn: 102651

just leave it in Developer/usr/local/lib.

llvm-svn: 102646

llvm-svn: 102637

llvm-svn: 102621

level metadata does not have any function local operands.
This would have caught the problem found in PR6112.

llvm-svn: 102620

llvm-svn: 102612

llvm-svn: 102606

Limit alignment in SmallVector 8, otherwise GCC assumes 16 byte alignment.
opetaror new, and malloc only return 8-byte aligned memory on 32-bit Linux,
which cause a crash if code is compiled with -O3 (or -ftree-vectorize) and some
SmallVector code is vectorized.

llvm-svn: 102604

llvm-svn: 102602

llvm-svn: 102599

should use esp / rsp to reference frame instead.

llvm-svn: 102596

llvm-svn: 102594

llvm-svn: 102590

llvm-svn: 102585

llvm-svn: 102584

llvm-svn: 102581

llvm-svn: 102577

llvm-svn: 102573

- Also, update dbg_value is the value is being re-matted from a frame slot, e.g. fixed slots for arguments.

llvm-svn: 102565

llvm-svn: 102558

Operand size override prefix to be part of their records.

llvm-svn: 102556

Replace r102368 with code that's less fragile. This creates DBG_VALUE instructions for function arguments early and insert them after instruction selection is done.

llvm-svn: 102554

llvm-svn: 102547

llvm-svn: 102545

llvm-svn: 102544

be installed.  Disable it by default.

llvm-svn: 102531

Add sizes non-floating point versions for the eh sjlj intrinsic expansions.

rdar://7895451

llvm-svn: 102526

Update tests. Now DBG_VALUE instruction is created only if alloca corresponding to llvm.dbg.declare is missing.

llvm-svn: 102524

Patch by Lars R in PR6961.

llvm-svn: 102523

metadata references in non-function-local MDNodes should drop to 
null.

llvm-svn: 102519

Before:
DBG_VALUE %RSI, 0, !-1; dbg:SimpleRegisterCoalescing.cpp:2707
Now:
DBG_VALUE %RSI, 0, !"this"; dbg:SimpleRegisterCoalescing.cpp:2707

llvm-svn: 102518

alignment of globals to the preferred alignment, but only when
there is no section specified on the global (by far the common
case).

llvm-svn: 102515

While lowering dbg_declare, emit DBG_VALUE machine instruction if alloca matching llvm.dbg.declare intrinsic is missing.

llvm-svn: 102513

update them. Computing kill flags is notoriously difficult, and the coalescer
would get it wrong sometimes, and it would completely skip physical registers.

Now we simply remove kill flags based on the live intervals after coalescing.
This is a few percent slower, but now we get correct kill flags for physical
registers after coalescing.

llvm-svn: 102510

instruction.

This instruction would crash the pass:

  INLINEASM <es:foo $0 $1>, 9, %FP0<kill>, 9, %FP0<kill>, 14, %EFLAGS<earlyclobber,def,dead>

Now it doesn't.

llvm-svn: 102509

after all.

llvm-svn: 102508