llvm-svn: 81498

thumb-foo-bar as an ARM target.

llvm-svn: 81497

Follow up to 81494. When the folded reload is narrowed to a 32-bit load then change the destination register to a 32-bit one or add a sub-register index.

llvm-svn: 81496

out an illegal "PHINODE" instruction.

llvm-svn: 81495

It's not legal to fold a load from a narrower stack slot into a wider instruction. If done, the instruction does a 64-bit load and that's not
safe. This can happen we a subreg_to_reg 0 has been coalesced. One
exception is when the instruction that folds the load is a move, then we
can simply turn it into a 32-bit load from the stack slot.                                                                                                                    

rdar://7170444

llvm-svn: 81494

llvm-svn: 81493

isn't legal.

llvm-svn: 81492

folder is now automatically adding.

llvm-svn: 81491

Also, treat the GNU __null as an integral constant expression to match
GCC's behavior.

llvm-svn: 81490

llvm-svn: 81487

llvm-svn: 81486

llvm-svn: 81485

llvm-svn: 81484

how to fold notionally-out-of-bounds array getelementptr indices instead
of just doing these in lib/Analysis/ConstantFolding.cpp, because it can
be done in a fairly general way without TargetData, and because not all
constants are visited by lib/Analysis/ConstantFolding.cpp. This enables
more constant folding.

Also, set the "inbounds" flag when the getelementptr indices are
one-past-the-end.

llvm-svn: 81483

llvm-svn: 81482

llvm-svn: 81481

llvm-svn: 81480

for block pointer.

llvm-svn: 81479

within the notional bounds of the static type of the getelementptr (which
is not the same as "inbounds") from GlobalOpt into a utility routine,
and use it in ConstantFold.cpp to check whether there are any mis-behaved
indices.

llvm-svn: 81478

llvm-svn: 81477

 - Noticed & patched by Nick Kledzik.

 - This may expose portability issues in the .S files, if so we should either
   disable them on non-Darwin, or even better make them portable.

llvm-svn: 81476

llvm-svn: 81475

llvm-svn: 81474

llvm-svn: 81473

llvm-svn: 81472

llvm-svn: 81471

llvm-svn: 81470

llvm-svn: 81469

llvm-svn: 81468

class into a candidate set. 

llvm-svn: 81467

llvm-svn: 81466

llvm-svn: 81465

post-decrement load/store.

llvm-svn: 81464

llvm-svn: 81463

llvm-svn: 81462

that things like .word can be parsed as target specific.  Moved parsing .word
out of AsmParser.cpp into X86AsmParser.cpp as it is 2 bytes on X86 and 4 bytes
for other targets that support the .word directive.

llvm-svn: 81461

Don't check use of a member function declaration used if the member function is virtual and the member reference expression doesn't explicitly qualify it. Fixes PR4878.

llvm-svn: 81460

llvm-svn: 81459

various widths.

llvm-svn: 81458

including segment offsets and (for 8-bit operands)
absolute offsets.

llvm-svn: 81457