llvm-svn: 85640

llvm-svn: 85637

unfolding loads for hoisting.  getOpcodeAfterMemoryUnfold returns the
opcode of the original operation without the load, not the load
itself, MachineLICM needs to know the operand index in order to get
the correct register class. Extend getOpcodeAfterMemoryUnfold to
return this information.

llvm-svn: 85622

llvm-svn: 85608

void f (int a1, int a2, int a3, int a4, int a5,...)

In ARMTargetLowering::LowerFormalArguments if the function has 4 or
more regular arguments we used to set VarArgsFrameIndex using an
offset of 0, which is only correct if the function has exactly 4
regular arguments.

llvm-svn: 85590

common tail, except when the OptimizeForSize function attribute is present.
Radar 7338114.

llvm-svn: 85441

otherwise unhoistable instructions in order to allow the loads
to be hoisted.

llvm-svn: 85364

llvm-svn: 85362

MachineLICM and other things which run before LiveVariables is run.

llvm-svn: 85360

if they have compatible encodings.

llvm-svn: 85359

llvm-svn: 85266

llvm-svn: 85210

llvm-svn: 85208

llvm-svn: 85207

llvm-svn: 85204

llvm-svn: 85127

strides for now, because it doesn't handle them correctly. This fixes a
miscompile of SingleSource/Benchmarks/Misc-C++/ray.

This problem was usually hidden because indvars transforms such induction
variables into negations of canonical induction variables.

llvm-svn: 85118

  bootstrapping. It's not safe to leave identity subreg_to_reg and insert_subreg
  around.
- Relax register scavenging to allow use of partially "not-live" registers. It's
  common for targets to operate on registers where the top bits are undef. e.g.
  s0 =
  d0 = insert_subreg d0<undef>, s0, 1
  ...
     = d0
  When the insert_subreg is eliminated by the coalescer, the scavenger used to
  complain. The previous fix was to keep to insert_subreg around. But that's
  brittle and it's overly conservative when we want to use the scavenger to 
  allocate registers. It's actually legal and desirable for other instructions
  to use the "undef" part of d0. e.g.
  s0 =
  d0 = insert_subreg d0<undef>, s0, 1
  ...
  s1 =
     = s1
     = d0
  We probably need add a "partial-undef" marker on machine operand so the
  machine verifier would not complain.

llvm-svn: 85091

or global after a function with conflicting names.  Update some testcases
that were accidentally depending on this behavior.

llvm-svn: 85081

llvm-svn: 85051

llvm-svn: 85050

the scale overflows pointer-sized arithmetic. This fixes PR5281.

llvm-svn: 84954

llvm-svn: 84932

Allow the target to select the level of anti-dependence breaking that should be performed by the post-RA scheduler. The default is none.

llvm-svn: 84911

llvm-svn: 84868

Move if-conversion before post-regalloc scheduling so the predicated instruction get scheduled properly.

llvm-svn: 84843

llvm-svn: 84813

Yay for ASCII graphics!

llvm-svn: 84808

to break up CFG diamonds by banishing one of the blocks to the end of
the function, which is bad for code density and branch size.

This does pessimize MultiSource/Benchmarks/Ptrdist/yacr2, the
benchmark cited as the reason for the change, however I've examined
the code and it looks more like a case of gaming a particular
branch than of being generally applicable.

llvm-svn: 84803

immediate operands when they will fit into the using instruction.

llvm-svn: 84778

llvm-svn: 84776

RMW preprocessing stuff was incorrect. Grab the stuff from x86 backend and disable some tests until it will be clever enough to handle them.

llvm-svn: 84775

llvm-svn: 84751

llvm-svn: 84727

llvm-svn: 84705

transform.

llvm-svn: 84683

llvm-svn: 84658

encounters an OEQ or UNE comparison, and update its callers to check
for this return status and recover. This fixes a problem resulting from
the LowerOperation hooks being called from LegalizeVectorOps, because
LegalizeVectorOps only lowers vectors, so OEQ and UNE comparisons may
still be at large. This fixes PR5092.

llvm-svn: 84640

tracked. Instead of trying to manually keep track of these locations
while doing complex modifications, just recompute them when they're needed.
This fixes a bug in which the TopMBB and BotMBB were not correctly updated,
leading to invalid transformations.

llvm-svn: 84598

llvm-svn: 84586