- This fixed a number of bugs in if-converter, tail merging, and post-allocation
  scheduler. If-converter now runs branch folding / tail merging first to
  maximize if-conversion opportunities.
- Also changed the t2IT instruction slightly. It now defines the ITSTATE
  register which is read by instructions in the IT block.
- Added Thumb2 specific hazard recognizer to ensure the scheduler doesn't
  change the instruction ordering in the IT block (since IT mask has been
  finalized). It also ensures no other instructions can be scheduled between
  instructions in the IT block.

This is not yet enabled.

llvm-svn: 106344

llvm-svn: 106342

llvm-svn: 106336

(as suggested in radar 8104405).

llvm-svn: 106318

llvm-svn: 106314

limit on the offset that can be materialized without using the register
scavenger.

llvm-svn: 106312

ARM tail calls.  Don't know if it works, but it
doesn't break Darwin.

llvm-svn: 106309

basic tests.

This has been well tested on Darwin but not elsewhere.
It should work provided the linker correctly resolves
  B.W  <label in other function>
which it has not seen before, at least from llvm-based
compilers.  I'm leaving the arm-tail-calls switch in
until I see if there's any problems because of that;
it might need to be disabled for some environments.

llvm-svn: 106299

MachineRegisterInfo doesn't have to confusingly allocate an extra
entry.

llvm-svn: 106296

Not turning them on yet.

llvm-svn: 106295

does for {flags}. If we create virtual registers of the CCR class, RegAllocFast
may try to spill them, and we can't do that.

llvm-svn: 106289

llvm-svn: 106279

addresses a longstanding deficiency noted in many FIXMEs scattered
across all the targets.

This effectively moves the problem up one level, replacing eleven
FIXMEs in the targets with eight FIXMEs in CodeGen, plus one path
through FastISel where we actually supply a DebugLoc, fixing Radar
7421831.

llvm-svn: 106243

ISD::MEMBARRIER. v7 and v7 ARM mode continue to use the custom lowering.

llvm-svn: 106204

previously would result in 'cannot yet select' errors.

llvm-svn: 106199

llvm-svn: 106173

This is before LiveVariables anyway, where these kill flags are recalculated.

llvm-svn: 106157

now, so there's no need to disable them.

llvm-svn: 106155

Make post-ra scheduling, anti-dep breaking, and register scavenger (conservatively) aware of predicated instructions. This enables ARM to move if-conversion before post-ra scheduler.

llvm-svn: 106091

llvm-svn: 106058

call must not be callee-saved; following x86, add a new
regclass to represent this.  Also fixes a couple of bugs.
Still disabled by default; Thumb doesn't work yet.

llvm-svn: 106053

llvm-svn: 106030

llvm-svn: 105994

llvm-svn: 105990

the combined load/store instruction. rdar://7797940

llvm-svn: 105982

immediate" operands.  These functions have so far only been used for VMOV
but they also apply to other NEON instructions with modified immediate
operands.  No functional changes.

llvm-svn: 105969

i64 and f64 types, but now it also handle Neon vector types, so the f64 result
of VMOVDRR may need to be converted to a Neon type.  Radar 8084742.

llvm-svn: 105845

the machine instruction representation of the immediate value to be encoded
into an integer with similar fields as the actual VMOV instruction.  This makes
things easier for the disassembler, since it can just stuff the bits into the
immediate operand, but harder for the asm printer since it has to decode the
value to be printed.  Testcase for the encoding will follow later when MC has
more support for ARM.

llvm-svn: 105836

llvm-svn: 105774

dbg_value immediately follows a sequence of ldr/str instructions that should
be combined into an ldm/stm and is the last instruction in the block, then
combine may end up being skipped.

llvm-svn: 105758

Allow target to place 2-address pass inserted copies in better spots. Thumb2 will use this to try to avoid breaking up IT blocks.

llvm-svn: 105745

llvm-svn: 105677

the same condition, it's important to make sure they are scheduled together
to avoid forming multiple IT blocks. I'm adding a pre-regalloc pass that forms
IT blocks early (by re-scheduling instructions and split basic blocks) to
attempt to fix this. This is not turned on by default since I am not sure this
is the right fix.

Another issue is llvm selects are modeled as two-address conditional moves.
This can be very bad when the copies before the conditional moves are not
coalesced away. Teach IT formation pass to move the copies above the IT block
(when legal) to avoid breaking the IT block.

llvm-svn: 105669

llvm-svn: 105653

immediates to avoid breaking the build.

llvm-svn: 105652

llvm-svn: 105634

llvm-svn: 105591

- change isShuffleMaskLegal to show that all shuffles with 32-bit and 64-bit
  elements are legal
- the Neon shuffle instructions do not support 64-bit elements, but we were
  not checking for that before lowering shuffles to use them
- remove some 64-bit element vduplane patterns that are no longer needed

llvm-svn: 105586

rdar://7797940

llvm-svn: 105557

In file included from X86InstrInfo.cpp:16:
X86GenInstrInfo.inc:2789: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2790: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2792: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2793: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2808: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2809: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2816: error: integer constant is too large for 'long' type
X86GenInstrInfo.inc:2817: error: integer constant is too large for 'long' type

llvm-svn: 105524