llvm-svn: 147005

llvm-svn: 147004

MIPS64 can generate constant +0.0 with a single DMTC1 instruction. 

llvm-svn: 146999

Use the spill slot alignment as well as the local variable alignment to
determine when the stack needs to be realigned. This works now that the
ARM target can always realign the stack by using a base pointer.

Still respect the ARMBaseRegisterInfo::canRealignStack() function
vetoing a realigned stack.  Don't use aligned spill code in that case.

llvm-svn: 146997

llvm-svn: 146981

The failure that I see in the current version is:

LLVM ERROR: Cannot select: 0x18b8f70: v4i64 = X86ISD::VZEXT_MOVL 0x18beee0 [ID=14]
  0x18beee0: v4i64 = insert_subvector 0x18b8c70, 0x18b9170, 0x18b9570 [ID=13]
    0x18b8c70: v4i64 = insert_subvector 0x18b9870, 0x18bf4e0, 0x18b9970 [ID=12]
      0x18b9870: v4i64 = undef [ID=4]
      0x18bf4e0: v2i64 = bitcast 0x18bf3e0 [ID=10]
        0x18bf3e0: v4i32 = BUILD_VECTOR 0x18b9770, 0x18b9770, 0x18b9770, 0x18b9770 [ID=8]
          0x18b9770: i32 = TargetConstant<0> [ID=6]
          0x18b9770: i32 = TargetConstant<0> [ID=6]
          0x18b9770: i32 = TargetConstant<0> [ID=6]
          0x18b9770: i32 = TargetConstant<0> [ID=6]
      0x18b9970: i32 = Constant<0> [ID=3]
    0x18b9170: v2i64 = undef [ORD=1] [ID=1]
    0x18b9570: i32 = Constant<2> [ID=5]

llvm-svn: 146975

use the zero-undefined variants of CTTZ and CTLZ. These are just simple
patterns for now, there is more to be done to make real world code using
these constructs be optimized and codegen'ed properly on X86.

The existing tests are spiffed up to check that we no longer generate
unnecessary cmov instructions, and that we generate the very important
'xor' to transform bsr which counts the index of the most significant
one bit to the number of leading (most significant) zero bits. Also they
now check that when the variant with defined zero result is used, the
cmov is still produced.

llvm-svn: 146974

We used to rely on the *eh_sjlj_setjmp instructions to mark that a function
with setjmp/longjmp exception handling clobbers all the registers.  But with
the recent reorganization of ARM EH, those eh_sjlj_setjmp instructions are
expanded away earlier, before PEI can see them to determine what registers to
save and restore.  Mark the dispatchsetup instruction in the same way, since
that instruction cannot be expanded early.  This also more accurately reflects
when the registers are clobbered.

llvm-svn: 146949

llvm-svn: 146923

llvm-svn: 146901

patterns emit a single LUi instruction instead of a pair of LUi and ORi.

llvm-svn: 146900

direct-object emitter should emit the appropriate shift instruction depending
on the shift amount.

llvm-svn: 146893

This change reduces the number of instructions generated.

For example, 
(load (add (sub $n0, $n1), (MipsLo got(s))))

results in the following sequence of instructions:
1. sub $n2, $n0, $n1
2. lw got(s)($n2)

Previously, three instructions were needed.
1. sub $n2, $n0, $n1
2. addiu $n3, $n2, got(s)
3. lw 0($n3)

llvm-svn: 146888

llvm-svn: 146805

llvm-svn: 146800

The bad sorting caused a misaligned basic block when building 176.vpr in
ARM mode.

<rdar://problem/10594653>

llvm-svn: 146767

Reenable the tests.

llvm-svn: 146750

Don't try to match 'unpackl/h v, v' for 32xi8 and 16xi16 when only AVX1 is supported. Fix 'unpackh v, v' for 256-bit types to understand 128-bit lanes.

llvm-svn: 146726

Patch by Elena Demikhovsky <elena.demikhovsky@intel.com>!

llvm-svn: 146689

Patch by Elena Demikhovsky <elena.demikhovsky@intel.com>!

llvm-svn: 146684

llvm-svn: 146678

llvm-svn: 146675

llvm-svn: 146666

Don't try to form FGETSIGN after legalization; it is possible in some cases, but the existing code can't do it correctly. PR11570.

llvm-svn: 146630

rdar://10566486

llvm-svn: 146625

On ARM, peephole optimization for ABS creates a trivial cfg triangle which tempts machine sink to sink instructions in code which is really straight line code. Sometimes this sinking may alter register allocator input such that use and def of a reg is divided by a branch in between, which may result in extra spills. Now mahine sink avoids sinking if final sink destination is post dominator.

Radar 10266272.

llvm-svn: 146604

emission is not supported yet, but a patch that adds the support should follow
soon.

llvm-svn: 146572

  to finalize MI bundles (i.e. add BUNDLE instruction and computing register def
  and use lists of the BUNDLE instruction) and a pass to unpack bundles.
- Teach more of MachineBasic and MachineInstr methods to be bundle aware.
- Switch Thumb2 IT block to MI bundles and delete the hazard recognizer hack to
  prevent IT blocks from being broken apart.

llvm-svn: 146542

llvm-svn: 146538

load and then move the result from a GPR to a FPR.

llvm-svn: 146502

elf-relsym.ll.

llvm-svn: 146470

test cases where there were a lot of relocations applied relative to a large
rodata section. Gas would create a symbol for each of these whereas we would
be relative to the beginning of the rodata section. This change mimics what
gas does.

Patch by Jack Carter.

llvm-svn: 146468

llvm-svn: 146455

llvm-svn: 146433

These modifiers simply select either the low or high D subregister of a Neon
Q register.  I've also removed the unimplemented 'p' modifier, which turns out
to be a bit different than the comment here suggests and as far as I can tell
was only intended for internal use in Apple's version of gcc.

llvm-svn: 146417

llvm-svn: 146412

I followed three heuristics for deciding whether to set 'true' or
'false':

- Everything target independent got 'true' as that is the expected
  common output of the GCC builtins.
- If the target arch only has one way of implementing this operation,
  set the flag in the way that exercises the most of codegen. For most
  architectures this is also the likely path from a GCC builtin, with
  'true' being set. It will (eventually) require lowering away that
  difference, and then lowering to the architecture's operation.
- Otherwise, set the flag differently dependending on which target
  operation should be tested.

Let me know if anyone has any issue with this pattern or would like
specific tests of another form. This should allow the x86 codegen to
just iteratively improve as I teach the backend how to differentiate
between the two forms, and everything else should remain exactly the
same.

llvm-svn: 146370

Fixed bug 9905: Failure in code selection for llvm intrinsics sqrt/exp (fix for FSQRT, FSIN, FCOS, FPOWI, FPOW, FLOG, FLOG2, FLOG10, FEXP, FEXP2). Third attempt: simplified checks in test for armv7-apple-darwin11.

llvm-svn: 146341

llvm-svn: 146332

Fixed bug 9905: Failure in code selection for llvm intrinsics sqrt/exp (fix for
FSQRT, FSIN, FCOS, FPOWI, FPOW, FLOG, FLOG2, FLOG10, FEXP, FEXP2). Second
attempt.

llvm-svn: 146328