llvm-svn: 116414

llvm-svn: 116412

arithmetic-with-carry-in instructions.

llvm-svn: 116384

and move to a custom operand encoder. Remove the last of the special handling
stuff from ARMMCCodeEmitter::EncodeInstruction.

llvm-svn: 116377

explicit handling of the instructions referencing it from the MC code
emitter.

llvm-svn: 116367

llvm-svn: 116321

ARM instructions.

llvm-svn: 116313

register operand.

llvm-svn: 116259

matching in tblgen to do the predicate operand.

llvm-svn: 116213

concept level stuff at this point, but it is generally working for those
instructions that know how to map the operands.

This patch fills in the register operands for add/sub/or/etc instructions
and adds the conditional execution predicate encoding.

llvm-svn: 116112

llvm-svn: 115884

llvm-svn: 115853

llvm-svn: 115845

  allow target to correctly compute latency for cases where static scheduling
  itineraries isn't sufficient. e.g. variable_ops instructions such as
  ARM::ldm.
  This also allows target without scheduling itineraries to compute operand
  latencies. e.g. X86 can return (approximated) latencies for high latency
  instructions such as division.
- Compute operand latencies for those defined by load multiple instructions,
  e.g. ldm and those used by store multiple instructions, e.g. stm.

llvm-svn: 115755

llvm-svn: 115370

llvm-svn: 115193

llvm-svn: 115160

Now that the MC lowering handles the expansion of the pseudos, kill the horrible
blobs of text.

llvm-svn: 115130

1. Cortex-a9 8-bit and 16-bit loads / stores AGU cycles are 1 cycle longer than 32-bit ones.
2. Cortex-a9 is out-of-order so model all read cycles as cycle 1.
3. Lots of other random fixes for A8 and A9.

llvm-svn: 115121

which require the use of the shifter-operand. This will be used to split
the ldr/str instructions such that those versions needing the shifter operand
can get a different scheduling itenerary, as in some cases, the use of the
shifter can cause different scheduling than the simpler forms.

llvm-svn: 115066

llvm-svn: 115010

Assign bitwise binary instructions different itinerary classes from ALU instructions such as add / sub.

llvm-svn: 115008

llvm-svn: 114780

Fix scheduling itinerary for pseudo mov immediate instructions which expand into two real instructions.

llvm-svn: 114766

reflection, this isn't going to achieve the purpose I intended it for.  Back to the drawing board!

llvm-svn: 114710

llvm-svn: 114706

llvm-svn: 114705

llvm-svn: 114703

(yet) recognize the 'trap' mnemonic, so we use .short/.long to emit the
opcode directly. On Darwin, however, we do want the mnemonic for more
readable assembly code and better disassembly.

Adjust the .td file to use the 'trap' mnemonic and handle using the binutils
workaround in the assembly printer. Also tweak the formatting of the opcode
values to make them consistent between the MC printer and the old printer.

llvm-svn: 114679

passed the root of the match, even though only a few patterns
actually needed this (one in X86, several in ARM [which should
be refactored anyway], and some in CellSPU that I don't feel 
like detangling).   Instead of requiring all ComplexPatterns to
take the dead root, have targets opt into getting the root by
putting SDNPWantRoot on the ComplexPattern.

llvm-svn: 114471

llvm-svn: 113584

llvm-svn: 113435

the testcases should be merged.

llvm-svn: 112711

int x(int t) {
  if (t & 256)
    return -26;
  return 0;
}

We generate this:

     tst.w   r0, #256
     mvn     r0, #25
     it      eq
     moveq   r0, #0

while gcc generates this:

     ands    r0, r0, #256
     it      ne
     mvnne   r0, #25
     bx      lr

Scandalous really!

During ISel time, we can look for this particular pattern. One where we have a
"MOVCC" that uses the flag off of a CMPZ that itself is comparing an AND
instruction to 0. Something like this (greatly simplified):

  %r0 = ISD::AND ...
  ARMISD::CMPZ %r0, 0         @ sets [CPSR]
  %r0 = ARMISD::MOVCC 0, -26  @ reads [CPSR]

All we have to do is convert the "ISD::AND" into an "ARM::ANDS" that sets [CPSR]
when it's zero. The zero value will all ready be in the %r0 register and we only
need to change it if the AND wasn't zero. Easy!

llvm-svn: 112664

llvm-svn: 112654

Make ARM add rN, sp, #imm instructions rematerializable. That's how the address of locals is calculated, so this should
help relieve register pressure a bit. Recalculating the local address is
almost always going to be better than spilling.

llvm-svn: 112503

llvm-svn: 112421

llvm-svn: 112394

llvm-svn: 112206

llvm-svn: 111277