Commit Graph

4 Commits

Author SHA1 Message Date
Jim Grosbach d3c7942f4a Use an explicit triple in testcase.
Make the test work better on non-darwin hosts. Hopefully.

llvm-svn: 213801
2014-07-23 20:46:32 +00:00
Jim Grosbach 724e438c62 [X86,AArch64] Extend vcmp w/ unary op combine to work w/ more constants.
The transform to constant fold unary operations with an AND across a
vector comparison applies when the constant is not a splat of a scalar
as well.

llvm-svn: 213800
2014-07-23 20:41:43 +00:00
Jim Grosbach 8f6f0858ec X86: restrict combine to when type sizes are safe.
The folding of unary operations through a vector compare and mask operation
is only safe if the unary operation result is of the same size as its input.
For example, it's not safe for [su]itofp from v4i32 to v4f64.

llvm-svn: 213799
2014-07-23 20:41:38 +00:00
Jim Grosbach f7502c4884 AArch64: Constant fold converting vector setcc results to float.
Since the result of a SETCC for AArch64 is 0 or -1 in each lane, we can
move unary operations, in this case [su]int_to_fp through the mask
operation and constant fold the operation away. Generally speaking:
  UNARYOP(AND(VECTOR_CMP(x,y), constant))
      --> AND(VECTOR_CMP(x,y), constant2)
where constant2 is UNARYOP(constant).

This implements the transform where UNARYOP is [su]int_to_fp.

For example, consider the simple function:
define <4 x float> @foo(<4 x float> %val, <4 x float> %test) nounwind {
  %cmp = fcmp oeq <4 x float> %val, %test
  %ext = zext <4 x i1> %cmp to <4 x i32>
  %result = sitofp <4 x i32> %ext to <4 x float>
  ret <4 x float> %result
}

Before this change, the code is generated as:
  fcmeq.4s  v0, v0, v1
  movi.4s v1, #0x1        // Integer splat value.
  and.16b v0, v0, v1      // Mask lanes based on the comparison.
  scvtf.4s  v0, v0        // Convert each lane to f32.
  ret

After, the code is improved to:
  fcmeq.4s  v0, v0, v1
  fmov.4s v1, #1.00000000 // f32 splat value.
  and.16b v0, v0, v1      // Mask lanes based on the comparison.
  ret

The svvtf.4s has been constant folded away and the floating point 1.0f
vector lanes are materialized directly via fmov.4s.

Rather than do the folding manually in the target code, teach getNode()
in the generic SelectionDAG to handle folding constant operands of
vector [su]int_to_fp nodes. It is reasonable (as noted in a FIXME) to do
additional constant folding there as well, but I don't have test cases
for those operations, so leaving them for another time when it becomes
appropriate.

rdar://17693791

llvm-svn: 213341
2014-07-18 00:40:52 +00:00