[llvm-bugs] [Bug 26914] New: [AVX512] lowerV4X128VectorShuffle doesn't correctly select input operands

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=26914

            Bug ID: 26914
           Summary: [AVX512] lowerV4X128VectorShuffle doesn't correctly
                    select input operands
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Windows NT
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
          Assignee: unassignedbugs at nondot.org
          Reporter: llvm-dev at redking.me.uk
                CC: elena.demikhovsky at intel.com, igor.breger at intel.com,
                    llvm-bugs at lists.llvm.org
    Classification: Unclassified

define <8 x double> @shuffle_v8f64_23014567(<8 x double> %a0, <8 x double> %a1)
{
  %1 = shufflevector <8 x double> %a1, <8 x double> undef, <8 x i32> <i32 2,
i32 3, i32 0, i32 1, i32 4, i32 5, i32 6, i32 7>
  ret <8 x double> %1
}

Gives:
vshuff64x2    $225, %zmm0, %zmm1, %zmm0 # zmm0 = zmm1[2,3,0,1],zmm0[4,5,6,7]
retq

I'd expect:
vshuff64x2    $225, %zmm1, %zmm1, %zmm0 # zmm0 = zmm1[2,3,0,1,4,5,6,7]
retq

Found whilst investigating adding X86ISD::SHUF128 to the target shuffle combine
list.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20160311/249dfad7/attachment-0001.html>