[PATCH] D134477: [X86] Lower vector interleave into unpck and perm

[Han Zhu via Phabricator via llvm-commits]

zhuhan0 added inline comments.


================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:18582-18606
   // There is a really nice hard cut-over between AVX1 and AVX2 that means we
   // can check for those subtargets here and avoid much of the subtarget
   // querying in the per-vector-type lowering routines. With AVX1 we have
   // essentially *zero* ability to manipulate a 256-bit vector with integer
   // types. Since we'll use floating point types there eventually, just
   // immediately cast everything to a float and operate entirely in that domain.
   if (VT.isInteger() && !Subtarget.hasAVX2()) {
----------------
v8i32 is also covered now because of the cast to v8f32 here.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D134477/new/

https://reviews.llvm.org/D134477