[PATCH] D111657: [SVE][CodeGen] Enable reciprocal estimates for scalable fdiv/fsqrt

[Kerry McLaughlin via Phabricator via llvm-commits]

kmclaughlin added inline comments.


================
Comment at: llvm/lib/Target/AArch64/AArch64ISelLowering.cpp:8150
+       (ST->hasSVE() &&
+        (VT == MVT::nxv2f32 || VT == MVT::nxv4f32 || VT == MVT::nxv2f64))) {
     if (ExtraSteps == TargetLoweringBase::ReciprocalEstimate::Unspecified)
----------------
paulwalker-arm wrote:
> Out of interest is there a reason we ignore f16 vectors here?
When I created this patch I thought that `DAGCombiner::BuildDivEstimate/BuildSqrtEstimate` didn't support f16 types, which was incorrect. I've added the f16 vector types here & added tests for these to sve-fp-reciprocal.ll.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D111657/new/

https://reviews.llvm.org/D111657