[PATCH] D154738: [SLP]Introduce isLegalVectorOp to check if the vector instruction is going to be scalarized.

[Alexey Bataev via Phabricator via llvm-commits]

ABataev added inline comments.


================
Comment at: llvm/test/Transforms/SLPVectorizer/AMDGPU/packed-math.ll:5
 
 ; FIXME: Should still like to vectorize the memory operations for VI
 
----------------
arsenm wrote:
> This fixme was fixed but has now been unfixed
Hmm, I checked VI tests and loooks like fmul <2 x half> tests are scalarized in the end (https://godbolt.org/z/hPcjoETWT). So, it does not worth it to vectorize them.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D154738/new/

https://reviews.llvm.org/D154738