[PATCH] D44585: [AMDGPU] Scalarize when scalar code cheaper than vector code.
Farhana Aleen via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Apr 10 15:46:00 PDT 2018
FarhanaAleen abandoned this revision.
FarhanaAleen added a comment.
Implementing this transformation in the CodeGen seem to be messy since there is no way to classify scalar operations other than hard-code them. I am exploring other viable clean solutions.
https://reviews.llvm.org/D44585
More information about the llvm-commits
mailing list