[PATCH] D155601: [AArch64][GISel] Additional FPExt vector lowering

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Jul 18 11:28:43 PDT 2023


arsenm added inline comments.


================
Comment at: llvm/test/CodeGen/AArch64/fpext.ll:204
+entry:
+  %c = fpext <3 x half> %a to <3 x float>
+  ret <3 x float> %c
----------------
tschuett wrote:
> What is the trick behind the difference?
the dag widened the vector and gisel scalarized it


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155601/new/

https://reviews.llvm.org/D155601



More information about the llvm-commits mailing list