[PATCH] D79718: [CGP] include funnel shift intrinsic when sinking splatted operands
Simon Pilgrim via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon May 11 09:39:57 PDT 2020
RKSimon added inline comments.
================
Comment at: llvm/lib/CodeGen/CodeGenPrepare.cpp:6439
+ // We are also using shift cost as a proxy for rotate and funnel shift ops.
if (!TLI->isVectorShiftByScalarCheap(SVI->getType()))
return false;
----------------
Please can you mention that we use it for funnels/rotates in isVectorShiftByScalarCheap descriptions in TargetLowering.h/X86ISelLowering.cpp
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D79718/new/
https://reviews.llvm.org/D79718
More information about the llvm-commits
mailing list