[PATCH] D29942: Add custom lowering for llvm.log.f32 intrinsic
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Feb 14 10:03:12 PST 2017
arsenm added a comment.
Needs tests
================
Comment at: lib/Target/AMDGPU/SIISelLowering.cpp:297
+ setOperationAction(ISD::FLOG, MVT::f32, Custom);
+ setOperationAction(ISD::FLOG, MVT::f64, Custom);
----------------
Should also handle f16?
================
Comment at: lib/Target/AMDGPU/SIISelLowering.cpp:3478
+ SDValue Log2Operand = DAG.getNode(ISD::FLOG2, SL, VT, Operand);
+ const SDValue Log2e = DAG.getConstantFP(M_LOG2E, SL, VT);
+
----------------
I've been meaning to remove the dependent on the math.h constants. Can you define this somewhere locally?
Repository:
rL LLVM
https://reviews.llvm.org/D29942
More information about the llvm-commits
mailing list