[PATCH] D40692: AMDGPU/EG: Add a new FeatureFMA and use it to selectively enable FMA instruction

Jan Vesely via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Dec 4 12:55:18 PST 2017


jvesely added inline comments.


================
Comment at: lib/Target/AMDGPU/R600ISelLowering.cpp:214-217
+  if (!Subtarget->hasFMA()) {
+    setOperationAction(ISD::FMA, MVT::f32, Expand);
+    setOperationAction(ISD::FMA, MVT::f64, Expand);
+  }
----------------
arsenm wrote:
> There's not much point in doing this since it will just turn into an unsupported libcall
it produces an error instead of assertion failure, so I can use 'not llc' in the test


Repository:
  rL LLVM

https://reviews.llvm.org/D40692





More information about the llvm-commits mailing list