[PATCH] D15736: AMDGPU/SI: Select non-uniform constant addrspace loads to flat instructions for HSA

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Wed Dec 23 20:52:45 PST 2015


arsenm accepted this revision.
arsenm added a comment.
This revision is now accepted and ready to land.

LGTM


================
Comment at: lib/Target/AMDGPU/SIInstrInfo.td:147-148
@@ -146,3 +146,4 @@
   return isFlatLoad(dyn_cast<LoadSDNode>(N)) ||
+         isConstantLoad(cast<LoadSDNode>(N), -1) ||
          isGlobalLoad(dyn_cast<LoadSDNode>(N));
 }]>;
----------------
global should be tested first since it is more common


http://reviews.llvm.org/D15736





More information about the llvm-commits mailing list