[PATCH] D27572: AMDGPU: llvm.amdgcn.interp.mov is a source of divergence

[Nicolai Hähnle via Phabricator via llvm-commits]

nhaehnle created this revision.
nhaehnle added reviewers: arsenm, tstellarAMD.
nhaehnle added a subscriber: llvm-commits.
Herald added subscribers: tony-tye, yaxunl, wdng, kzhuravl.

While the result is constant across a single primitive, each pixel
shader wave can have pixels from multiple primitives.


https://reviews.llvm.org/D27572

Files:
  lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp


Index: lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
===================================================================
--- lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -241,6 +241,7 @@
   case Intrinsic::amdgcn_workitem_id_x:
   case Intrinsic::amdgcn_workitem_id_y:
   case Intrinsic::amdgcn_workitem_id_z:
+  case Intrinsic::amdgcn_interp_mov:
   case Intrinsic::amdgcn_interp_p1:
   case Intrinsic::amdgcn_interp_p2:
   case Intrinsic::amdgcn_mbcnt_hi:


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D27572.80755.patch
Type: text/x-patch
Size: 518 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161208/fb422f3e/attachment.bin>