[llvm] r356470 - [AMDGPU] Add convergent attribute to WWM.

[Neil Henning via llvm-commits]

Author: sheredom
Date: Tue Mar 19 09:32:24 2019
New Revision: 356470

URL: http://llvm.org/viewvc/llvm-project?rev=356470&view=rev
Log:
[AMDGPU] Add convergent attribute to WWM.

Add the convergent attribute to the WWM intrinsic to stop it ever being
sunk out of cfg.

Differential Revision: https://reviews.llvm.org/D59536

Modified:
    llvm/trunk/include/llvm/IR/IntrinsicsAMDGPU.td

Modified: llvm/trunk/include/llvm/IR/IntrinsicsAMDGPU.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/include/llvm/IR/IntrinsicsAMDGPU.td?rev=356470&r1=356469&r2=356470&view=diff
==============================================================================
--- llvm/trunk/include/llvm/IR/IntrinsicsAMDGPU.td (original)
+++ llvm/trunk/include/llvm/IR/IntrinsicsAMDGPU.td Tue Mar 19 09:32:24 2019
@@ -1361,7 +1361,7 @@ def int_amdgcn_kill : Intrinsic<[], [llv
 // enabled, with a few exceptions: - Phi nodes with require WWM return an
 // undefined value.
 def int_amdgcn_wwm : Intrinsic<[llvm_any_ty],
-  [LLVMMatchType<0>], [IntrNoMem, IntrSpeculatable]
+  [LLVMMatchType<0>], [IntrNoMem, IntrSpeculatable, IntrConvergent]
 >;
 
 // Given a value, copies it while setting all the inactive lanes to a given