Index: lib/Target/AMDGPU/SIInstrInfo.td
===================================================================
--- lib/Target/AMDGPU/SIInstrInfo.td
+++ lib/Target/AMDGPU/SIInstrInfo.td
@@ -2119,8 +2119,8 @@
                          list<dag> pat32, list<dag> pat64,
                          string revOp, bit useSGPRInput> {
 
-  let SchedRW = [Write32Bit, WriteSALU] in {
-    let Uses = !if(useSGPRInput, [VCC, EXEC], [EXEC]), Defs = [VCC] in {
+  let SchedRW = [Write32Bit] in {
+    let Uses = !if(useSGPRInput, [VCC, EXEC], [EXEC]) in {
       defm _e32 : VOP2_m <op, opName, p, pat32, revOp>;
     }
 
Index: test/CodeGen/AMDGPU/cndmask-no-def-vcc.ll
===================================================================
--- /dev/null
+++ test/CodeGen/AMDGPU/cndmask-no-def-vcc.ll
@@ -0,0 +1,27 @@
+; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+
+; Produces error after adding an implicit deff to v_cndmask_b32
+
+; GCN-LABEL: {{^}}vcc_shrink_vcc_def:
+; GCN: v_cmp_eq_i32_e64 vcc, 0, s{{[0-9]+}}
+; GCN: v_cndmask_b32_e32 v{{[0-9]+}}, 1.0, v{{[0-9]+}}, vcc
+; GCN: v_cndmask_b32_e64 v1, 0, 1, s{{\[[0-9]+:[0-9]+\]}}
+define void @vcc_shrink_vcc_def(float %arg, i32 %arg1, float %arg2, i32 %arg3) {
+bb0:
+  %tmp = icmp sgt i32 %arg1, 4
+  %c = icmp eq i32 %arg3, 0
+  %tmp4 = select i1 %c, float %arg, float 1.000000e+00
+  %tmp5 = fcmp ogt float %arg2, 0.000000e+00
+  %tmp6 = fcmp olt float %arg2, 1.000000e+00
+  %tmp7 = fcmp olt float %arg, %tmp4
+  %tmp8 = and i1 %tmp5, %tmp6
+  %tmp9 = and i1 %tmp8, %tmp7
+  br i1 %tmp9, label %bb1, label %bb2
+
+bb1:
+  store volatile i32 0, i32 addrspace(1)* undef
+  br label %bb2
+
+bb2:
+  ret void
+}