Index: lib/Target/AMDGPU/SIShrinkInstructions.cpp
===================================================================
--- lib/Target/AMDGPU/SIShrinkInstructions.cpp
+++ lib/Target/AMDGPU/SIShrinkInstructions.cpp
@@ -241,6 +241,32 @@
           }
         }
       }
+      // Combine adjacent s_nops to use the immediate operand encoding how long
+      // to wait.
+      //
+      // s_nop N
+      // s_nop M
+      //  =>
+      // s_nop (N + M)
+      if (MI.getOpcode() == AMDGPU::S_NOP &&
+          Next != MBB.end() &&
+          (*Next).getOpcode() == AMDGPU::S_NOP) {
+
+        MachineInstr &NextMI = *Next;
+        // The instruction encodes the amount to wait with an offset of 1,
+        // i.e. 0 is wait 1 cycle. Convert both to cycles and then convert back
+        // after adding.
+        uint8_t Nop0 = MI.getOperand(0).getImm() + 1;
+        uint8_t Nop1 = NextMI.getOperand(0).getImm() + 1;
+
+        // Make sure we don't overflow the bounds.
+        if (Nop0 + Nop1 <= 8) {
+          NextMI.getOperand(0).setImm(Nop0 + Nop1 - 1);
+          MI.eraseFromParent();
+        }
+
+        continue;
+      }
 
       // FIXME: We also need to consider movs of constant operands since
       // immediate operands are not folded if they have more than one use, and