diff --git a/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp b/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
--- a/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
+++ b/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
@@ -449,24 +449,10 @@
   return true;
 }
 
-// If the use operand doesn't care about the value, this may be an operand only
-// used for register indexing, in which case it is unsafe to fold.
 bool SIFoldOperands::isUseSafeToFold(const MachineInstr &MI,
                                      const MachineOperand &UseMO) const {
-  if (UseMO.isUndef() || TII->isSDWA(MI))
-    return false;
-
-  switch (MI.getOpcode()) {
-  case AMDGPU::V_MOV_B32_e32:
-  case AMDGPU::V_MOV_B32_e64:
-  case AMDGPU::V_MOV_B64_PSEUDO:
-  case AMDGPU::V_MOV_B64_e32:
-  case AMDGPU::V_MOV_B64_e64:
-    // Do not fold into an indirect mov.
-    return !MI.hasRegisterImplicitUseOperand(AMDGPU::M0);
-  }
-
-  return true;
+  // Operands of SDWA instructions must be registers.
+  return !TII->isSDWA(MI);
 }
 
 // Find a def of the UseReg, check if it is a reg_sequence and find initializers
diff --git a/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp b/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
--- a/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
+++ b/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
@@ -637,9 +637,6 @@
   if (!TRI->isVGPR(*MRI, X))
     return nullptr;
 
-  if (MovT.hasRegisterImplicitUseOperand(AMDGPU::M0))
-    return nullptr;
-
   const unsigned SearchLimit = 16;
   unsigned Count = 0;
   bool KilledT = false;
@@ -654,8 +651,7 @@
          MovY->getOpcode() != AMDGPU::COPY) ||
         !MovY->getOperand(1).isReg()        ||
         MovY->getOperand(1).getReg() != T   ||
-        MovY->getOperand(1).getSubReg() != Tsub ||
-        MovY->hasRegisterImplicitUseOperand(AMDGPU::M0))
+        MovY->getOperand(1).getSubReg() != Tsub)
       continue;
 
     Register Y = MovY->getOperand(0).getReg();
@@ -688,9 +684,6 @@
         MovX = nullptr;
         break;
       }
-      // Implicit use of M0 is an indirect move.
-      if (I->hasRegisterImplicitUseOperand(AMDGPU::M0))
-        continue;
 
       if (Size > 1 && (I->getNumImplicitOperands() > (I->isCopy() ? 0U : 1U)))
         continue;
diff --git a/llvm/test/CodeGen/AMDGPU/v_swap_b32.mir b/llvm/test/CodeGen/AMDGPU/v_swap_b32.mir
--- a/llvm/test/CodeGen/AMDGPU/v_swap_b32.mir
+++ b/llvm/test/CodeGen/AMDGPU/v_swap_b32.mir
@@ -848,7 +848,7 @@
 ...
 
 # GCN-LABEL: name: indirect_mov_t
-# GCN:      $vgpr2 = V_MOV_B32_e32 killed $vgpr0, implicit $exec, implicit $m0
+# GCN:      $vgpr2 = V_MOV_B32_indirect_read killed $vgpr0, implicit $exec, implicit $m0
 # GCN-NEXT: $vgpr0 = V_MOV_B32_e32 killed $vgpr1, implicit $exec
 # GCN-NEXT: $vgpr1 = V_MOV_B32_e32 killed $vgpr2, implicit $exec
 # GCN-NEXT: S_SETPC_B64_return $sgpr30_sgpr31, implicit $vgpr0, implicit $vgpr1
@@ -859,7 +859,7 @@
 body:             |
   bb.0:
     liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
-    $vgpr2 = V_MOV_B32_e32 killed $vgpr0, implicit $exec, implicit $m0
+    $vgpr2 = V_MOV_B32_indirect_read killed $vgpr0, implicit $exec, implicit $m0
     $vgpr0 = V_MOV_B32_e32 killed $vgpr1, implicit $exec
     $vgpr1 = V_MOV_B32_e32 killed $vgpr2, implicit $exec
     S_SETPC_B64_return $sgpr30_sgpr31, implicit $vgpr0, implicit $vgpr1
@@ -867,7 +867,7 @@
 
 # GCN-LABEL: name: indirect_mov_x
 # GCN:      $vgpr2 = V_MOV_B32_e32 killed $vgpr0, implicit $exec
-# GCN-NEXT: $vgpr0 = V_MOV_B32_e32 killed $vgpr1, implicit $exec, implicit $m0
+# GCN-NEXT: $vgpr0 = V_MOV_B32_indirect_read killed $vgpr1, implicit $exec, implicit $m0
 # GCN-NEXT: $vgpr1 = V_MOV_B32_e32 killed $vgpr2, implicit $exec
 # GCN-NEXT: S_SETPC_B64_return $sgpr30_sgpr31, implicit $vgpr0, implicit $vgpr1
 
@@ -878,7 +878,7 @@
   bb.0:
     liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
     $vgpr2 = V_MOV_B32_e32 killed $vgpr0, implicit $exec
-    $vgpr0 = V_MOV_B32_e32 killed $vgpr1, implicit $exec, implicit $m0
+    $vgpr0 = V_MOV_B32_indirect_read killed $vgpr1, implicit $exec, implicit $m0
     $vgpr1 = V_MOV_B32_e32 killed $vgpr2, implicit $exec
     S_SETPC_B64_return $sgpr30_sgpr31, implicit $vgpr0, implicit $vgpr1
 ...
@@ -886,7 +886,7 @@
 # GCN-LABEL: name: indirect_mov_y
 # GCN:      $vgpr2 = V_MOV_B32_e32 killed $vgpr0, implicit $exec
 # GCN-NEXT: $vgpr0 = V_MOV_B32_e32 killed $vgpr1, implicit $exec
-# GCN-NEXT: $vgpr1 = V_MOV_B32_e32 killed $vgpr2, implicit $exec, implicit $m0
+# GCN-NEXT: $vgpr1 = V_MOV_B32_indirect_read killed $vgpr2, implicit $exec, implicit $m0
 # GCN-NEXT: S_SETPC_B64_return $sgpr30_sgpr31, implicit $vgpr0, implicit $vgpr1
 
 ---
@@ -897,7 +897,7 @@
     liveins: $vgpr0, $vgpr1, $sgpr30_sgpr31
     $vgpr2 = V_MOV_B32_e32 killed $vgpr0, implicit $exec
     $vgpr0 = V_MOV_B32_e32 killed $vgpr1, implicit $exec
-    $vgpr1 = V_MOV_B32_e32 killed $vgpr2, implicit $exec, implicit $m0
+    $vgpr1 = V_MOV_B32_indirect_read killed $vgpr2, implicit $exec, implicit $m0
     S_SETPC_B64_return $sgpr30_sgpr31, implicit $vgpr0, implicit $vgpr1
 ...