diff --git a/externals/llvm-project b/externals/llvm-project
index 585523750..f9031f00f 160000
--- a/externals/llvm-project
+++ b/externals/llvm-project
@@ -1 +1 @@
-Subproject commit 585523750e2bbe374d1cb3bf4ff9d53de29b9593
+Subproject commit f9031f00f2c90bc0af274b45ec3e169b5250a688
diff --git a/test/Conversion/TorchToSCF/basic.mlir b/test/Conversion/TorchToSCF/basic.mlir
index dd64e99b8..ccd1b7998 100644
--- a/test/Conversion/TorchToSCF/basic.mlir
+++ b/test/Conversion/TorchToSCF/basic.mlir
@@ -124,8 +124,8 @@ func.func @torch.prim.loop$while(%arg0: !torch.int) -> !torch.float {
 // CHECK-NEXT:      %[[VAL_1:.*]] = torch_c.to_f64 %[[TORCH_VAL_1]]
 // CHECK-NEXT:      scf.yield %[[BLOCK_CONDITION]], %[[VAL_0]], %[[VAL_1]] : i1, f64, f64
 // CHECK-NEXT:    }
-// CHECK-NEXT:    %[[TORCH_LOOP_0:.*]] = torch_c.from_f64 %[[LOOP]]#0
-// CHECK-NEXT:    %[[TORCH_LOOP_1:.*]] = torch_c.from_f64 %[[LOOP]]#1
+// CHECK-DAG:    %[[TORCH_LOOP_0:.*]] = torch_c.from_f64 %[[LOOP]]#0
+// CHECK-DAG:    %[[TORCH_LOOP_1:.*]] = torch_c.from_f64 %[[LOOP]]#1
 // CHECK-NEXT:    return %[[TORCH_LOOP_0]], %[[TORCH_LOOP_1]] : !torch.float, !torch.float
 func.func @torch.prim.loop$while_with_multiple_values() -> (!torch.float, !torch.float) {
   %float3.200000e00 = torch.constant.float 3.200000e+00
@@ -198,8 +198,8 @@ func.func @torch.prim.Loop$for(%arg0: !torch.int) -> !torch.float {
 // CHECK-NEXT:      %[[VAL_1:.*]] = torch_c.to_f64 %[[TORCH_VAL_1]]
 // CHECK-NEXT:      scf.yield %[[VAL_0]], %[[VAL_1]] : f64, f64
 // CHECK-NEXT:    }
-// CHECK-NEXT:    %[[RETURN_0:.*]] = torch_c.from_f64 %[[LOOP]]#0
-// CHECK-NEXT:    %[[RETURN_1:.*]] = torch_c.from_f64 %[[LOOP]]#1
+// CHECK-DAG:    %[[RETURN_0:.*]] = torch_c.from_f64 %[[LOOP]]#0
+// CHECK-DAG:    %[[RETURN_1:.*]] = torch_c.from_f64 %[[LOOP]]#1
 // CHECK-NEXT:    return %[[RETURN_0]], %[[RETURN_1]] : !torch.float, !torch.float
 // CHECK-NEXT:  }
 func.func @torch.prim.Loop$for_with_multiple_results(%arg0: !torch.int) -> (!torch.float, !torch.float) {
diff --git a/test/Conversion/TorchToStablehlo/basic.mlir b/test/Conversion/TorchToStablehlo/basic.mlir
index 0690fb339..c46328095 100644
--- a/test/Conversion/TorchToStablehlo/basic.mlir
+++ b/test/Conversion/TorchToStablehlo/basic.mlir
@@ -40,10 +40,8 @@ func.func @torch.prim.NumToTensor.Scalar$basic() -> !torch.vtensor<[], si64> {
 
 // CHECK-LABEL:   func.func @torch.aten.contiguous(
 // CHECK-SAME:                                     %[[VAL_0:.*]]: !torch.vtensor<[4,64],f32>) -> !torch.vtensor<[4,64],f32> {
-// CHECK:           %[[VAL_1:.*]] = torch_c.to_builtin_tensor %[[VAL_0]] : !torch.vtensor<[4,64],f32> -> tensor<4x64xf32>
 // CHECK:           %int0 = torch.constant.int 0
-// CHECK:           %[[VAL_2:.*]] = torch_c.from_builtin_tensor %[[VAL_1]] : tensor<4x64xf32> -> !torch.vtensor<[4,64],f32>
-// CHECK:           return %[[VAL_2]] : !torch.vtensor<[4,64],f32>
+// CHECK:           return %[[VAL_0]] : !torch.vtensor<[4,64],f32>
 func.func @torch.aten.contiguous(%arg0: !torch.vtensor<[4,64],f32>) -> !torch.vtensor<[4,64],f32> {
   %int0 = torch.constant.int 0
   %0 = torch.aten.contiguous %arg0, %int0 : !torch.vtensor<[4,64],f32>, !torch.int -> !torch.vtensor<[4,64],f32>
diff --git a/test/Conversion/TorchToStablehlo/pooling.mlir b/test/Conversion/TorchToStablehlo/pooling.mlir
index 537ed9ca5..f44d51c9f 100644
--- a/test/Conversion/TorchToStablehlo/pooling.mlir
+++ b/test/Conversion/TorchToStablehlo/pooling.mlir
@@ -103,8 +103,8 @@ func.func @torch.aten.max_pool2d$padding(%arg0: !torch.vtensor<[?,?,?,?],f32>) -
 // CHECK:         %[[T21:.*]] = stablehlo.select %[[T18]], %[[T19]], %[[T20]] : tensor<i1>, tensor<i64>
 // CHECK:         stablehlo.return %[[T17]], %[[T21]] : tensor<f32>, tensor<i64>
 // CHECK:         }) : (tensor<?x?x?xf32>, tensor<?x?x?xi64>, tensor<f32>, tensor<i64>) -> (tensor<?x?x?xf32>, tensor<?x?x?xi64>)
-// CHECK:         %[[T14:.*]] = torch_c.from_builtin_tensor %[[T13]]#0 : tensor<?x?x?xf32> -> !torch.vtensor<[?,?,?],f32>
-// CHECK:         %[[T15:.*]] = torch_c.from_builtin_tensor %[[T13]]#1 : tensor<?x?x?xi64> -> !torch.vtensor<[?,?,?],si64>
+// CHECK-DAG:         %[[T14:.*]] = torch_c.from_builtin_tensor %[[T13]]#0 : tensor<?x?x?xf32> -> !torch.vtensor<[?,?,?],f32>
+// CHECK-DAG:         %[[T15:.*]] = torch_c.from_builtin_tensor %[[T13]]#1 : tensor<?x?x?xi64> -> !torch.vtensor<[?,?,?],si64>
 // CHECK:         return %[[T14]], %[[T15]] : !torch.vtensor<[?,?,?],f32>, !torch.vtensor<[?,?,?],si64>
 func.func @torch.aten.max_pool2d_with_indices(%arg0: !torch.vtensor<[?,?,?],f32>) -> (!torch.vtensor<[?,?,?],f32>, !torch.vtensor<[?,?,?],si64>) {
   %int3 = torch.constant.int 3
diff --git a/test/Conversion/TorchToTosa/basic.mlir b/test/Conversion/TorchToTosa/basic.mlir
index 35007f2a2..3972e2fd4 100644
--- a/test/Conversion/TorchToTosa/basic.mlir
+++ b/test/Conversion/TorchToTosa/basic.mlir
@@ -801,10 +801,8 @@ func.func @torch.aten.unsqueeze$negative_dim(%arg0: !torch.vtensor<[4,3],si32> )
 
 // CHECK-LABEL:   func.func @torch.aten.contiguous$basic(
 // CHECK-SAME:                                      %[[VAL_0:.*]]: !torch.vtensor<[?,?],f32>) -> !torch.vtensor<[?,?],f32> {
-// CHECK:           %[[VAL_1:.*]] = torch_c.to_builtin_tensor %[[VAL_0]] : !torch.vtensor<[?,?],f32> -> tensor<?x?xf32>
 // CHECK:           %[[VAL_2:.*]] = torch.constant.int 0
-// CHECK:           %[[VAL_3:.*]] = torch_c.from_builtin_tensor %[[VAL_1]] : tensor<?x?xf32> -> !torch.vtensor<[?,?],f32>
-// CHECK:           return %[[VAL_3]] : !torch.vtensor<[?,?],f32>
+// CHECK:           return %[[VAL_0]] : !torch.vtensor<[?,?],f32>
 // CHECK:         }
 func.func @torch.aten.contiguous$basic(%arg0: !torch.vtensor<[?,?],f32> ) -> !torch.vtensor<[?,?],f32> {
   %int0 = torch.constant.int 0
diff --git a/test/Dialect/TMTensor/bufferize.mlir b/test/Dialect/TMTensor/bufferize.mlir
index 0fd0e2dcc..7c4a5798c 100644
--- a/test/Dialect/TMTensor/bufferize.mlir
+++ b/test/Dialect/TMTensor/bufferize.mlir
@@ -13,8 +13,8 @@
 // CHECK:             %[[OUT_CURRENT_ELEMENT:.*]] = arith.addi %[[OUT_PREV_ELEMENT]], %[[IN_ELEMENT]] : i32
 // CHECK:             tm_tensor.yield %[[OUT_CURRENT_ELEMENT]] : i32
 // CHECK:           }
-// CHECK:           %[[OUT_TENSOR_NEW:.*]] = bufferization.to_tensor %[[OUT_MEMREF_NEW]] : memref<128xi32>
-// CHECK:           %[[ACC_TENSOR_NEW:.*]] = bufferization.to_tensor %[[ACC_MEMREF_NEW]] : memref<i32>
+// CHECK-DAG:           %[[OUT_TENSOR_NEW:.*]] = bufferization.to_tensor %[[OUT_MEMREF_NEW]] : memref<128xi32>
+// CHECK-DAG:           %[[ACC_TENSOR_NEW:.*]] = bufferization.to_tensor %[[ACC_MEMREF_NEW]] : memref<i32>
 // CHECK:           return %[[OUT_TENSOR_NEW]], %[[ACC_TENSOR_NEW]] : tensor<128xi32>, tensor<i32>
 func.func @scan_1d_inclusive(%in: tensor<128xi32>, %out: tensor<128xi32>, %acc: tensor<i32>) -> (tensor<128xi32>, tensor<i32>) {
   %ret_out, %ret_acc = tm_tensor.scan dimension(0) inclusive(true)
@@ -41,8 +41,8 @@ func.func @scan_1d_inclusive(%in: tensor<128xi32>, %out: tensor<128xi32>, %acc:
 // CHECK:             %[[OUT_CURRENT_ELEMENT:.*]] = arith.addi %[[OUT_PREV_ELEMENT]], %[[IN_ELEMENT]] : i32
 // CHECK:             tm_tensor.yield %[[OUT_CURRENT_ELEMENT]] : i32
 // CHECK:           }
-// CHECK:           %[[OUT_TENSOR_NEW:.*]] = bufferization.to_tensor %[[OUT_MEMREF_NEW]] : memref<128xi32>
-// CHECK:           %[[ACC_TENSOR_NEW:.*]] = bufferization.to_tensor %[[ACC_MEMREF_NEW]] : memref<i32>
+// CHECK-DAG:           %[[OUT_TENSOR_NEW:.*]] = bufferization.to_tensor %[[OUT_MEMREF_NEW]] : memref<128xi32>
+// CHECK-DAG:           %[[ACC_TENSOR_NEW:.*]] = bufferization.to_tensor %[[ACC_MEMREF_NEW]] : memref<i32>
 // CHECK:           return %[[OUT_TENSOR_NEW]], %[[ACC_TENSOR_NEW]] : tensor<128xi32>, tensor<i32>
 func.func @scan_1d_exclusive(%in: tensor<128xi32>, %out: tensor<128xi32>, %acc: tensor<i32>) -> (tensor<128xi32>, tensor<i32>) {
   %ret_out, %ret_acc = tm_tensor.scan dimension(0) inclusive(false)