diff --git a/xla/hlo/utils/BUILD b/xla/hlo/utils/BUILD
index b4900fc092e4b..c3e47f05dca03 100644
--- a/xla/hlo/utils/BUILD
+++ b/xla/hlo/utils/BUILD
@@ -119,6 +119,7 @@ cc_library(
         "//xla/hlo/ir:tile_assignment",
         "//xla/service:call_graph",
         "//xla/service:dot_as_convolution_util",
+        "//xla/service:gather_scatter_utils",
         "@com_google_absl//absl/algorithm:container",
         "@com_google_absl//absl/container:btree",
         "@com_google_absl//absl/container:flat_hash_map",
diff --git a/xla/hlo/utils/hlo_sharding_util.cc b/xla/hlo/utils/hlo_sharding_util.cc
index def884e17ef0c..1c60203f48d45 100644
--- a/xla/hlo/utils/hlo_sharding_util.cc
+++ b/xla/hlo/utils/hlo_sharding_util.cc
@@ -54,6 +54,7 @@ limitations under the License.
 #include "xla/protobuf_util.h"
 #include "xla/service/call_graph.h"
 #include "xla/service/dot_as_convolution_util.h"
+#include "xla/service/gather_scatter_utils.h"
 #include "xla/shape.h"
 #include "xla/shape_util.h"
 #include "xla/util.h"
@@ -1513,8 +1514,8 @@ absl::InlinedVector<int64_t, 1> GetGatherScatterOperandPassthroughOperandDims(
   absl::InlinedVector<int64_t, 1> passthrough_dims;
   int64_t collapsed_or_batching = 0;
   for (int64_t i = 0; i < operand_shape.rank(); ++i) {
-    if (absl::c_linear_search(collapsed_or_inserted_dims, i) ||
-        absl::c_linear_search(operand_batching_dims, i)) {
+    if (IsCollapsedOrBatchingDim(collapsed_or_inserted_dims,
+                                 operand_batching_dims, i)) {
       collapsed_or_batching++;
       continue;
     }
@@ -1546,8 +1547,8 @@ GetGatherScatterOperandPassthroughOutputOrUpdateDims(
   absl::InlinedVector<int64_t, 1> passthrough_dims;
   int64_t collapsed_or_batching = 0;
   for (int64_t i = 0; i < operand_shape.rank(); ++i) {
-    if (absl::c_linear_search(collapsed_or_inserted_dims, i) ||
-        absl::c_linear_search(operand_batching_dims, i)) {
+    if (IsCollapsedOrBatchingDim(collapsed_or_inserted_dims,
+                                 operand_batching_dims, i)) {
       collapsed_or_batching++;
       continue;
     }
@@ -1581,8 +1582,8 @@ std::optional<HloSharding> PassthroughOperandToGatherOutputOrScatterUpdate(
   DimensionVector passthrough_tile(output_or_update_rank, 1);
   int64_t collapsed_or_batching = 0;
   for (int64_t i = 0; i < operand_shape.rank(); ++i) {
-    if (absl::c_linear_search(collapsed_or_inserted_dims, i) ||
-        absl::c_linear_search(operand_batching_dims, i)) {
+    if (IsCollapsedOrBatchingDim(collapsed_or_inserted_dims,
+                                 operand_batching_dims, i)) {
       collapsed_or_batching++;
       continue;
     }
@@ -1636,8 +1637,8 @@ std::optional<HloSharding> PassthroughGatherOutputOrScatterUpdateToOperand(
   // Relevant dims have shardings passed to the operand.
   DimensionVector relevant_output_or_update_dims;
   for (int64_t i = 0; i < operand_shape.rank(); ++i) {
-    if (absl::c_linear_search(collapsed_or_inserted_dims, i) ||
-        absl::c_linear_search(operand_batching_dims, i)) {
+    if (IsCollapsedOrBatchingDim(collapsed_or_inserted_dims,
+                                 operand_batching_dims, i)) {
       collapsed_or_batching++;
       continue;
     }
@@ -1770,8 +1771,8 @@ std::vector<int64_t> GetScatterSliceSize(const Shape& operand_shape,
   std::vector<int64_t> slice_size(operand_shape.rank(), 1);
   int64_t num_update_window_dims = 0;
   for (int64_t i = 0; i < operand_shape.rank(); ++i) {
-    if (absl::c_linear_search(dnums.inserted_window_dims(), i) ||
-        absl::c_linear_search(dnums.input_batching_dims(), i)) {
+    if (IsCollapsedOrBatchingDim(dnums.inserted_window_dims(),
+                                 dnums.input_batching_dims(), i)) {
       continue;
     }
     slice_size[i] = update_shape.dimensions(
diff --git a/xla/service/gather_expander.cc b/xla/service/gather_expander.cc
index 095cb318b2833..d0190d02d7c22 100644
--- a/xla/service/gather_expander.cc
+++ b/xla/service/gather_expander.cc
@@ -222,8 +222,8 @@ HloInstruction* CreateGatherLoopAccumulatorInitValue(
   accumulator_state_shape_dims.reserve(1 + slice_sizes.size());
   accumulator_state_shape_dims.push_back(gather_loop_trip_count);
   for (int64_t i = 0; i < slice_sizes.size(); i++) {
-    if (!absl::c_linear_search(dim_numbers.collapsed_slice_dims(), i) &&
-        !absl::c_linear_search(dim_numbers.operand_batching_dims(), i)) {
+    if (!IsCollapsedOrBatchingDim(dim_numbers.collapsed_slice_dims(),
+                                  dim_numbers.operand_batching_dims(), i)) {
       accumulator_state_shape_dims.push_back(slice_sizes[i]);
     }
   }
diff --git a/xla/service/gather_scatter_utils.cc b/xla/service/gather_scatter_utils.cc
index 1b75952475466..98473a0264ecf 100644
--- a/xla/service/gather_scatter_utils.cc
+++ b/xla/service/gather_scatter_utils.cc
@@ -214,4 +214,10 @@ absl::StatusOr<HloInstruction*> ExpandIndexVectorIntoOperandSpace(
   return MakeConcatHlo(expanded_index_components, /*dimension=*/0);
 }
 
+bool IsCollapsedOrBatchingDim(absl::Span<const int64_t> collapsed_dims,
+                              absl::Span<const int64_t> batching_dims,
+                              int64_t dim) {
+  return absl::c_linear_search(collapsed_dims, dim) ||
+         absl::c_linear_search(batching_dims, dim);
+}
 }  // namespace xla
diff --git a/xla/service/gather_scatter_utils.h b/xla/service/gather_scatter_utils.h
index cf1368373583b..e73127003261d 100644
--- a/xla/service/gather_scatter_utils.h
+++ b/xla/service/gather_scatter_utils.h
@@ -62,6 +62,10 @@ absl::StatusOr<HloInstruction*> ExpandIndexVectorIntoOperandSpace(
     absl::Span<const int64_t> operand_batching_dims,
     HloInstruction* index_vector, HloInstruction* induction_var);
 
+// Returns true if the given dimension is a collapsed or batching dimension.
+bool IsCollapsedOrBatchingDim(absl::Span<const int64_t> collapsed_dims,
+                              absl::Span<const int64_t> batching_dims,
+                              int64_t dim);
 }  // namespace xla
 
 #endif  // XLA_SERVICE_GATHER_SCATTER_UTILS_H_