diff --git a/src/jit/impl/mlir/executable_cuda.cpp b/src/jit/impl/mlir/executable_cuda.cpp
index 638ec697ae2832ac2f25fb9e0f3beee12aa454ae..4f7c8629e96fb37a664273c4e01d07c4adc5e285 100644
--- a/src/jit/impl/mlir/executable_cuda.cpp
+++ b/src/jit/impl/mlir/executable_cuda.cpp
@@ -36,13 +36,14 @@ template <int out_dim, typename ctype>
 void setup_and_launch(const JITExecutor* fusion_opr, CUfunction func,
                       int block_size) {
     auto&& args = fusion_opr->args();
-    std::vector<StridedMemRefType<ctype, out_dim>> param_holders;
+    size_t num_memrefs = args.inputs.size() + args.outputs.size();
+    std::vector<StridedMemRefType<ctype, out_dim>> param_holders(num_memrefs);
     std::vector<void*> params;
 
     auto set_params = [&param_holders, &params](
-                              void* ptr, const megdnn::TensorLayout& layout) {
-        param_holders.push_back(StridedMemRefType<ctype, out_dim>{});
-        StridedMemRefType<ctype, out_dim>& desc = param_holders.back();
+                              size_t idx, void* ptr,
+                              const megdnn::TensorLayout& layout) {
+        auto& desc = param_holders[idx];
         desc.basePtr = static_cast<ctype*>(ptr);
         params.push_back(&(desc.basePtr));
         desc.data = static_cast<ctype*>(ptr);
@@ -56,9 +57,12 @@ void setup_and_launch(const JITExecutor* fusion_opr, CUfunction func,
             params.push_back(&(desc.strides[i]));
         }
     };
+
+    size_t idx = 0;
     for (const auto& arg : args.inputs) {
-        set_params(arg.from->dev_tensor().raw_ptr(), arg.from->layout());
+        set_params(idx++, arg.from->dev_tensor().raw_ptr(), arg.from->layout());
     }
+
     int64_t nr_elements = 0;
     for (const auto& arg : args.outputs) {
         if (nr_elements == 0) {
@@ -73,8 +77,13 @@ void setup_and_launch(const JITExecutor* fusion_opr, CUfunction func,
                        arg.from->layout().to_string().c_str());
         }
 
-        set_params(arg.from->dev_tensor().raw_ptr(), arg.from->layout());
+        set_params(idx++, arg.from->dev_tensor().raw_ptr(), arg.from->layout());
     }
+
+    mgb_assert(param_holders.size() == num_memrefs,
+               "calling push_back method of param_holders is unsafe as it "
+               "might cause reallocation of std::vector");
+
     const CompNodeEnv& env =
             CompNodeEnv::from_comp_node(fusion_opr->comp_node());