Modify the implementation of BlockXReduce to fit more scenes (#39554)

* Modify the implementation of BlockYReduce to fit more scenes

* fix

* fix

paddle/phi/kernels/primitive/compute_primitives.h

@@ -110,7 +110,11 @@ __device__ __forceinline__ T BlockXReduce(T val, ReduceOp reducer) {
     T temp = paddle::platform::CudaShuffleDownSync(mask, val, stride);
     val = reducer(val, temp);
   }
-  return val;
+  if (threadIdx.x == 0) {
+    shared[threadIdx.y] = val;
+  }
+  __syncthreads();
+  return shared[threadIdx.y];
 }
 
 /**