fix(dnn/cuda): add block size limit for culass gemm algo

GitOrigin-RevId: c0940e4535

dnn/src/cuda/matrix_mul/cutlass_float32_simt.cpp

@@ -22,10 +22,20 @@ using namespace cutlass_wrapper;
 
 bool MatrixMulForwardImpl::AlgoFloat32SIMT::is_available(
         const SizeArgs& args) const {
-    return args.opr->param().format == param::MatrixMul::Format::DEFAULT &&
-           args.layout_a.dtype == dtype::Float32() &&
-           args.layout_b.dtype == dtype::Float32() &&
-           args.layout_c.dtype == dtype::Float32();
+    bool available =
+            args.opr->param().format == param::MatrixMul::Format::DEFAULT &&
+            args.layout_a.dtype == dtype::Float32() &&
+            args.layout_b.dtype == dtype::Float32() &&
+            args.layout_c.dtype == dtype::Float32();
+    int n = args.layout_c.shape[1];
+    auto&& device_prop = cuda::current_device_prop();
+    int y_grid_limit = device_prop.maxGridSize[1];
+    // limit y grid
+    available &= ((n + m_algo_param.threadblock_n - 1) /
+                          m_algo_param.threadblock_n <=
+                  y_grid_limit);
+
+    return available;
 }
 
 size_t MatrixMulForwardImpl::AlgoFloat32SIMT::get_workspace_in_bytes(