diff --git a/configure.py b/configure.py
index b4d7778..8d6c6e5 100644
--- a/configure.py
+++ b/configure.py
@@ -1324,9 +1324,9 @@
 
   cuda_libraries = ['cuda', 'cudnn']
   if is_linux():
-    if 'TF_TENSORRT_VERSION' in environ_cp:  # if env variable exists
+    if environ_cp.get('TF_NEED_TENSORRT', None):
       cuda_libraries.append('tensorrt')
-    if environ_cp.get('TF_NCCL_VERSION', None):  # if env variable not empty
+    if environ_cp.get('TF_NCCL_VERSION', None):
       cuda_libraries.append('nccl')
 
   proc = subprocess.Popen(
@@ -1453,8 +1453,12 @@
         cuda_env_names = [
             'TF_CUDA_VERSION', 'TF_CUBLAS_VERSION', 'TF_CUDNN_VERSION',
             'TF_TENSORRT_VERSION', 'TF_NCCL_VERSION', 'TF_CUDA_PATHS',
-            'CUDA_TOOLKIT_PATH'
+            # Items below are for backwards compatibility when not using
+            # TF_CUDA_PATHS.
+            'CUDA_TOOLKIT_PATH', 'CUDNN_INSTALL_PATH', 'NCCL_INSTALL_PATH',
+            'NCCL_HDR_PATH', 'TENSORRT_INSTALL_PATH'
         ]
+        # Note: set_action_env_var above already writes to bazelrc.
         for name in cuda_env_names:
           if name in environ_cp:
             write_action_env_to_bazelrc(name, environ_cp[name])
diff --git a/third_party/gpus/cuda_configure.bzl b/third_party/gpus/cuda_configure.bzl
index 607210b..380e642 100644
--- a/third_party/gpus/cuda_configure.bzl
+++ b/third_party/gpus/cuda_configure.bzl
@@ -335,10 +335,8 @@
     return "\n".join(inc_entries)
 
 def enable_cuda(repository_ctx):
-    if "TF_NEED_CUDA" in repository_ctx.os.environ:
-        enable_cuda = repository_ctx.os.environ["TF_NEED_CUDA"].strip()
-        return enable_cuda == "1"
-    return False
+    """Returns whether to build with CUDA support."""
+    return int(repository_ctx.os.environ.get("TF_NEED_CUDA", False))
 
 def matches_version(environ_version, detected_version):
     """Checks whether the user-specified version matches the detected version.
diff --git a/third_party/tensorrt/tensorrt_configure.bzl b/third_party/tensorrt/tensorrt_configure.bzl
index 5c2dded..728c318 100644
--- a/third_party/tensorrt/tensorrt_configure.bzl
+++ b/third_party/tensorrt/tensorrt_configure.bzl
@@ -18,6 +18,7 @@
 _TENSORRT_INSTALL_PATH = "TENSORRT_INSTALL_PATH"
 _TF_TENSORRT_CONFIG_REPO = "TF_TENSORRT_CONFIG_REPO"
 _TF_TENSORRT_VERSION = "TF_TENSORRT_VERSION"
+_TF_NEED_TENSORRT = "TF_NEED_TENSORRT"
 
 _TF_TENSORRT_LIBS = ["nvinfer", "nvinfer_plugin"]
 _TF_TENSORRT_HEADERS = ["NvInfer.h", "NvUtils.h", "NvInferPlugin.h"]
@@ -43,6 +44,10 @@
         "%{tensorrt_libs}": "[]",
     })
 
+def enable_tensorrt(repository_ctx):
+    """Returns whether to build with TensorRT support."""
+    return int(repository_ctx.os.environ.get(_TF_NEED_TENSORRT, False))
+
 def _tensorrt_configure_impl(repository_ctx):
     """Implementation of the tensorrt_configure repository rule."""
     if _TF_TENSORRT_CONFIG_REPO in repository_ctx.os.environ:
@@ -56,7 +61,7 @@
         )
         return
 
-    if _TF_TENSORRT_VERSION not in repository_ctx.os.environ:
+    if not enable_tensorrt(repository_ctx):
         _create_dummy_repository(repository_ctx)
         return
 
@@ -99,6 +104,7 @@
         _TENSORRT_INSTALL_PATH,
         _TF_TENSORRT_VERSION,
         _TF_TENSORRT_CONFIG_REPO,
+        _TF_NEED_TENSORRT,
         "TF_CUDA_PATHS",
     ],
 )