From 53fedbc7bdf4dea35b76b0f935f7eb7cd71e7d53 Mon Sep 17 00:00:00 2001
From: krishung5 <krish@nvidia.com>
Date: Tue, 24 Oct 2023 12:55:23 -0700
Subject: [PATCH] Fix CPU only build

---
 src/infer_response.cc | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/infer_response.cc b/src/infer_response.cc
index 8e3e2712..09737b26 100644
--- a/src/infer_response.cc
+++ b/src/infer_response.cc
@@ -249,9 +249,11 @@ InferResponse::Send(
   }
 
   bool cuda_copy = false;
+#ifdef TRITON_ENABLE_GPU
   // This variable is used to avoid printing the same message multiple times
   // when the output tensor is failed to be allocated from the CUDA memory pool.
   bool log_warning = true;
+#endif  // TRITON_ENABLE_GPU
 
   for (auto& output_tensor : OutputTensors()) {
     // FIXME: for decoupled models we will skip the requested output names.