From cf8d5b9b76339fa79b2e4def914b560eb9249caa Mon Sep 17 00:00:00 2001
From: Jiaming Yuan <jm.yuan@outlook.com>
Date: Sun, 17 Mar 2019 16:59:15 +0800
Subject: [PATCH] Mark CUDA 10.1 as unsupported. (#4265)

---
 doc/build.rst                 | 7 ++++---
 src/common/device_helpers.cuh | 9 +++++----
 2 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/doc/build.rst b/doc/build.rst
index f5b561616..139e2c0fd 100644
--- a/doc/build.rst
+++ b/doc/build.rst
@@ -28,7 +28,7 @@ This page gives instructions on how to build and install XGBoost from scratch on
 .. note:: Use of Git submodules
 
   XGBoost uses Git submodules to manage dependencies. So when you clone the repo, remember to specify ``--recursive`` option:
-  
+
   .. code-block:: bash
 
     git clone --recursive https://github.com/dmlc/xgboost
@@ -185,7 +185,9 @@ Building with GPU support
 =========================
 XGBoost can be built with GPU support for both Linux and Windows using CMake. GPU support works with the Python package as well as the CLI version. See `Installing R package with GPU support`_ for special instructions for R.
 
-An up-to-date version of the CUDA toolkit is required.
+An up-to-date version of the CUDA toolkit is required.  Please note that we
+skipped the support for compiling XGBoost with NVCC 10.1 due a small bug in its
+spliter, see `#4264 <https://github.com/dmlc/xgboost/issues/4264>`_.
 
 From the command line on Linux starting from the XGBoost directory:
 
@@ -448,4 +450,3 @@ Trouble Shooting
    .. code-block:: bash
 
      git clone https://github.com/dmlc/xgboost --recursive
-
diff --git a/src/common/device_helpers.cuh b/src/common/device_helpers.cuh
index 5e2461ef7..a26342411 100644
--- a/src/common/device_helpers.cuh
+++ b/src/common/device_helpers.cuh
@@ -26,8 +26,9 @@
 #include "../common/io.h"
 #endif
 
-// Uncomment to enable
-#define TIMERS
+#if __CUDACC_VER_MAJOR__ == 10 && __CUDACC_VER_MINOR__ == 1
+#error "CUDA 10.1 is not supported, see #4264."
+#endif
 
 namespace dh {
 
@@ -893,14 +894,14 @@ class AllReducer {
     int nccl_nranks = std::accumulate(device_counts.begin(),
                         device_counts.end(), 0);
     nccl_rank += nccl_rank_offset;
-    
+
     GroupStart();
     for (size_t i = 0; i < device_ordinals.size(); i++) {
       int dev = device_ordinals.at(i);
       dh::safe_cuda(cudaSetDevice(dev));
       dh::safe_nccl(ncclCommInitRank(
         &comms.at(i),
-        nccl_nranks, id, 
+        nccl_nranks, id,
         nccl_rank));
 
       nccl_rank++;