From 1a7ac522f8cc88764ee418d2385a65baaf1a6692 Mon Sep 17 00:00:00 2001 From: akerr Date: Thu, 20 Sep 2018 11:04:03 -0700 Subject: [PATCH] Clarification to README --- README.md | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index c53a42f4bcce4..c5b796eb23a90 100644 --- a/README.md +++ b/README.md @@ -56,7 +56,10 @@ when compiled with CUDA 10.0. # Compatibility -CUTLASS requires CUDA 9 but performs best with [CUDA 10.0 Toolkit](ttps://developer.nvidia.com/cuda-toolkit) or later. +CUTLASS performs best when compiled with the [CUDA 10.0 Toolkit](ttps://developer.nvidia.com/cuda-toolkit). +It is compatible with CUDA 9.0, 9.1, and 9.2, but these versions of the CUDA Toolkit do not support new Turing WMMA features. + +We have tested the following environments. |**Operating System** | **Compiler** | |-----------------|----------| @@ -64,7 +67,7 @@ CUTLASS requires CUDA 9 but performs best with [CUDA 10.0 Toolkit](ttps://develo | | Microsoft Visual Studio 2017| | Ubuntu 14.04 | GCC 4.8.2 | | Ubuntu 16.04 | GCC 5.4.0 | - +| Ubuntu 18.04 | GCC 7.3.0 | CUTLASS runs successfully on the following NVIDIA GPUs, and it is expected to be efficient on any Maxwell-, Pascal-, or Volta-architecture NVIDIA GPU.