From 7e8a04d9857c9ed3685d7d933bf2f5a6981b3bb5 Mon Sep 17 00:00:00 2001 From: Trung Nguyen Date: Thu, 12 Dec 2019 09:57:49 -0600 Subject: [PATCH] More minor update to README --- lib/gpu/README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/gpu/README b/lib/gpu/README index 92be2d408c..2d98749a40 100644 --- a/lib/gpu/README +++ b/lib/gpu/README @@ -130,7 +130,7 @@ on the head node of a GPU cluster, this library may not be installed, so you may need to copy it over from one of the compute nodes (best into this directory). Recent CUDA toolkits starting from CUDA 9 provide a dummy libcuda.so library (typically under $(CUDA_HOME)/lib64/stubs), that can be used for -linking (but not for running). +linking. The gpu library supports 3 precision modes as determined by the CUDA_PRECISION variable: