Merge pull request #344 from ggruszczynski/develop

sm_75

Author: llaniewski

src/Makefile.in.Rt

@@ -25,7 +25,7 @@ LIBS := @R_LIB@ @LIBS@ @RCPP_LD@ -Xlinker -rpath -Xlinker $(MPI_LIBS) -L$(MPI_LI
 
 TARGETS := CLB.so
 
-NVCC := @CUDA_HOME@/bin/nvcc -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20  -gencode arch=compute_60,code=sm_60
+NVCC := @CUDA_HOME@/bin/nvcc -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20  -gencode arch=compute_60,code=sm_60, -gencode arch=compute_75,code=sm_75
 
 all: $(TARGETS) 
 

src/configure.ac

@@ -51,7 +51,7 @@ AC_ARG_ENABLE([cpp11],
 
 AC_ARG_WITH([cuda-arch],
 	AC_HELP_STRING([--with-cuda-arch=arch],
-		[specify the desired CUDA architecture (sm_11/sm_13/sm_20/sm_30/sm_60)]))
+		[specify the desired CUDA architecture (sm_11/sm_13/sm_20/sm_30/sm_60/sm_75)]))
 
 AC_ARG_WITH([nlopt],
 	AC_HELP_STRING([--with-nlopt=nlopt],
@@ -285,7 +285,14 @@ then
 		AC_MSG_RESULT([${CUDA_ARCH}])
 		AC_DEFINE([GRID3D], [1], [Using 3D block grid in CUDA])
 
-    CUDA_SUFF="60"
+    	CUDA_SUFF="60"
+	
+	elif test "$CUDA_ARCH" == "sm_75"
+	then
+		AC_MSG_RESULT([${CUDA_ARCH}])
+		AC_DEFINE([GRID3D], [1], [Using 3D block grid in CUDA])
+
+    	CUDA_SUFF="75"
 	else
 		AC_MSG_ERROR([unknown arch ${CUDA_ARCH}])
 	fi