Merge branch 'master' of github.com:UoB-HPC/GPU-STREAM

README.md

@@ -13,6 +13,41 @@ Build the OpenCL and CUDA binaries with `make`
 
 Run the OpenCL version with `./gpu-stream-ocl` and the CUDA version with `./gpu-stream-cuda`
 
+Android
+-------
+
+Assuming you have a recent Android NDK available, you can use the
+toolchain that it provides to build GPU-STREAM. You should first
+use the NDK to generate a standalone toolchain:
+
+    # Select a directory to install the toolchain to
+    ANDROID_NATIVE_TOOLCHAIN=/path/to/toolchain
+
+    ${NDK}/build/tools/make-standalone-toolchain.sh \
+      --platform=android-14 \
+      --toolchain=arm-linux-androideabi-4.8 \
+      --install-dir=${ANDROID_NATIVE_TOOLCHAIN}
+
+Make sure that the OpenCL headers and library (libOpenCL.so) are
+available in `${ANDROID_NATIVE_TOOLCHAIN}/sysroot/usr/`.
+
+You should then be able to build GPU-STREAM:
+
+    make CXX=${ANDROID_NATIVE_TOOLCHAIN}/bin/arm-linux-androideabi-g++
+
+Copy the executable and OpenCL kernels to the device:
+
+    adb push gpu-stream-ocl /data/local/tmp
+    adb push ocl-stream-kernels.cl /data/local/tmp
+
+Run GPU-STREAM from an adb shell:
+
+    adb shell
+    cd /data/local/tmp
+
+    # Use float if device doesn't support double, and reduce array size
+    ./gpu-stream-ocl --float -n 6 -s 10000000
+
 Results
 -------
 

common.cpp

@@ -38,18 +38,18 @@
 
 // Default array size 50 * 2^20 (50*8 Mebibytes double precision)
 // Use binary powers of two so divides 1024
-int ARRAY_SIZE = 52428800;
+unsigned int ARRAY_SIZE = 52428800;
 
-int NTIMES = 10;
+unsigned int NTIMES = 10;
 
 bool useFloat = false;
 
-int deviceIndex = 0;
+unsigned int deviceIndex = 0;
 
-int parseInt(const char *str, int *output)
+int parseUInt(const char *str, unsigned int *output)
 {
     char *next;
-    *output = strtol(str, &next, 10);
+    *output = strtoul(str, &next, 10);
     return !strlen(next);
 }
 
@@ -64,7 +64,7 @@ void parseArguments(int argc, char *argv[])
         }
         else if (!strcmp(argv[i], "--device"))
         {
-            if (++i >= argc || !parseInt(argv[i], &deviceIndex))
+            if (++i >= argc || !parseUInt(argv[i], &deviceIndex))
             {
                 std::cout << "Invalid device index" << std::endl;
                 exit(1);
@@ -72,7 +72,7 @@ void parseArguments(int argc, char *argv[])
         }
         else if (!strcmp(argv[i], "--arraysize") || !strcmp(argv[i], "-s"))
         {
-            if (++i >= argc || !parseInt(argv[i], &ARRAY_SIZE))
+            if (++i >= argc || !parseUInt(argv[i], &ARRAY_SIZE))
             {
                 std::cout << "Invalid array size" << std::endl;
                 exit(1);
@@ -80,7 +80,7 @@ void parseArguments(int argc, char *argv[])
         }
         else if (!strcmp(argv[i], "--numtimes") || !strcmp(argv[i], "-n"))
         {
-            if (++i >= argc || !parseInt(argv[i], &NTIMES))
+            if (++i >= argc || !parseUInt(argv[i], &NTIMES))
             {
                 std::cout << "Invalid number of times" << std::endl;
                 exit(1);

common.h

@@ -36,6 +36,7 @@
 
 #include <iomanip>
 #include <iostream>
+#include <cstdlib>
 #include <cstring>
 #include <limits>
 #include <stdexcept>
@@ -46,12 +47,12 @@ extern void parseArguments(int argc, char *argv[]);
 
 extern void listDevices(void);
 
-extern int ARRAY_SIZE;
-extern int NTIMES;
+extern unsigned int ARRAY_SIZE;
+extern unsigned int NTIMES;
 
 extern bool useFloat;
 
-extern int deviceIndex;
+extern unsigned int deviceIndex;
 
 
 template < typename T >