From ad8a61545ac8318e9ad23df716d00a90b3edd6a3 Mon Sep 17 00:00:00 2001
From: Julia Jiang <julia.jiang@amd.com>
Date: Mon, 21 Sep 2020 11:15:40 -0400
Subject: [PATCH] SWDEV-253250-update instruction for hip sample

Change-Id: Iebc2887a3687e6c565999f4faf1cd4cf893e24cd
---
 samples/0_Intro/square/README.md | 42 ++++++++++++++++++++++++++------
 1 file changed, 34 insertions(+), 8 deletions(-)

diff --git a/samples/0_Intro/square/README.md b/samples/0_Intro/square/README.md
index c185903993..0bbb2f7e39 100644
--- a/samples/0_Intro/square/README.md
+++ b/samples/0_Intro/square/README.md
@@ -1,13 +1,39 @@
 # Square.md
 
-Simple test which shows how to use hipify-perl to port CUDA code to HIP.  
-See related [blog](http://gpuopen.com/hip-to-be-squared-an-introductory-hip-tutorial) that explains the example. 
+Simple test which shows how to use hipify-perl to port CUDA code to HIP.
+See related [blog](http://gpuopen.com/hip-to-be-squared-an-introductory-hip-tutorial) that explains the example.
 Now it is even simpler and requires no manual modification to the hipified source code - just hipify and compile:
 
-1. Add hip/bin path to the PATH  :
-    <code>export PATH=$PATH:[MYHIP]/bin</code>
+- Add hip/bin path to the PATH
 
-2. <code>$ make </code>
-   Make runs these steps.  This can be performed on either CUDA or AMD platform:
-   <code>hipify-perl square.cu > square.cpp </code>    # convert cuda code to hip code
-   <code>hipcc square.cpp</code>                       # compile into executable
+```
+$ export PATH=$PATH:[MYHIP]/bin
+```
+
+- Define environment variable
+
+```
+$ export HIP_PATH=[MYHIP]
+```
+
+- Build executible file
+
+```
+$ cd ~/hip/samples/0_Intro/square
+$ make
+/home/user/hip/bin/hipify-perl square.cu > square.cpp
+/home/user/hip/bin/hipcc  square.cpp -o square.out
+/home/user/hip/bin/hipcc -use-staticlib  square.cpp -o square.out.static
+```
+- Execute file
+```
+$ ./square.out
+info: running on device Navi 14 [Radeon Pro W5500]
+info: allocate host mem (  7.63 MB)
+info: allocate device mem (  7.63 MB)
+info: copy Host2Device
+info: launch 'vector_square' kernel
+info: copy Device2Host
+info: check result
+PASSED!
+```