Repository: systemml
Updated Branches:
  refs/heads/master f69047ea4 -> 5da8132ea


[SYSTEMML-445] Guard JCudaKernels with sysml.gpu.sync.postProcess flag

Project: http://git-wip-us.apache.org/repos/asf/systemml/repo
Commit: http://git-wip-us.apache.org/repos/asf/systemml/commit/5da8132e
Tree: http://git-wip-us.apache.org/repos/asf/systemml/tree/5da8132e
Diff: http://git-wip-us.apache.org/repos/asf/systemml/diff/5da8132e

Branch: refs/heads/master
Commit: 5da8132ea8c165d6d1a65c2c293fc87aaa15a2a3
Parents: f69047e
Author: Niketan Pansare <[email protected]>
Authored: Wed Jan 31 21:38:29 2018 -0800
Committer: Niketan Pansare <[email protected]>
Committed: Wed Jan 31 21:38:29 2018 -0800

----------------------------------------------------------------------
 .../sysml/runtime/instructions/gpu/context/JCudaKernels.java     | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/systemml/blob/5da8132e/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/JCudaKernels.java
----------------------------------------------------------------------
diff --git 
a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/JCudaKernels.java
 
b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/JCudaKernels.java
index d22110d..181b08d 100644
--- 
a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/JCudaKernels.java
+++ 
b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/JCudaKernels.java
@@ -27,6 +27,7 @@ import java.io.IOException;
 import java.io.InputStream;
 import java.util.HashMap;
 
+import org.apache.sysml.api.DMLScript;
 import org.apache.sysml.runtime.DMLRuntimeException;
 import org.apache.sysml.runtime.io.IOUtilFunctions;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
@@ -110,7 +111,8 @@ public class JCudaKernels {
                checkResult(cuLaunchKernel(function, config.gridDimX, 
config.gridDimY, config.gridDimZ, config.blockDimX,
                                config.blockDimY, config.blockDimZ, 
config.sharedMemBytes, config.stream, Pointer.to(kernelParams),
                                null));
-               JCuda.cudaDeviceSynchronize();
+               if(DMLScript.SYNCHRONIZE_GPU)
+                       JCuda.cudaDeviceSynchronize();
        }
 
        public static void checkResult(int cuResult) throws DMLRuntimeException 
{

Reply via email to