You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@systemml.apache.org by mb...@apache.org on 2017/04/22 23:15:41 UTC

incubator-systemml git commit: [MINOR] Fix gpu backend compiler warnings and javadoc issues

Repository: incubator-systemml
Updated Branches:
  refs/heads/master dd373ee01 -> 56e6b6e94


[MINOR] Fix gpu backend compiler warnings and javadoc issues

Project: http://git-wip-us.apache.org/repos/asf/incubator-systemml/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-systemml/commit/56e6b6e9
Tree: http://git-wip-us.apache.org/repos/asf/incubator-systemml/tree/56e6b6e9
Diff: http://git-wip-us.apache.org/repos/asf/incubator-systemml/diff/56e6b6e9

Branch: refs/heads/master
Commit: 56e6b6e94f6f98ab5152b2eb633f7d93a6cab1e6
Parents: dd373ee
Author: Matthias Boehm <mb...@gmail.com>
Authored: Sat Apr 22 16:17:38 2017 -0700
Committer: Matthias Boehm <mb...@gmail.com>
Committed: Sat Apr 22 16:17:38 2017 -0700

----------------------------------------------------------------------
 .../java/org/apache/sysml/api/DMLScript.java    |  4 +--
 .../gpu/AggregateUnaryGPUInstruction.java       |  1 -
 .../gpu/ConvolutionGPUInstruction.java          |  1 -
 .../instructions/gpu/MMTSJGPUInstruction.java   |  1 -
 .../gpu/MatrixBuiltinGPUInstruction.java        |  1 -
 .../MatrixMatrixArithmeticGPUInstruction.java   |  1 -
 .../gpu/MatrixMatrixAxpyGPUInstruction.java     |  1 -
 .../instructions/gpu/ReorgGPUInstruction.java   |  1 -
 .../instructions/gpu/context/CSRPointer.java    | 23 ++++++++------
 .../instructions/gpu/context/GPUContext.java    |  8 +++--
 .../gpu/context/GPUContextPool.java             |  2 ++
 .../instructions/gpu/context/GPUObject.java     | 33 +++++++++++++-------
 .../instructions/gpu/context/JCudaKernels.java  |  4 +--
 .../runtime/matrix/data/LibMatrixCUDA.java      |  3 +-
 .../runtime/matrix/data/LibMatrixReorg.java     |  1 +
 .../runtime/matrix/data/SparseBlockCSR.java     |  2 +-
 .../test/integration/AutomatedTestBase.java     |  4 ---
 17 files changed, 51 insertions(+), 40 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/api/DMLScript.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/api/DMLScript.java b/src/main/java/org/apache/sysml/api/DMLScript.java
index febbf13..ab059a0 100644
--- a/src/main/java/org/apache/sysml/api/DMLScript.java
+++ b/src/main/java/org/apache/sysml/api/DMLScript.java
@@ -845,11 +845,11 @@ public class DMLScript
 			ec = ExecutionContextFactory.createContext(rtprog);
 			if (DMLScript.USE_ACCELERATOR && ec != null){
 				gCtx = GPUContextPool.getFromPool();
-				gCtx.initializeThread();
-				ec.setGPUContext(gCtx);
 				if (gCtx == null) {
 					throw new DMLRuntimeException("GPU : Could not create GPUContext, either no GPU or all GPUs currently in use");
 				}
+				gCtx.initializeThread();
+				ec.setGPUContext(gCtx);
 			}
 			rtprog.execute( ec );  
 			

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/AggregateUnaryGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/AggregateUnaryGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/AggregateUnaryGPUInstruction.java
index bfe30f9..8bdd397 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/AggregateUnaryGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/AggregateUnaryGPUInstruction.java
@@ -27,7 +27,6 @@ import org.apache.sysml.runtime.functionobjects.ReduceCol;
 import org.apache.sysml.runtime.functionobjects.ReduceRow;
 import org.apache.sysml.runtime.instructions.InstructionUtils;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.AggregateUnaryOperator;
 import org.apache.sysml.runtime.matrix.operators.Operator;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/ConvolutionGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/ConvolutionGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/ConvolutionGPUInstruction.java
index e806f3b..497b182 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/ConvolutionGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/ConvolutionGPUInstruction.java
@@ -26,7 +26,6 @@ import org.apache.sysml.runtime.controlprogram.context.ExecutionContext;
 import org.apache.sysml.runtime.functionobjects.SwapIndex;
 import org.apache.sysml.runtime.instructions.InstructionUtils;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.ReorgOperator;
 import org.apache.sysml.runtime.util.ConvolutionUtils;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/MMTSJGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MMTSJGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MMTSJGPUInstruction.java
index af97141..c147a6f 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MMTSJGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MMTSJGPUInstruction.java
@@ -32,7 +32,6 @@ import org.apache.sysml.runtime.controlprogram.caching.MatrixObject;
 import org.apache.sysml.runtime.controlprogram.context.ExecutionContext;
 import org.apache.sysml.runtime.instructions.InstructionUtils;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.Operator;
 import org.apache.sysml.utils.GPUStatistics;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixBuiltinGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixBuiltinGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixBuiltinGPUInstruction.java
index 3fc8a98..ce25dec 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixBuiltinGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixBuiltinGPUInstruction.java
@@ -23,7 +23,6 @@ import org.apache.sysml.runtime.DMLRuntimeException;
 import org.apache.sysml.runtime.controlprogram.caching.MatrixObject;
 import org.apache.sysml.runtime.controlprogram.context.ExecutionContext;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.Operator;
 import org.apache.sysml.utils.GPUStatistics;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixArithmeticGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixArithmeticGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixArithmeticGPUInstruction.java
index a1520a9..9573a1a 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixArithmeticGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixArithmeticGPUInstruction.java
@@ -23,7 +23,6 @@ import org.apache.sysml.runtime.DMLRuntimeException;
 import org.apache.sysml.runtime.controlprogram.caching.MatrixObject;
 import org.apache.sysml.runtime.controlprogram.context.ExecutionContext;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.BinaryOperator;
 import org.apache.sysml.runtime.matrix.operators.Operator;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixAxpyGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixAxpyGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixAxpyGPUInstruction.java
index 28254c6..58905d6 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixAxpyGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/MatrixMatrixAxpyGPUInstruction.java
@@ -26,7 +26,6 @@ import org.apache.sysml.runtime.controlprogram.context.ExecutionContext;
 import org.apache.sysml.runtime.instructions.InstructionUtils;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
 import org.apache.sysml.runtime.instructions.cp.ScalarObject;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.Operator;
 import org.apache.sysml.utils.GPUStatistics;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/ReorgGPUInstruction.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/ReorgGPUInstruction.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/ReorgGPUInstruction.java
index 424a2c5..53d56a3 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/ReorgGPUInstruction.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/ReorgGPUInstruction.java
@@ -25,7 +25,6 @@ import org.apache.sysml.runtime.controlprogram.context.ExecutionContext;
 import org.apache.sysml.runtime.functionobjects.SwapIndex;
 import org.apache.sysml.runtime.instructions.InstructionUtils;
 import org.apache.sysml.runtime.instructions.cp.CPOperand;
-import org.apache.sysml.runtime.instructions.gpu.context.GPUContext;
 import org.apache.sysml.runtime.matrix.data.LibMatrixCUDA;
 import org.apache.sysml.runtime.matrix.operators.Operator;
 import org.apache.sysml.runtime.matrix.operators.ReorgOperator;

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/CSRPointer.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/CSRPointer.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/CSRPointer.java
index c25bd22..ef549a1 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/CSRPointer.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/CSRPointer.java
@@ -86,25 +86,25 @@ public class CSRPointer {
     allocateMatDescrPointer();
   }
 
-  private Pointer allocate(String instName, long size) throws DMLRuntimeException {
-    return getGPUContext().allocate(instName, size);
-  }
+//  private Pointer allocate(String instName, long size) throws DMLRuntimeException {
+//    return getGPUContext().allocate(instName, size);
+//  }
 
   private Pointer allocate(long size) throws DMLRuntimeException {
     return getGPUContext().allocate(size);
   }
 
-  private void cudaFreeHelper(Pointer toFree) throws DMLRuntimeException {
-    getGPUContext().cudaFreeHelper(toFree);
-  }
+//  private void cudaFreeHelper(Pointer toFree) throws DMLRuntimeException {
+//    getGPUContext().cudaFreeHelper(toFree);
+//  }
 
   private void cudaFreeHelper(Pointer toFree, boolean eager) throws DMLRuntimeException {
     getGPUContext().cudaFreeHelper(toFree, eager);
   }
 
-  private void cudaFreeHelper(String instName, Pointer toFree, boolean eager) throws DMLRuntimeException {
-    getGPUContext().cudaFreeHelper(instName, toFree, eager);
-  }
+//  private void cudaFreeHelper(String instName, Pointer toFree, boolean eager) throws DMLRuntimeException {
+//    getGPUContext().cudaFreeHelper(instName, toFree, eager);
+//  }
 
   private static long getDoubleSizeOf(long numElems) {
     return numElems * ((long)jcuda.Sizeof.DOUBLE);
@@ -235,6 +235,7 @@ public class CSRPointer {
    * Estimates the number of non-zero elements from the result of a sparse matrix multiplication C = A * B
    * and returns the {@link CSRPointer} to C with the appropriate GPU memory.
    *
+   * @param gCtx ?
    * @param handle a valid {@link cusparseHandle}
    * @param A      Sparse Matrix A on GPU
    * @param transA 'T' if A is to be transposed, 'N' otherwise
@@ -298,6 +299,7 @@ public class CSRPointer {
   /**
    * Factory method to allocate an empty CSR Sparse matrix on the GPU
    *
+   * @param gCtx ?
    * @param nnz2 number of non-zeroes
    * @param rows number of rows
    * @return a {@link CSRPointer} instance that encapsulates the CSR matrix on GPU
@@ -449,6 +451,8 @@ public class CSRPointer {
 
   /**
    * Calls cudaFree lazily on the allocated {@link Pointer} instances
+   * 
+   * @throws DMLRuntimeException ?
    */
   public void deallocate() throws DMLRuntimeException {
     deallocate(false);
@@ -458,6 +462,7 @@ public class CSRPointer {
    * Calls cudaFree lazily or eagerly on the allocated {@link Pointer} instances
    *
    * @param eager whether to do eager or lazy cudaFrees
+   * @throws DMLRuntimeException ?
    */
   public void deallocate(boolean eager) throws DMLRuntimeException {
     if (nnz > 0) {

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContext.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContext.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContext.java
index 708f291..d6f3a71 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContext.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContext.java
@@ -155,7 +155,6 @@ public class GPUContext {
     cudaSetDevice(deviceNum);
   }
 
-  @SuppressWarnings("unused")
   public static int cudaGetDevice() {
     int[] device = new int[1];
     JCuda.cudaGetDevice(device);
@@ -492,6 +491,7 @@ public class GPUContext {
    * Gets the device properties for the active GPU (set with cudaSetDevice())
    *
    * @return the device properties
+   * @throws DMLRuntimeException ?
    */
   public cudaDeviceProp getGPUProperties() throws DMLRuntimeException {
     return GPUContextPool.getGPUProperties(deviceNum);
@@ -501,6 +501,7 @@ public class GPUContext {
    * Gets the maximum number of threads per block for "active" GPU
    *
    * @return the maximum number of threads per block
+   * @throws DMLRuntimeException ?
    */
   public int getMaxThreadsPerBlock() throws DMLRuntimeException {
     cudaDeviceProp deviceProps = getGPUProperties();
@@ -511,6 +512,7 @@ public class GPUContext {
    * Gets the maximum number of blocks supported by the active cuda device
    *
    * @return the maximum number of blocks supported
+   * @throws DMLRuntimeException ?
    */
   public int getMaxBlocks() throws DMLRuntimeException {
     cudaDeviceProp deviceProp = getGPUProperties();
@@ -521,6 +523,7 @@ public class GPUContext {
    * Gets the shared memory per block supported by the active cuda device
    *
    * @return the shared memory per block
+   * @throws DMLRuntimeException ?
    */
   public long getMaxSharedMemory() throws DMLRuntimeException {
     cudaDeviceProp deviceProp = getGPUProperties();
@@ -531,6 +534,7 @@ public class GPUContext {
    * Gets the warp size supported by the active cuda device
    *
    * @return the warp size
+   * @throws DMLRuntimeException ?
    */
   public int getWarpSize() throws DMLRuntimeException {
     cudaDeviceProp deviceProp = getGPUProperties();
@@ -574,7 +578,7 @@ public class GPUContext {
   /**
    * Clears all memory used by this {@link GPUContext}
    * Be careful to ensure that no memory is currently being used before invoking this
-   * @throws DMLRuntimeException
+   * @throws DMLRuntimeException ?
    */
   public void clearMemory() throws DMLRuntimeException {
     while (allocatedGPUObjects.isEmpty()) {

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContextPool.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContextPool.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContextPool.java
index 6452651..1d0b5c8 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContextPool.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUContextPool.java
@@ -61,6 +61,7 @@ public class GPUContextPool {
    * Also sets behaviour for J{Cuda, Cudnn, Cublas, Cusparse} in case of error
    * Initializes the CUDA driver
    * All these need be done once, and not per GPU
+   * @throws DMLRuntimeException ?
    */
   public synchronized static void initializeGPU() throws DMLRuntimeException {
     GPUContext.LOG.info("Initializing CUDA");
@@ -109,6 +110,7 @@ public class GPUContextPool {
   /**
    * Gets an initialized GPUContext from a pool of GPUContexts, each linked to a GPU
    * @return null if not more GPUContexts in pool, a valid GPUContext otherwise
+   * @throws DMLRuntimeException ?
    */
   public static synchronized GPUContext getFromPool() throws DMLRuntimeException {
     if (!initialized) initializeGPU();

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUObject.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUObject.java b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUObject.java
index 3a1fafa..dd5ba41 100644
--- a/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUObject.java
+++ b/src/main/java/org/apache/sysml/runtime/instructions/gpu/context/GPUObject.java
@@ -96,9 +96,9 @@ public class GPUObject {
 	/** Enclosing {@link MatrixObject} instance */
 	protected MatrixObject mat = null;
 
-	private Pointer allocate(String instName, long size) throws DMLRuntimeException {
-		return getGPUContext().allocate(instName, size);
-	}
+//	private Pointer allocate(String instName, long size) throws DMLRuntimeException {
+//		return getGPUContext().allocate(instName, size);
+//	}
 
 	private Pointer allocate(long size) throws DMLRuntimeException {
 		return getGPUContext().allocate(size);
@@ -108,9 +108,9 @@ public class GPUObject {
 		getGPUContext().cudaFreeHelper(toFree);
 	}
 
-	private void cudaFreeHelper(Pointer toFree, boolean eager) throws DMLRuntimeException {
-		getGPUContext().cudaFreeHelper(toFree, eager);
-	}
+//	private void cudaFreeHelper(Pointer toFree, boolean eager) throws DMLRuntimeException {
+//		getGPUContext().cudaFreeHelper(toFree, eager);
+//	}
 
 	private void cudaFreeHelper(String instName, Pointer toFree, boolean eager) throws DMLRuntimeException {
 		getGPUContext().cudaFreeHelper(instName, toFree, eager);
@@ -200,7 +200,6 @@ public class GPUObject {
 	 * @return the debug string
 	 * @throws DMLRuntimeException  if DMLRuntimeException occurs
 	 */
-	@SuppressWarnings("unused")
 	public static String debugString(Pointer A, long rows, long cols) throws DMLRuntimeException {
 		StringBuffer sb = new StringBuffer();
 		int len = toIntExact(rows * cols);
@@ -230,6 +229,8 @@ public class GPUObject {
 	 * Make sure to call {@link #addReadLock()} after this to set appropriate state, if you are not sure what you are doing.
 	 * Needed for operations like {@link JCusparse#cusparseDcsrgemm(cusparseHandle, int, int, int, int, int, cusparseMatDescr, int, Pointer, Pointer, Pointer, cusparseMatDescr, int, Pointer, Pointer, Pointer, cusparseMatDescr, Pointer, Pointer, Pointer)}
 	 * @param sparseMatrixPtr CSR (compressed sparse row) pointer
+	 * 
+	 * @throws DMLRuntimeException ?
 	 */
 	public void setSparseMatrixCudaPointer(CSRPointer sparseMatrixPtr) throws DMLRuntimeException {
 		this.jcudaSparseMatrixPtr = sparseMatrixPtr;
@@ -245,6 +246,7 @@ public class GPUObject {
 	 * Make sure to call {@link #addReadLock()} after this to set appropriate state, if you are not sure what you are doing.
 	 *
 	 * @param densePtr dense pointer
+	 * @throws DMLRuntimeException ?
 	 */
 	public void setDenseMatrixCudaPointer(Pointer densePtr) throws DMLRuntimeException{
 		this.jcudaDenseMatrixPtr = densePtr;
@@ -543,7 +545,7 @@ public class GPUObject {
 	/**
 	 * if the data is allocated on the GPU and is dirty, it is copied back to the host memory
 	 * @return true if a copy to host happened, false otherwise
-	 * @throws CacheException
+	 * @throws CacheException ?
 	 */
 	public boolean acquireHostRead() throws CacheException {
 		boolean copied = false;
@@ -825,17 +827,26 @@ public class GPUObject {
 
 	}
 
-	/** Pointer to dense matrix */
+	/** 
+	 * Pointer to dense matrix 
+	 * @return ?
+	 */
 	public Pointer getJcudaDenseMatrixPtr() {
 		return jcudaDenseMatrixPtr;
 	}
 
-	/** Pointer to sparse matrix */
+	/** 
+	 * Pointer to sparse matrix
+	 * @return ?
+	 */
 	public CSRPointer getJcudaSparseMatrixPtr() {
 		return jcudaSparseMatrixPtr;
 	}
 
-	/** Whether this block is dirty on the GPU */
+	/** 
+	 * Whether this block is dirty on the GPU 
+	 * @return ?
+	 */
 	public boolean isDirty() {
 		return dirty;
 	}

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/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 b4cb87d..ac11df9 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
@@ -48,7 +48,7 @@ public class JCudaKernels {
 	private final static String ptxFileName = "/kernels/SystemML.ptx";
 	private HashMap<String, CUfunction> kernels = new HashMap<String, CUfunction>();
 	private CUmodule module;
-	private final int deviceNum;
+//	private final int deviceNum;
 	
 	/**
 	 * Loads the kernels in the file ptxFileName. Though cubin files are also supported, we will stick with
@@ -57,7 +57,7 @@ public class JCudaKernels {
 	 * @throws DMLRuntimeException if DMLRuntimeException occurs
 	 */
 	JCudaKernels(int deviceNum) throws DMLRuntimeException {
-		this.deviceNum = deviceNum;
+//		this.deviceNum = deviceNum;
 		module = new CUmodule();
 		// Load the kernels specified in the ptxFileName file
 		checkResult(cuModuleLoadDataEx(module, initKernels(ptxFileName), 0, new int[0], Pointer.to(new int[0])));

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixCUDA.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixCUDA.java b/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixCUDA.java
index 3c32137..9ddd87a 100644
--- a/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixCUDA.java
+++ b/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixCUDA.java
@@ -56,7 +56,6 @@ import static jcuda.jcusparse.JCusparse.cusparseDcsrgemm;
 import static jcuda.jcusparse.JCusparse.cusparseDcsrmv;
 import static jcuda.jcusparse.cusparseOperation.CUSPARSE_OPERATION_NON_TRANSPOSE;
 import static jcuda.jcusparse.cusparseOperation.CUSPARSE_OPERATION_TRANSPOSE;
-import static jcuda.runtime.JCuda.cudaDeviceSynchronize;
 import static jcuda.runtime.JCuda.cudaMemcpy;
 import static jcuda.runtime.cudaMemcpyKind.cudaMemcpyDeviceToDevice;
 import static jcuda.runtime.cudaMemcpyKind.cudaMemcpyDeviceToHost;
@@ -2810,7 +2809,7 @@ public class LibMatrixCUDA {
 			if (GPUStatistics.DISPLAY_STATISTICS) GPUStatistics.maintainCPMiscTimes(instName, GPUInstruction.MISC_TIMER_SPARSE_ALLOCATE_LIB, System.nanoTime() - t1);
 
 			out.getGPUObject(gCtx).setSparseMatrixCudaPointer(C);
-			long sizeOfC = CSRPointer.estimateSize(C.nnz, out.getNumRows());
+			//long sizeOfC = CSRPointer.estimateSize(C.nnz, out.getNumRows());
 			out.getGPUObject(gCtx).addReadLock();
 			if (GPUStatistics.DISPLAY_STATISTICS) t0 = System.nanoTime();
 			JCusparse.cusparseDcsrgeam(getCusparseHandle(gCtx), m, n, alphaPtr, A.descr, (int)A.nnz, A.val, A.rowPtr, A.colInd, betaPtr,

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixReorg.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixReorg.java b/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixReorg.java
index edf69c1..bfaf4c8 100644
--- a/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixReorg.java
+++ b/src/main/java/org/apache/sysml/runtime/matrix/data/LibMatrixReorg.java
@@ -642,6 +642,7 @@ public class LibMatrixReorg
 	 * @param rows ?
 	 * @param cast ?
 	 * @param ignore ?
+	 * @param k degree of parallelism
 	 * @return output matrix
 	 * @throws DMLRuntimeException if DMLRuntimeException occurs
 	 */

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/main/java/org/apache/sysml/runtime/matrix/data/SparseBlockCSR.java
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/runtime/matrix/data/SparseBlockCSR.java b/src/main/java/org/apache/sysml/runtime/matrix/data/SparseBlockCSR.java
index 9ddff9e..a4e680d 100644
--- a/src/main/java/org/apache/sysml/runtime/matrix/data/SparseBlockCSR.java
+++ b/src/main/java/org/apache/sysml/runtime/matrix/data/SparseBlockCSR.java
@@ -171,7 +171,7 @@ public class SparseBlockCSR extends SparseBlock
 	 * 
 	 * @param nnz number of non-zeros to read
 	 * @param in data input stream of ijv triples, ordered by ij
-	 * @throws IOException
+	 * @throws IOException if deserialization error occurs
 	 */
 	public void initUltraSparse(int nnz, DataInput in) 
 		throws IOException 

http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/56e6b6e9/src/test/java/org/apache/sysml/test/integration/AutomatedTestBase.java
----------------------------------------------------------------------
diff --git a/src/test/java/org/apache/sysml/test/integration/AutomatedTestBase.java b/src/test/java/org/apache/sysml/test/integration/AutomatedTestBase.java
index b20f66a..d1cba1d 100644
--- a/src/test/java/org/apache/sysml/test/integration/AutomatedTestBase.java
+++ b/src/test/java/org/apache/sysml/test/integration/AutomatedTestBase.java
@@ -65,10 +65,6 @@ import org.apache.sysml.runtime.util.MapReduceTool;
 import org.apache.sysml.test.utils.TestUtils;
 import org.apache.sysml.utils.ParameterBuilder;
 import org.apache.sysml.utils.Statistics;
-import org.apache.wink.json4j.JSONObject;
-import org.junit.After;
-import org.junit.Assert;
-import org.junit.Before;
 
 
 /**