[
https://issues.apache.org/jira/browse/MAHOUT-1885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15824865#comment-15824865
]
ASF GitHub Bot commented on MAHOUT-1885:
----------------------------------------
Github user andrewpalumbo commented on a diff in the pull request:
https://github.com/apache/mahout/pull/261#discussion_r96336102
--- Diff:
viennacl/src/main/scala/org/apache/mahout/viennacl/opencl/GPUMMul.scala ---
@@ -0,0 +1,455 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.mahout.viennacl.opencl
+
+import org.apache.mahout.logging._
+import org.apache.mahout.math
+import org.apache.mahout.math._
+import org.apache.mahout.math.flavor.{BackEnum, TraversingStructureEnum}
+import org.apache.mahout.math.function.Functions
+import org.apache.mahout.math.scalabindings.RLikeOps._
+import org.apache.mahout.math.scalabindings._
+import org.apache.mahout.viennacl.opencl.javacpp.Functions._
+import org.apache.mahout.viennacl.opencl.javacpp.LinalgFunctions._
+import org.apache.mahout.viennacl.opencl.javacpp.{CompressedMatrix,
Context, DenseRowMatrix}
+
+import scala.collection.JavaConversions._
+
+object GPUMMul extends MMBinaryFunc {
+
+ private final implicit val log = getLog(GPUMMul.getClass)
+
+ override def apply(a: Matrix, b: Matrix, r: Option[Matrix]): Matrix = {
+
+ require(a.ncol == b.nrow, "Incompatible matrix sizes in matrix
multiplication.")
+
+ val (af, bf) = (a.getFlavor, b.getFlavor)
+ val backs = (af.getBacking, bf.getBacking)
+ val sd = (af.getStructure, math.scalabindings.densityAnalysis(a),
bf.getStructure, densityAnalysis(b))
+
+
+ try {
+
+ val alg: MMulAlg = backs match {
+
+ // Both operands are jvm memory backs.
+ case (BackEnum.JVMMEM, BackEnum.JVMMEM) ⇒
--- End diff --
Agreed (obviously this is an old comment due to my underestimation of what
a move across the country and a new job can do to you + a nephew destroying my
Mahout machine)- I don't like keeping both modules around but, quoting myself
from 6 months ago:
> We may be able to get away with only one Viennacl module by setting an
environment variable and checking for it in Context.scala if it exists we
"#define VIENNACL_WITH_OPENCL" which will make the module OpenMP only. The
Problem With This Is That We Can't Test GPU in this module.. without checking
for the environment var.
I guess the above is a possibility, but not especially simple. For no,
though all tests are passing.. though i am seeing that some of the eg. spark
tests are not running `GPUMmul` when Built for `-Pviennacl`. So I may need to
fix some imports somewhere.
Either way I 'd like to get the base Architecture nailed down with this
release.
> Inital Implementation of VCL Bindings
> -------------------------------------
>
> Key: MAHOUT-1885
> URL: https://issues.apache.org/jira/browse/MAHOUT-1885
> Project: Mahout
> Issue Type: Improvement
> Components: Math
> Affects Versions: 0.12.2
> Reporter: Andrew Palumbo
> Assignee: Andrew Palumbo
> Fix For: 0.13.0
>
>
> Push a working experimental branch of VCL bindings into master. There is
> still a lot of work to be done. All tests are passing, At the moment there
> am opening this JIRA mostly to get a number for PR and to test profiles
> against on travis.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
