From b8cc5635d581d3b3ea9950ce8359681ae01491a2 Mon Sep 17 00:00:00 2001 From: Mehdi Goli Date: Wed, 23 Nov 2016 16:30:41 +0000 Subject: Removing unsupported device from test case; cleaning the tensor device sycl. --- unsupported/Eigen/CXX11/src/Tensor/TensorReductionSycl.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'unsupported/Eigen/CXX11/src/Tensor/TensorReductionSycl.h') diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorReductionSycl.h b/unsupported/Eigen/CXX11/src/Tensor/TensorReductionSycl.h index f293869ee..2f7468d56 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorReductionSycl.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorReductionSycl.h @@ -135,8 +135,7 @@ struct FullReducer { /// if the shared memory is less than the GRange, we set shared_mem size to the TotalSize and in this case one kernel would be created for recursion to reduce all to one. if (GRange < outTileSize) outTileSize=GRange; // getting final out buffer at the moment the created buffer is true because there is no need for assign -// auto out_buffer =dev.template get_sycl_buffer::type>(self.dimensions().TotalSize(), output); - auto out_buffer =dev.get_sycl_buffer(self.dimensions().TotalSize(), output); + auto out_buffer =dev.get_sycl_buffer(output); /// creating the shared memory for calculating reduction. /// This one is used to collect all the reduced value of shared memory as we dont have global barrier on GPU. Once it is saved we can @@ -191,7 +190,7 @@ struct InnerReducer { typedef const typename Self::ChildType HostExpr; /// this is the child of reduction typedef typename TensorSycl::internal::createPlaceHolderExpression::Type PlaceHolderExpr; auto functors = TensorSycl::internal::extractFunctors(self.impl()); - size_t range, GRange, tileSize; + typename Self::Index range, GRange, tileSize; dev.parallel_for_setup(num_coeffs_to_preserve, tileSize, range, GRange); // getting final out buffer at the moment the created buffer is true because there is no need for assign /// creating the shared memory for calculating reduction. @@ -204,7 +203,7 @@ struct InnerReducer { dev.sycl_queue().submit([&](cl::sycl::handler &cgh) { // create a tuple of accessors from Evaluator auto tuple_of_accessors = TensorSycl::internal::createTupleOfAccessors(cgh, self.impl()); - auto output_accessor = dev.template get_sycl_accessor(num_coeffs_to_preserve,cgh, output); + auto output_accessor = dev.template get_sycl_accessor(cgh, output); cgh.parallel_for( cl::sycl::nd_range<1>(cl::sycl::range<1>(GRange), cl::sycl::range<1>(tileSize)), [=](cl::sycl::nd_item<1> itemID) { typedef typename TensorSycl::internal::ConvertToDeviceExpression::Type DevExpr; -- cgit v1.2.3