Visible to Intel only — GUID: GUID-32310FD9-44AE-483E-BABF-665A11E151F4
Introduction to SYCL Essentials
Module 1: oneAPI Intro
Module 2: DPCPP Program Structure
Module 3: DPCPP Unified Shared Memory
Module 4: DPCPP Sub-Groups
Module 5: Intel® Advisor
Module 6: VTune™ Profiler
Module 7: DPCPP Library
Module 8: DPCPP Reduction
Module 9: DPCPP Buffers And Accessors In Depth
Module 10: DPCPP Graphs Scheduling Data Management
Module 11: Intel® Distribution for GDB
Module 12: DPCPP Local Memory And Atomics
Visible to Intel only — GUID: GUID-32310FD9-44AE-483E-BABF-665A11E151F4
SYCL matrix_multiplication Using USM
Matrix multiplication is a fundamental operation in linear algebra. It plays a crucial role in various scientific and engineering applications, such as image processing, simulation, and machine learning. The operation involves taking the dot product of rows from one matrix with the columns of another matrix to produce a resulting matrix. While conceptually simple, efficient computation of matrix multiplication becomes a challenge when dealing with large matrices due to the inherent parallelism that can be leveraged for optimization. This is where SYCL comes into play. SYCL is a programming model and API that enables developers to write code that can be executed on a variety of heterogeneous platforms, including CPUs, GPUs, FPGAs, and other accelerators, using a single-source approach.
In this context, leveraging SYCL for matrix multiplication offers the potential to harness the power of various computing devices while maintaining a unified codebase. By expressing parallelism with SYCL, developers can achieve optimized matrix multiplication routines that efficiently distribute computation across different devices. In the code below we present the matrix multiplication code in SYCL format using USM:
//==============================================================
// Copyright © Intel Corporation
//
// SPDX-License-Identifier: MIT
// =============================================================
#include <CL/sycl.hpp>
using namespace sycl;
static const int N = 3;
int main() {
queue q;
// STEP 1 : Get device information
std::cout << "Device : " << q.get_device().get_info<info::device::name>() << "\n";
// STEP 2 : Initialize vectors in shared memory
int *data1 = malloc_shared<int>(N * N, q);
int *data2 = malloc_shared<int>(N * N, q);
int *dataR = malloc_shared<int>(N * N, q);
// STEP 3 : Asign values to vectors
for (int i = 0; i < N * N; i++) {
data1[i] = 10;
data2[i] = 20;
dataR[i] = 0;
}
// STEP 4 : Proceed with the calculation
q.parallel_for(range<2>{N,N}, [=](item<2> item){
const int j = item.get_id(0);
const int i = item.get_id(1);
for (int k = 0; k < N; ++k)
dataR[i*N+j] += data1[i*N+k] * data2[k*N+j];
}).wait();
// STEP 5 : Print output values
for (int i = 0; i < N * N; i++) std::cout << dataR[i] << " ";
std::cout << "\n";
//# STEP 6 : Release memory
free(data1, q);
free(data2, q);
free(dataR, q);
return 0;
}This program produces the following output:
Device : Intel(R) UHD Graphics [0x9a60]
600 600 600 600 600 600 600 600 600