Kernel Class Files¶
Each kernel in the Suite is implemented in a class whose header and
implementation files reside in src subdirectory named for the group in
which the kernel lives. A kernel class is responsible for implementing all
operations that manage data, execute, and record execution timing, checksum,
and other information for each variant and tuning of a kernel. To properly
integrate into the RAJA Performance Suite framework, the kernel class must be
a subclass of the KernelBase base class that defines the interface for
kernels in the Suite. The KernelBase.hpp header file resides in the
src/common directory.
Continuing with the example we started discussing above, we add the
ADD.hpp header file for the ADD class to the stream directory
along with multiple implementation files. We describe the contents of these
files in the following sections:
ADD.cppcontains methods to set up and tear down the memory for the ADD kernel, and compute and record a checksum on the result after it executes. It also specifies ADD kernel information in theADDclass constructor.
ADD-Seq.cppcontains sequential CPU variants and tunings of the kernel.
ADD-OMP.cppcontains OpenMP CPU multithreading variants and tunings of the kernel.
ADD-OMPTarget.cppcontains OpenMP target offload variants and tunings of the kernel.
ADD-Cuda.cppcontains CUDA GPU variants and tunings of the kernel.
FOO-Hip.cppcontains HIP GPU variants and tunings of the kernel.
Note
All kernels in the Suite follow the same file organization and implementation pattern. Inspection of the files for any kernel helps to understand the overall organization.
Important
If a new execution back-end variant is added that is not listed
here, that variant should be placed in a file named to clearly
distinguish the back-end implementation, such as
ADD-<backend>.cpp. Keeping the variants for each back-end
in a separate file helps to understand compiler optimization
when looking at generated assembly code, for example.
Kernel class header file¶
In its entirety, the ADD kernel class header file ADD.hpp is:
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~//
// Copyright (c) 2017-25, Lawrence Livermore National Security, LLC
// and RAJA Performance Suite project contributors.
// See the RAJAPerf/LICENSE file for details.
//
// SPDX-License-Identifier: (BSD-3-Clause)
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~//
///
/// ADD kernel reference implementation:
///
/// for (Index_type i = ibegin; i < iend; ++i ) {
/// c[i] = a[i] + b[i];
/// }
///
#ifndef RAJAPerf_Stream_ADD_HPP
#define RAJAPerf_Stream_ADD_HPP
#define ADD_DATA_SETUP \
Real_ptr a = m_a; \
Real_ptr b = m_b; \
Real_ptr c = m_c;
#define ADD_BODY \
c[i] = a[i] + b[i];
#include "common/KernelBase.hpp"
namespace rajaperf
{
class RunParams;
namespace stream
{
class ADD : public KernelBase
{
public:
ADD(const RunParams& params);
~ADD();
void setUp(VariantID vid, size_t tune_idx);
void updateChecksum(VariantID vid, size_t tune_idx);
void tearDown(VariantID vid, size_t tune_idx);
void runSeqVariant(VariantID vid, size_t tune_idx);
void runOpenMPVariant(VariantID vid, size_t tune_idx);
void runCudaVariant(VariantID vid, size_t tune_idx);
void runHipVariant(VariantID vid, size_t tune_idx);
void runOpenMPTargetVariant(VariantID vid, size_t tune_idx);
void runSyclVariant(VariantID vid, size_t tune_idx);
void runKokkosVariant(VariantID vid, size_t tune_idx);
void setCudaTuningDefinitions(VariantID vid);
void setHipTuningDefinitions(VariantID vid);
void setSyclTuningDefinitions(VariantID vid);
template < size_t block_size >
void runCudaVariantImpl(VariantID vid);
template < size_t block_size >
void runHipVariantImpl(VariantID vid);
template < size_t work_group_size >
void runSyclVariantImpl(VariantID vid);
private:
static const size_t default_gpu_block_size = 256;
using gpu_block_sizes_type = integer::make_gpu_block_size_list_type<default_gpu_block_size>;
Real_ptr m_a;
Real_ptr m_b;
Real_ptr m_c;
};
} // end namespace stream
} // end namespace rajaperf
#endif // closing endif for header file include guard
The key ingredients of a kernel class header file are:
Copyright statement at the top of the file.
Note
Each file in the RAJA Performance Suite must start with a boilerplate comment for the project copyright information.
Reference implementation, which is a comment section that shows the kernel as it appears in the original code in which it was taken. This is helpful to understand the origin and intent of the original.
Uniquely-named include guard that guards the contents of the header file.
Macro definitions that contain source lines of code that appear in multiple places in the kernel class implementation, such as setting data pointers and operations in the kernel body. While macros obfuscate the code somewhat, we use them to reduce the amount of code we maintain and ensure consistency.
Class definition derived from the
KernelBaseclass. We describe this in more detail below.
Note
All types, methods, etc. in the RAJA Performance Suite reside in the
rajaperfnamespace.In addition, each kernel class lives in the namespace of the kernel group of which the kernel is a member. For example, here, the
ADDclass is in thestreamnamespace.Each kernel class must be derived from the
KernelBaseclass so that the kernel implementation integrates properly into the Suite.
The class must provide a constructor that takes a reference to a RunParams
object, which contains input parameters for running the Suite – we’ll say more
about this later. The class constructor may or may not allocate storage for
a class object. If it does, the storage should be deallocated in the class
destructor.
Several methods in the KernelBase class are pure virtual and the derived
kernel class must provide implementations of those methods. These methods
take a VariantID argument and a tuning index. They include: setUp,
updateChecksum, and tearDown, and methods to run the different kernel
variants. While these method names are descriptive of what they do, we’ll
provide more details about them when we describe the class implementation in
the next section.
Other methods in the code above, such as setCudaTuningDefinitions are
virtual in the KernelBase class and so they may be provided optionally by
the kernel class for kernel specific operations.
Lastly, any data members used in the class implementation are defined,
typically in a private member section so they don’t bleed out of the
kernel class. For example, in the ADD class, we see data members for
GPU block sizes. Also, there are pointer members to hold data arrays for
the kernel. Here we have m_a, m_b``, and m_c for the three arrays
used in the ADD kernel. Note that we use the convention to prefix class data
members with m_.