97 lines
4.1 KiB
C++
97 lines
4.1 KiB
C++
/*
|
|
* Copyright (C) 2019-2024 Intel Corporation
|
|
*
|
|
* SPDX-License-Identifier: MIT
|
|
*
|
|
*/
|
|
|
|
#pragma once
|
|
#include "shared/source/helpers/engine_node_helper.h"
|
|
#include "shared/source/os_interface/os_context.h"
|
|
|
|
#include "opencl/source/command_queue/hardware_interface_base.inl"
|
|
|
|
namespace NEO {
|
|
|
|
template <typename GfxFamily>
|
|
inline void HardwareInterface<GfxFamily>::getDefaultDshSpace(
|
|
const size_t &offsetInterfaceDescriptorTable,
|
|
CommandQueue &commandQueue,
|
|
const MultiDispatchInfo &multiDispatchInfo,
|
|
size_t &totalInterfaceDescriptorTableSize,
|
|
IndirectHeap *dsh,
|
|
LinearStream *commandStream) {
|
|
|
|
size_t numDispatches = multiDispatchInfo.size();
|
|
totalInterfaceDescriptorTableSize *= numDispatches;
|
|
|
|
dsh->getSpace(totalInterfaceDescriptorTableSize);
|
|
}
|
|
|
|
template <typename GfxFamily>
|
|
template <typename WalkerType>
|
|
inline void HardwareInterface<GfxFamily>::programWalker(
|
|
LinearStream &commandStream,
|
|
Kernel &kernel,
|
|
CommandQueue &commandQueue,
|
|
IndirectHeap &dsh,
|
|
IndirectHeap &ioh,
|
|
IndirectHeap &ssh,
|
|
const DispatchInfo &dispatchInfo,
|
|
HardwareInterfaceWalkerArgs &walkerArgs) {
|
|
|
|
auto walkerCmdBuf = allocateWalkerSpace<WalkerType>(commandStream, kernel);
|
|
WalkerType walkerCmd = GfxFamily::cmdInitGpgpuWalker;
|
|
uint32_t dim = dispatchInfo.getDim();
|
|
uint32_t simd = kernel.getKernelInfo().getMaxSimdSize();
|
|
auto &rootDeviceEnvironment = commandQueue.getDevice().getRootDeviceEnvironment();
|
|
|
|
size_t startWorkGroups[3] = {walkerArgs.startOfWorkgroups->x, walkerArgs.startOfWorkgroups->y, walkerArgs.startOfWorkgroups->z};
|
|
size_t numWorkGroups[3] = {walkerArgs.numberOfWorkgroups->x, walkerArgs.numberOfWorkgroups->y, walkerArgs.numberOfWorkgroups->z};
|
|
auto threadGroupCount = static_cast<uint32_t>(walkerArgs.numberOfWorkgroups->x * walkerArgs.numberOfWorkgroups->y * walkerArgs.numberOfWorkgroups->z);
|
|
|
|
if (walkerArgs.currentTimestampPacketNodes && walkerArgs.currentTimestampPacketNodes->peekNodes().size() > 0 &&
|
|
commandQueue.getGpgpuCommandStreamReceiver().peekTimestampPacketWriteEnabled()) {
|
|
auto timestampPacketNode = walkerArgs.currentTimestampPacketNodes->peekNodes().at(walkerArgs.currentDispatchIndex);
|
|
GpgpuWalkerHelper<GfxFamily>::setupTimestampPacket(&commandStream, &walkerCmd, timestampPacketNode, rootDeviceEnvironment);
|
|
}
|
|
|
|
auto isCcsUsed = EngineHelpers::isCcs(commandQueue.getGpgpuEngine().osContext->getEngineType());
|
|
auto kernelUsesLocalIds = HardwareCommandsHelper<GfxFamily>::kernelUsesLocalIds(kernel);
|
|
|
|
GpgpuWalkerHelper<GfxFamily>::setGpgpuWalkerThreadData(&walkerCmd, kernel.getKernelInfo().kernelDescriptor,
|
|
startWorkGroups,
|
|
numWorkGroups, walkerArgs.localWorkSizes, simd, dim,
|
|
false, false, 0u);
|
|
|
|
HardwareCommandsHelper<GfxFamily>::template sendIndirectState<WalkerType, INTERFACE_DESCRIPTOR_DATA>(
|
|
commandStream,
|
|
dsh,
|
|
ioh,
|
|
ssh,
|
|
kernel,
|
|
kernel.getKernelStartAddress(true, kernelUsesLocalIds, isCcsUsed, false),
|
|
simd,
|
|
walkerArgs.localWorkSizes,
|
|
threadGroupCount,
|
|
walkerArgs.offsetInterfaceDescriptorTable,
|
|
walkerArgs.interfaceDescriptorIndex,
|
|
walkerArgs.preemptionMode,
|
|
&walkerCmd,
|
|
nullptr,
|
|
kernelUsesLocalIds,
|
|
0,
|
|
commandQueue.getDevice());
|
|
|
|
EncodeWalkerArgs encodeWalkerArgs{
|
|
kernel.getKernelInfo().kernelDescriptor, // kernelDescriptor
|
|
kernel.getExecutionType(), // kernelExecutionType
|
|
RequiredDispatchWalkOrder::none, // requiredDispatchWalkOrder
|
|
0, // additionalSizeParam
|
|
0, // maxFrontEndThreads
|
|
false}; // requiredSystemFence
|
|
EncodeDispatchKernel<GfxFamily>::encodeAdditionalWalkerFields(rootDeviceEnvironment, walkerCmd, encodeWalkerArgs);
|
|
*walkerCmdBuf = walkerCmd;
|
|
}
|
|
} // namespace NEO
|