Media
/
compute-runtime
mirror of https://github.com/intel/compute-runtime.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
compute-runtime/opencl/source/kernel/kernel.cpp

2674 lines
126 KiB
C++
Raw Blame History

/*
* Copyright (C) 2018-2021 Intel Corporation
*
* SPDX-License-Identifier: MIT
*
*/
#include "opencl/source/kernel/kernel.h"
#include "shared/source/built_ins/built_ins.h"
#include "shared/source/command_stream/command_stream_receiver.h"
#include "shared/source/debug_settings/debug_settings_manager.h"
#include "shared/source/device_binary_format/patchtokens_decoder.h"
#include "shared/source/gmm_helper/gmm_helper.h"
#include "shared/source/helpers/aligned_memory.h"
#include "shared/source/helpers/api_specific_config.h"
#include "shared/source/helpers/basic_math.h"
#include "shared/source/helpers/debug_helpers.h"
#include "shared/source/helpers/get_info.h"
#include "shared/source/helpers/hw_helper.h"
#include "shared/source/helpers/kernel_helpers.h"
#include "shared/source/helpers/ptr_math.h"
#include "shared/source/kernel/kernel_arg_descriptor_extended_device_side_enqueue.h"
#include "shared/source/kernel/kernel_arg_descriptor_extended_vme.h"
#include "shared/source/memory_manager/memory_manager.h"
#include "shared/source/memory_manager/unified_memory_manager.h"
#include "opencl/source/accelerators/intel_accelerator.h"
#include "opencl/source/accelerators/intel_motion_estimation.h"
#include "opencl/source/built_ins/builtins_dispatch_builder.h"
#include "opencl/source/cl_device/cl_device.h"
#include "opencl/source/command_queue/command_queue.h"
#include "opencl/source/command_queue/gpgpu_walker.h"
#include "opencl/source/context/context.h"
#include "opencl/source/device_queue/device_queue.h"
#include "opencl/source/execution_model/device_enqueue.h"
#include "opencl/source/gtpin/gtpin_notify.h"
#include "opencl/source/helpers/cl_hw_helper.h"
#include "opencl/source/helpers/dispatch_info.h"
#include "opencl/source/helpers/get_info_status_mapper.h"
#include "opencl/source/helpers/per_thread_data.h"
#include "opencl/source/helpers/sampler_helpers.h"
#include "opencl/source/helpers/surface_formats.h"
#include "opencl/source/kernel/image_transformer.h"
#include "opencl/source/kernel/kernel.inl"
#include "opencl/source/kernel/kernel_info_cl.h"
#include "opencl/source/mem_obj/buffer.h"
#include "opencl/source/mem_obj/image.h"
#include "opencl/source/mem_obj/pipe.h"
#include "opencl/source/memory_manager/mem_obj_surface.h"
#include "opencl/source/platform/platform.h"
#include "opencl/source/program/block_kernel_manager.h"
#include "opencl/source/program/kernel_info.h"
#include "opencl/source/sampler/sampler.h"
#include "patch_list.h"
#include <algorithm>
#include <cstdint>
#include <vector>
using namespace iOpenCL;
namespace NEO {
class Surface;
uint32_t Kernel::dummyPatchLocation = 0xbaddf00d;
Kernel::Kernel(Program *programArg, const KernelInfo &kernelInfoArg, ClDevice &clDeviceArg, bool schedulerKernel)
: isParentKernel(kernelInfoArg.kernelDescriptor.kernelAttributes.flags.usesDeviceSideEnqueue),
isSchedulerKernel(schedulerKernel),
executionEnvironment(programArg->getExecutionEnvironment()),
program(programArg),
clDevice(clDeviceArg),
kernelInfo(kernelInfoArg) {
program->retain();
program->retainForKernel();
imageTransformer.reset(new ImageTransformer);
if (kernelInfoArg.kernelDescriptor.kernelAttributes.simdSize == 1u) {
maxKernelWorkGroupSize = HwHelper::get(getHardwareInfo().platform.eRenderCoreFamily).getMaxThreadsForWorkgroup(getHardwareInfo(), static_cast<uint32_t>(getDevice().getDevice().getDeviceInfo().maxNumEUsPerSubSlice));
} else {
maxKernelWorkGroupSize = static_cast<uint32_t>(clDevice.getSharedDeviceInfo().maxWorkGroupSize);
}
slmTotalSize = kernelInfoArg.kernelDescriptor.kernelAttributes.slmInlineSize;
}
Kernel::~Kernel() {
delete[] crossThreadData;
crossThreadData = nullptr;
crossThreadDataSize = 0;
if (privateSurface) {
program->peekExecutionEnvironment().memoryManager->checkGpuUsageAndDestroyGraphicsAllocations(privateSurface);
privateSurface = nullptr;
}
if (kernelReflectionSurface) {
program->peekExecutionEnvironment().memoryManager->freeGraphicsMemory(kernelReflectionSurface);
kernelReflectionSurface = nullptr;
}
for (uint32_t i = 0; i < patchedArgumentsNum; i++) {
if (SAMPLER_OBJ == getKernelArguments()[i].type) {
auto sampler = castToObject<Sampler>(kernelArguments.at(i).object);
if (sampler) {
sampler->decRefInternal();
}
}
}
kernelArgHandlers.clear();
program->releaseForKernel();
program->release();
}
// If dstOffsetBytes is not an invalid offset, then patches dst at dstOffsetBytes
// with src casted to DstT type.
template <typename DstT, typename SrcT>
inline void patch(const SrcT &src, void *dst, CrossThreadDataOffset dstOffsetBytes) {
if (isValidOffset(dstOffsetBytes)) {
DstT *patchLocation = reinterpret_cast<DstT *>(ptrOffset(dst, dstOffsetBytes));
*patchLocation = static_cast<DstT>(src);
}
}
void Kernel::patchWithImplicitSurface(void *ptrToPatchInCrossThreadData, GraphicsAllocation &allocation, const ArgDescPointer &arg) {
if ((nullptr != crossThreadData) && isValidOffset(arg.stateless)) {
auto pp = ptrOffset(crossThreadData, arg.stateless);
uintptr_t addressToPatch = reinterpret_cast<uintptr_t>(ptrToPatchInCrossThreadData);
patchWithRequiredSize(pp, arg.pointerSize, addressToPatch);
if (DebugManager.flags.AddPatchInfoCommentsForAUBDump.get()) {
PatchInfoData patchInfoData(addressToPatch, 0u, PatchInfoAllocationType::KernelArg, reinterpret_cast<uint64_t>(crossThreadData), arg.stateless, PatchInfoAllocationType::IndirectObjectHeap, arg.pointerSize);
this->patchInfoDataList.push_back(patchInfoData);
}
}
void *ssh = getSurfaceStateHeap();
if ((nullptr != ssh) & isValidOffset(arg.bindful)) {
auto surfaceState = ptrOffset(ssh, arg.bindful);
void *addressToPatch = reinterpret_cast<void *>(allocation.getGpuAddressToPatch());
size_t sizeToPatch = allocation.getUnderlyingBufferSize();
Buffer::setSurfaceState(&clDevice.getDevice(), surfaceState, false, false, sizeToPatch, addressToPatch, 0, &allocation, 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
}
cl_int Kernel::initialize() {
this->kernelHasIndirectAccess = false;
auto pClDevice = &getDevice();
auto rootDeviceIndex = pClDevice->getRootDeviceIndex();
reconfigureKernel();
auto &hwInfo = pClDevice->getHardwareInfo();
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
auto &kernelDescriptor = kernelInfo.kernelDescriptor;
const auto &implicitArgs = kernelDescriptor.payloadMappings.implicitArgs;
const auto &explicitArgs = kernelDescriptor.payloadMappings.explicitArgs;
auto maxSimdSize = kernelInfo.getMaxSimdSize();
const auto &heapInfo = kernelInfo.heapInfo;
if (maxSimdSize != 1 && maxSimdSize < hwHelper.getMinimalSIMDSize()) {
return CL_INVALID_KERNEL;
}
crossThreadDataSize = kernelDescriptor.kernelAttributes.crossThreadDataSize;
// now allocate our own cross-thread data, if necessary
if (crossThreadDataSize) {
crossThreadData = new char[crossThreadDataSize];
if (kernelInfo.crossThreadData) {
memcpy_s(crossThreadData, crossThreadDataSize,
kernelInfo.crossThreadData, crossThreadDataSize);
} else {
memset(crossThreadData, 0x00, crossThreadDataSize);
}
auto crossThread = reinterpret_cast<uint32_t *>(crossThreadData);
auto setArgsIfValidOffset = [&](uint32_t *&crossThreadData, NEO::CrossThreadDataOffset offset, uint32_t value) {
if (isValidOffset(offset)) {
crossThreadData = ptrOffset(crossThread, offset);
*crossThreadData = value;
}
};
setArgsIfValidOffset(maxWorkGroupSizeForCrossThreadData, implicitArgs.maxWorkGroupSize, maxKernelWorkGroupSize);
setArgsIfValidOffset(dataParameterSimdSize, implicitArgs.simdSize, maxSimdSize);
setArgsIfValidOffset(preferredWkgMultipleOffset, implicitArgs.preferredWkgMultiple, maxSimdSize);
setArgsIfValidOffset(parentEventOffset, implicitArgs.deviceSideEnqueueParentEvent, undefined<uint32_t>);
}
// allocate our own SSH, if necessary
sshLocalSize = heapInfo.SurfaceStateHeapSize;
if (sshLocalSize) {
pSshLocal = std::make_unique<char[]>(sshLocalSize);
// copy the ssh into our local copy
memcpy_s(pSshLocal.get(), sshLocalSize,
heapInfo.pSsh, heapInfo.SurfaceStateHeapSize);
}
numberOfBindingTableStates = kernelDescriptor.payloadMappings.bindingTable.numEntries;
localBindingTableOffset = kernelDescriptor.payloadMappings.bindingTable.tableOffset;
// patch crossthread data and ssh with inline surfaces, if necessary
auto perHwThreadPrivateMemorySize = kernelDescriptor.kernelAttributes.perHwThreadPrivateMemorySize;
if (perHwThreadPrivateMemorySize) {
privateSurfaceSize = KernelHelper::getPrivateSurfaceSize(perHwThreadPrivateMemorySize, pClDevice->getSharedDeviceInfo().computeUnitsUsedForScratch);
DEBUG_BREAK_IF(privateSurfaceSize == 0);
if (privateSurfaceSize > std::numeric_limits<uint32_t>::max()) {
return CL_OUT_OF_RESOURCES;
}
privateSurface = executionEnvironment.memoryManager->allocateGraphicsMemoryWithProperties(
{rootDeviceIndex,
static_cast<size_t>(privateSurfaceSize),
GraphicsAllocation::AllocationType::PRIVATE_SURFACE,
pClDevice->getDeviceBitfield()});
if (privateSurface == nullptr) {
return CL_OUT_OF_RESOURCES;
}
const auto &privateMemoryAddress = kernelDescriptor.payloadMappings.implicitArgs.privateMemoryAddress;
patchWithImplicitSurface(reinterpret_cast<void *>(privateSurface->getGpuAddressToPatch()), *privateSurface, privateMemoryAddress);
}
if (isValidOffset(kernelDescriptor.payloadMappings.implicitArgs.globalConstantsSurfaceAddress.stateless)) {
DEBUG_BREAK_IF(program->getConstantSurface(rootDeviceIndex) == nullptr);
uintptr_t constMemory = isBuiltIn ? (uintptr_t)program->getConstantSurface(rootDeviceIndex)->getUnderlyingBuffer() : (uintptr_t)program->getConstantSurface(rootDeviceIndex)->getGpuAddressToPatch();
const auto &arg = kernelDescriptor.payloadMappings.implicitArgs.globalConstantsSurfaceAddress;
patchWithImplicitSurface(reinterpret_cast<void *>(constMemory), *program->getConstantSurface(rootDeviceIndex), arg);
}
if (isValidOffset(kernelDescriptor.payloadMappings.implicitArgs.globalVariablesSurfaceAddress.stateless)) {
DEBUG_BREAK_IF(program->getGlobalSurface(rootDeviceIndex) == nullptr);
uintptr_t globalMemory = isBuiltIn ? (uintptr_t)program->getGlobalSurface(rootDeviceIndex)->getUnderlyingBuffer() : (uintptr_t)program->getGlobalSurface(rootDeviceIndex)->getGpuAddressToPatch();
const auto &arg = kernelDescriptor.payloadMappings.implicitArgs.globalVariablesSurfaceAddress;
patchWithImplicitSurface(reinterpret_cast<void *>(globalMemory), *program->getGlobalSurface(rootDeviceIndex), arg);
}
// Patch Surface State Heap
bool useGlobalAtomics = kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics;
if (isValidOffset(kernelDescriptor.payloadMappings.implicitArgs.deviceSideEnqueueEventPoolSurfaceAddress.bindful)) {
auto surfaceState = ptrOffset(reinterpret_cast<uintptr_t *>(getSurfaceStateHeap()),
kernelDescriptor.payloadMappings.implicitArgs.deviceSideEnqueueEventPoolSurfaceAddress.bindful);
Buffer::setSurfaceState(&pClDevice->getDevice(), surfaceState, false, false, 0, nullptr, 0, nullptr, 0, 0, useGlobalAtomics, areMultipleSubDevicesInContext());
}
if (isValidOffset(kernelDescriptor.payloadMappings.implicitArgs.deviceSideEnqueueDefaultQueueSurfaceAddress.bindful)) {
auto surfaceState = ptrOffset(reinterpret_cast<uintptr_t *>(getSurfaceStateHeap()),
kernelDescriptor.payloadMappings.implicitArgs.deviceSideEnqueueDefaultQueueSurfaceAddress.bindful);
Buffer::setSurfaceState(&pClDevice->getDevice(), surfaceState, false, false, 0, nullptr, 0, nullptr, 0, 0, useGlobalAtomics, areMultipleSubDevicesInContext());
}
setThreadArbitrationPolicy(hwHelper.getDefaultThreadArbitrationPolicy());
if (false == kernelInfo.kernelDescriptor.kernelAttributes.flags.requiresSubgroupIndependentForwardProgress) {
setThreadArbitrationPolicy(ThreadArbitrationPolicy::AgeBased);
}
patchBlocksSimdSize();
auto &clHwHelper = ClHwHelper::get(hwInfo.platform.eRenderCoreFamily);
auxTranslationRequired = HwHelper::renderCompressedBuffersSupported(hwInfo) && clHwHelper.requiresAuxResolves(kernelInfo);
if (DebugManager.flags.ForceAuxTranslationEnabled.get() != -1) {
auxTranslationRequired &= !!DebugManager.flags.ForceAuxTranslationEnabled.get();
}
if (auxTranslationRequired) {
program->getContextPtr()->setResolvesRequiredInKernels(true);
}
if (isParentKernel) {
program->allocateBlockPrivateSurfaces(*pClDevice);
}
if (program->isKernelDebugEnabled() && isValidOffset(kernelDescriptor.payloadMappings.implicitArgs.systemThreadSurfaceAddress.bindful)) {
debugEnabled = true;
}
auto numArgs = explicitArgs.size();
slmSizes.resize(numArgs);
this->kernelHasIndirectAccess |= kernelInfo.kernelDescriptor.kernelAttributes.hasNonKernelArgLoad ||
kernelInfo.kernelDescriptor.kernelAttributes.hasNonKernelArgStore ||
kernelInfo.kernelDescriptor.kernelAttributes.hasNonKernelArgAtomic;
provideInitializationHints();
// resolve the new kernel info to account for kernel handlers
// I think by this time we have decoded the binary and know the number of args etc.
// double check this assumption
bool usingBuffers = false;
kernelArguments.resize(numArgs);
kernelArgHandlers.resize(numArgs);
kernelArgRequiresCacheFlush.resize(numArgs);
for (uint32_t i = 0; i < numArgs; ++i) {
storeKernelArg(i, NONE_OBJ, nullptr, nullptr, 0);
// set the argument handler
const auto &arg = explicitArgs[i];
if (arg.is<ArgDescriptor::ArgTPointer>()) {
if (arg.getTraits().addressQualifier == KernelArgMetadata::AddrLocal) {
kernelArgHandlers[i] = &Kernel::setArgLocal;
} else if (arg.getTraits().typeQualifiers.pipeQual) {
kernelArgHandlers[i] = &Kernel::setArgPipe;
kernelArguments[i].type = PIPE_OBJ;
} else if (arg.getExtendedTypeInfo().isDeviceQueue) {
kernelArgHandlers[i] = &Kernel::setArgDevQueue;
kernelArguments[i].type = DEVICE_QUEUE_OBJ;
} else {
kernelArgHandlers[i] = &Kernel::setArgBuffer;
kernelArguments[i].type = BUFFER_OBJ;
usingBuffers = true;
allBufferArgsStateful &= static_cast<uint32_t>(arg.as<ArgDescPointer>().isPureStateful());
}
} else if (arg.is<ArgDescriptor::ArgTImage>()) {
kernelArgHandlers[i] = &Kernel::setArgImage;
kernelArguments[i].type = IMAGE_OBJ;
usingImages = true;
} else if (arg.is<ArgDescriptor::ArgTSampler>()) {
if (arg.getExtendedTypeInfo().isAccelerator) {
kernelArgHandlers[i] = &Kernel::setArgAccelerator;
} else {
kernelArgHandlers[i] = &Kernel::setArgSampler;
kernelArguments[i].type = SAMPLER_OBJ;
}
} else {
kernelArgHandlers[i] = &Kernel::setArgImmediate;
}
}
if (usingImages && !usingBuffers) {
usingImagesOnly = true;
}
return CL_SUCCESS;
}
cl_int Kernel::cloneKernel(Kernel *pSourceKernel) {
// copy cross thread data to store arguments set to source kernel with clSetKernelArg on immediate data (non-pointer types)
memcpy_s(crossThreadData, crossThreadDataSize,
pSourceKernel->crossThreadData, pSourceKernel->crossThreadDataSize);
DEBUG_BREAK_IF(pSourceKernel->crossThreadDataSize != crossThreadDataSize);
// copy arguments set to source kernel with clSetKernelArg or clSetKernelArgSVMPointer
for (uint32_t i = 0; i < pSourceKernel->kernelArguments.size(); i++) {
if (0 == pSourceKernel->getKernelArgInfo(i).size) {
// skip copying arguments that haven't been set to source kernel
continue;
}
switch (pSourceKernel->kernelArguments[i].type) {
case NONE_OBJ:
// all arguments with immediate data (non-pointer types) have been copied in cross thread data
storeKernelArg(i, NONE_OBJ, nullptr, nullptr, pSourceKernel->getKernelArgInfo(i).size);
patchedArgumentsNum++;
kernelArguments[i].isPatched = true;
break;
case SVM_OBJ:
setArgSvm(i, pSourceKernel->getKernelArgInfo(i).size, const_cast<void *>(pSourceKernel->getKernelArgInfo(i).value),
pSourceKernel->getKernelArgInfo(i).pSvmAlloc, pSourceKernel->getKernelArgInfo(i).svmFlags);
break;
case SVM_ALLOC_OBJ:
setArgSvmAlloc(i, const_cast<void *>(pSourceKernel->getKernelArgInfo(i).value),
(GraphicsAllocation *)pSourceKernel->getKernelArgInfo(i).object);
break;
default:
setArg(i, pSourceKernel->getKernelArgInfo(i).size, pSourceKernel->getKernelArgInfo(i).value);
break;
}
}
// copy additional information other than argument values set to source kernel with clSetKernelExecInfo
for (auto &gfxAlloc : pSourceKernel->kernelSvmGfxAllocations) {
kernelSvmGfxAllocations.push_back(gfxAlloc);
}
for (auto &gfxAlloc : pSourceKernel->kernelUnifiedMemoryGfxAllocations) {
kernelUnifiedMemoryGfxAllocations.push_back(gfxAlloc);
}
this->isBuiltIn = pSourceKernel->isBuiltIn;
return CL_SUCCESS;
}
cl_int Kernel::getInfo(cl_kernel_info paramName, size_t paramValueSize,
void *paramValue, size_t *paramValueSizeRet) const {
cl_int retVal;
const void *pSrc = nullptr;
size_t srcSize = GetInfo::invalidSourceSize;
cl_uint numArgs = 0;
const _cl_program *prog;
const _cl_context *ctxt;
cl_uint refCount = 0;
uint64_t nonCannonizedGpuAddress = 0llu;
switch (paramName) {
case CL_KERNEL_FUNCTION_NAME:
pSrc = kernelInfo.kernelDescriptor.kernelMetadata.kernelName.c_str();
srcSize = kernelInfo.kernelDescriptor.kernelMetadata.kernelName.length() + 1;
break;
case CL_KERNEL_NUM_ARGS:
srcSize = sizeof(cl_uint);
numArgs = static_cast<cl_uint>(kernelInfo.kernelDescriptor.payloadMappings.explicitArgs.size());
pSrc = &numArgs;
break;
case CL_KERNEL_CONTEXT:
ctxt = &program->getContext();
srcSize = sizeof(ctxt);
pSrc = &ctxt;
break;
case CL_KERNEL_PROGRAM:
prog = program;
srcSize = sizeof(prog);
pSrc = &prog;
break;
case CL_KERNEL_REFERENCE_COUNT:
refCount = static_cast<cl_uint>(pMultiDeviceKernel->getRefApiCount());
srcSize = sizeof(refCount);
pSrc = &refCount;
break;
case CL_KERNEL_ATTRIBUTES:
pSrc = kernelInfo.kernelDescriptor.kernelMetadata.kernelLanguageAttributes.c_str();
srcSize = kernelInfo.kernelDescriptor.kernelMetadata.kernelLanguageAttributes.length() + 1;
break;
case CL_KERNEL_BINARY_PROGRAM_INTEL:
pSrc = getKernelHeap();
srcSize = getKernelHeapSize();
break;
case CL_KERNEL_BINARY_GPU_ADDRESS_INTEL:
nonCannonizedGpuAddress = GmmHelper::decanonize(kernelInfo.kernelAllocation->getGpuAddress());
pSrc = &nonCannonizedGpuAddress;
srcSize = sizeof(nonCannonizedGpuAddress);
break;
default:
getAdditionalInfo(paramName, pSrc, srcSize);
break;
}
auto getInfoStatus = GetInfo::getInfo(paramValue, paramValueSize, pSrc, srcSize);
retVal = changeGetInfoStatusToCLResultType(getInfoStatus);
GetInfo::setParamValueReturnSize(paramValueSizeRet, srcSize, getInfoStatus);
return retVal;
}
cl_int Kernel::getArgInfo(cl_uint argIndex, cl_kernel_arg_info paramName, size_t paramValueSize,
void *paramValue, size_t *paramValueSizeRet) const {
cl_int retVal;
const void *pSrc = nullptr;
size_t srcSize = GetInfo::invalidSourceSize;
const auto &args = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs;
if (argIndex >= args.size()) {
retVal = CL_INVALID_ARG_INDEX;
return retVal;
}
const auto &argTraits = args[argIndex].getTraits();
const auto &argMetadata = kernelInfo.kernelDescriptor.explicitArgsExtendedMetadata[argIndex];
cl_kernel_arg_address_qualifier addressQualifier;
cl_kernel_arg_access_qualifier accessQualifier;
cl_kernel_arg_type_qualifier typeQualifier;
switch (paramName) {
case CL_KERNEL_ARG_ADDRESS_QUALIFIER:
addressQualifier = asClKernelArgAddressQualifier(argTraits.getAddressQualifier());
srcSize = sizeof(addressQualifier);
pSrc = &addressQualifier;
break;
case CL_KERNEL_ARG_ACCESS_QUALIFIER:
accessQualifier = asClKernelArgAccessQualifier(argTraits.getAccessQualifier());
srcSize = sizeof(accessQualifier);
pSrc = &accessQualifier;
break;
case CL_KERNEL_ARG_TYPE_QUALIFIER:
typeQualifier = asClKernelArgTypeQualifier(argTraits.typeQualifiers);
srcSize = sizeof(typeQualifier);
pSrc = &typeQualifier;
break;
case CL_KERNEL_ARG_TYPE_NAME:
srcSize = argMetadata.type.length() + 1;
pSrc = argMetadata.type.c_str();
break;
case CL_KERNEL_ARG_NAME:
srcSize = argMetadata.argName.length() + 1;
pSrc = argMetadata.argName.c_str();
break;
default:
break;
}
auto getInfoStatus = GetInfo::getInfo(paramValue, paramValueSize, pSrc, srcSize);
retVal = changeGetInfoStatusToCLResultType(getInfoStatus);
GetInfo::setParamValueReturnSize(paramValueSizeRet, srcSize, getInfoStatus);
return retVal;
}
cl_int Kernel::getWorkGroupInfo(cl_kernel_work_group_info paramName,
size_t paramValueSize, void *paramValue,
size_t *paramValueSizeRet) const {
cl_int retVal = CL_INVALID_VALUE;
const void *pSrc = nullptr;
size_t srcSize = GetInfo::invalidSourceSize;
struct size_t3 {
size_t val[3];
} requiredWorkGroupSize;
cl_ulong localMemorySize;
const auto &kernelDescriptor = kernelInfo.kernelDescriptor;
size_t preferredWorkGroupSizeMultiple = 0;
cl_ulong scratchSize;
cl_ulong privateMemSize;
size_t maxWorkgroupSize;
const auto &hwInfo = clDevice.getHardwareInfo();
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
auto &clHwHelper = ClHwHelper::get(hwInfo.platform.eRenderCoreFamily);
GetInfoHelper info(paramValue, paramValueSize, paramValueSizeRet);
switch (paramName) {
case CL_KERNEL_WORK_GROUP_SIZE:
maxWorkgroupSize = maxKernelWorkGroupSize;
if (DebugManager.flags.UseMaxSimdSizeToDeduceMaxWorkgroupSize.get()) {
auto divisionSize = CommonConstants::maximalSimdSize / kernelInfo.getMaxSimdSize();
maxWorkgroupSize /= divisionSize;
}
srcSize = sizeof(maxWorkgroupSize);
pSrc = &maxWorkgroupSize;
break;
case CL_KERNEL_COMPILE_WORK_GROUP_SIZE:
requiredWorkGroupSize.val[0] = kernelDescriptor.kernelAttributes.requiredWorkgroupSize[0];
requiredWorkGroupSize.val[1] = kernelDescriptor.kernelAttributes.requiredWorkgroupSize[1];
requiredWorkGroupSize.val[2] = kernelDescriptor.kernelAttributes.requiredWorkgroupSize[2];
srcSize = sizeof(requiredWorkGroupSize);
pSrc = &requiredWorkGroupSize;
break;
case CL_KERNEL_LOCAL_MEM_SIZE:
localMemorySize = kernelInfo.kernelDescriptor.kernelAttributes.slmInlineSize;
srcSize = sizeof(localMemorySize);
pSrc = &localMemorySize;
break;
case CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE:
preferredWorkGroupSizeMultiple = kernelInfo.getMaxSimdSize();
if (hwHelper.isFusedEuDispatchEnabled(hwInfo)) {
preferredWorkGroupSizeMultiple *= 2;
}
srcSize = sizeof(preferredWorkGroupSizeMultiple);
pSrc = &preferredWorkGroupSizeMultiple;
break;
case CL_KERNEL_SPILL_MEM_SIZE_INTEL:
scratchSize = kernelDescriptor.kernelAttributes.perThreadScratchSize[0];
srcSize = sizeof(scratchSize);
pSrc = &scratchSize;
break;
case CL_KERNEL_PRIVATE_MEM_SIZE:
privateMemSize = clHwHelper.getKernelPrivateMemSize(kernelInfo);
srcSize = sizeof(privateMemSize);
pSrc = &privateMemSize;
break;
default:
getAdditionalWorkGroupInfo(paramName, pSrc, srcSize);
break;
}
auto getInfoStatus = GetInfo::getInfo(paramValue, paramValueSize, pSrc, srcSize);
retVal = changeGetInfoStatusToCLResultType(getInfoStatus);
GetInfo::setParamValueReturnSize(paramValueSizeRet, srcSize, getInfoStatus);
return retVal;
}
cl_int Kernel::getSubGroupInfo(cl_kernel_sub_group_info paramName,
size_t inputValueSize, const void *inputValue,
size_t paramValueSize, void *paramValue,
size_t *paramValueSizeRet) const {
size_t numDimensions = 0;
size_t WGS = 1;
auto maxSimdSize = static_cast<size_t>(kernelInfo.getMaxSimdSize());
auto maxRequiredWorkGroupSize = static_cast<size_t>(kernelInfo.getMaxRequiredWorkGroupSize(getMaxKernelWorkGroupSize()));
auto largestCompiledSIMDSize = static_cast<size_t>(kernelInfo.getMaxSimdSize());
GetInfoHelper info(paramValue, paramValueSize, paramValueSizeRet);
if ((paramName == CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT) ||
(paramName == CL_KERNEL_MAX_NUM_SUB_GROUPS) ||
(paramName == CL_KERNEL_COMPILE_NUM_SUB_GROUPS)) {
if (clDevice.areOcl21FeaturesEnabled() == false) {
return CL_INVALID_OPERATION;
}
}
if ((paramName == CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE_KHR) ||
(paramName == CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE_KHR)) {
if (!inputValue) {
return CL_INVALID_VALUE;
}
if (inputValueSize % sizeof(size_t) != 0) {
return CL_INVALID_VALUE;
}
numDimensions = inputValueSize / sizeof(size_t);
if (numDimensions == 0 ||
numDimensions > static_cast<size_t>(clDevice.getDeviceInfo().maxWorkItemDimensions)) {
return CL_INVALID_VALUE;
}
}
if (paramName == CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT) {
if (!paramValue) {
return CL_INVALID_VALUE;
}
if (paramValueSize % sizeof(size_t) != 0) {
return CL_INVALID_VALUE;
}
numDimensions = paramValueSize / sizeof(size_t);
if (numDimensions == 0 ||
numDimensions > static_cast<size_t>(clDevice.getDeviceInfo().maxWorkItemDimensions)) {
return CL_INVALID_VALUE;
}
}
switch (paramName) {
case CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE_KHR: {
return changeGetInfoStatusToCLResultType(info.set<size_t>(maxSimdSize));
}
case CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE_KHR: {
for (size_t i = 0; i < numDimensions; i++) {
WGS *= ((size_t *)inputValue)[i];
}
return changeGetInfoStatusToCLResultType(
info.set<size_t>((WGS / maxSimdSize) + std::min(static_cast<size_t>(1), WGS % maxSimdSize))); // add 1 if WGS % maxSimdSize != 0
}
case CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT: {
auto subGroupsNum = *(size_t *)inputValue;
auto workGroupSize = subGroupsNum * largestCompiledSIMDSize;
// return workgroup size in first dimension, the rest shall be 1 in positive case
if (workGroupSize > maxRequiredWorkGroupSize) {
workGroupSize = 0;
}
// If no work group size can accommodate the requested number of subgroups, return 0 in each element of the returned array.
switch (numDimensions) {
case 1:
return changeGetInfoStatusToCLResultType(info.set<size_t>(workGroupSize));
case 2:
struct size_t2 {
size_t val[2];
} workGroupSize2;
workGroupSize2.val[0] = workGroupSize;
workGroupSize2.val[1] = (workGroupSize > 0) ? 1 : 0;
return changeGetInfoStatusToCLResultType(info.set<size_t2>(workGroupSize2));
default:
struct size_t3 {
size_t val[3];
} workGroupSize3;
workGroupSize3.val[0] = workGroupSize;
workGroupSize3.val[1] = (workGroupSize > 0) ? 1 : 0;
workGroupSize3.val[2] = (workGroupSize > 0) ? 1 : 0;
return changeGetInfoStatusToCLResultType(info.set<size_t3>(workGroupSize3));
}
}
case CL_KERNEL_MAX_NUM_SUB_GROUPS: {
// round-up maximum number of subgroups
return changeGetInfoStatusToCLResultType(info.set<size_t>(Math::divideAndRoundUp(maxRequiredWorkGroupSize, largestCompiledSIMDSize)));
}
case CL_KERNEL_COMPILE_NUM_SUB_GROUPS: {
return changeGetInfoStatusToCLResultType(info.set<size_t>(static_cast<size_t>(kernelInfo.kernelDescriptor.kernelMetadata.compiledSubGroupsNumber)));
}
case CL_KERNEL_COMPILE_SUB_GROUP_SIZE_INTEL: {
return changeGetInfoStatusToCLResultType(info.set<size_t>(kernelInfo.kernelDescriptor.kernelMetadata.requiredSubGroupSize));
}
default:
return CL_INVALID_VALUE;
}
}
const void *Kernel::getKernelHeap() const {
return kernelInfo.heapInfo.pKernelHeap;
}
size_t Kernel::getKernelHeapSize() const {
return kernelInfo.heapInfo.KernelHeapSize;
}
void Kernel::substituteKernelHeap(void *newKernelHeap, size_t newKernelHeapSize) {
KernelInfo *pKernelInfo = const_cast<KernelInfo *>(&kernelInfo);
void **pKernelHeap = const_cast<void **>(&pKernelInfo->heapInfo.pKernelHeap);
*pKernelHeap = newKernelHeap;
auto &heapInfo = pKernelInfo->heapInfo;
heapInfo.KernelHeapSize = static_cast<uint32_t>(newKernelHeapSize);
pKernelInfo->isKernelHeapSubstituted = true;
auto memoryManager = executionEnvironment.memoryManager.get();
auto currentAllocationSize = pKernelInfo->kernelAllocation->getUnderlyingBufferSize();
bool status = false;
const auto &hwInfo = clDevice.getHardwareInfo();
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
size_t isaPadding = hwHelper.getPaddingForISAAllocation();
if (currentAllocationSize >= newKernelHeapSize + isaPadding) {
auto &hwInfo = clDevice.getDevice().getHardwareInfo();
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
status = MemoryTransferHelper::transferMemoryToAllocation(hwHelper.isBlitCopyRequiredForLocalMemory(hwInfo, *pKernelInfo->getGraphicsAllocation()),
clDevice.getDevice(), pKernelInfo->getGraphicsAllocation(), 0, newKernelHeap,
static_cast<size_t>(newKernelHeapSize));
} else {
memoryManager->checkGpuUsageAndDestroyGraphicsAllocations(pKernelInfo->kernelAllocation);
pKernelInfo->kernelAllocation = nullptr;
status = pKernelInfo->createKernelAllocation(clDevice.getDevice(), isBuiltIn);
}
UNRECOVERABLE_IF(!status);
}
bool Kernel::isKernelHeapSubstituted() const {
return kernelInfo.isKernelHeapSubstituted;
}
uint64_t Kernel::getKernelId() const {
return kernelInfo.kernelId;
}
void Kernel::setKernelId(uint64_t newKernelId) {
KernelInfo *pKernelInfo = const_cast<KernelInfo *>(&kernelInfo);
pKernelInfo->kernelId = newKernelId;
}
uint32_t Kernel::getStartOffset() const {
return this->startOffset;
}
void Kernel::setStartOffset(uint32_t offset) {
this->startOffset = offset;
}
void *Kernel::getSurfaceStateHeap() const {
return pSshLocal.get();
}
size_t Kernel::getDynamicStateHeapSize() const {
return kernelInfo.heapInfo.DynamicStateHeapSize;
}
const void *Kernel::getDynamicStateHeap() const {
return kernelInfo.heapInfo.pDsh;
}
size_t Kernel::getSurfaceStateHeapSize() const {
return sshLocalSize;
}
size_t Kernel::getNumberOfBindingTableStates() const {
return numberOfBindingTableStates;
}
void Kernel::resizeSurfaceStateHeap(void *pNewSsh, size_t newSshSize, size_t newBindingTableCount, size_t newBindingTableOffset) {
pSshLocal.reset(static_cast<char *>(pNewSsh));
sshLocalSize = static_cast<uint32_t>(newSshSize);
numberOfBindingTableStates = newBindingTableCount;
localBindingTableOffset = newBindingTableOffset;
}
void Kernel::markArgPatchedAndResolveArgs(uint32_t argIndex) {
if (!kernelArguments[argIndex].isPatched) {
patchedArgumentsNum++;
kernelArguments[argIndex].isPatched = true;
}
resolveArgs();
}
cl_int Kernel::setArg(uint32_t argIndex, size_t argSize, const void *argVal) {
cl_int retVal = CL_SUCCESS;
bool updateExposedKernel = true;
auto argWasUncacheable = false;
if (kernelInfo.builtinDispatchBuilder != nullptr) {
updateExposedKernel = kernelInfo.builtinDispatchBuilder->setExplicitArg(argIndex, argSize, argVal, retVal);
}
if (updateExposedKernel) {
if (argIndex >= kernelArgHandlers.size()) {
return CL_INVALID_ARG_INDEX;
}
argWasUncacheable = kernelArguments[argIndex].isStatelessUncacheable;
auto argHandler = kernelArgHandlers[argIndex];
retVal = (this->*argHandler)(argIndex, argSize, argVal);
}
if (retVal == CL_SUCCESS) {
auto argIsUncacheable = kernelArguments[argIndex].isStatelessUncacheable;
statelessUncacheableArgsCount += (argIsUncacheable ? 1 : 0) - (argWasUncacheable ? 1 : 0);
markArgPatchedAndResolveArgs(argIndex);
}
return retVal;
}
cl_int Kernel::setArg(uint32_t argIndex, uint32_t argVal) {
return setArg(argIndex, sizeof(argVal), &argVal);
}
cl_int Kernel::setArg(uint32_t argIndex, uint64_t argVal) {
return setArg(argIndex, sizeof(argVal), &argVal);
}
cl_int Kernel::setArg(uint32_t argIndex, cl_mem argVal) {
return setArg(argIndex, sizeof(argVal), &argVal);
}
cl_int Kernel::setArg(uint32_t argIndex, cl_mem argVal, uint32_t mipLevel) {
auto retVal = setArgImageWithMipLevel(argIndex, sizeof(argVal), &argVal, mipLevel);
if (retVal == CL_SUCCESS) {
markArgPatchedAndResolveArgs(argIndex);
}
return retVal;
}
void *Kernel::patchBufferOffset(const ArgDescPointer &argAsPtr, void *svmPtr, GraphicsAllocation *svmAlloc) {
if (isUndefinedOffset(argAsPtr.bufferOffset)) {
return svmPtr;
}
void *ptrToPatch = svmPtr;
if (svmAlloc != nullptr) {
ptrToPatch = reinterpret_cast<void *>(svmAlloc->getGpuAddressToPatch());
}
constexpr uint32_t minimumAlignment = 4;
ptrToPatch = alignDown(ptrToPatch, minimumAlignment);
DEBUG_BREAK_IF(ptrDiff(svmPtr, ptrToPatch) != static_cast<uint32_t>(ptrDiff(svmPtr, ptrToPatch)));
uint32_t offsetToPatch = static_cast<uint32_t>(ptrDiff(svmPtr, ptrToPatch));
patch<uint32_t, uint32_t>(offsetToPatch, getCrossThreadData(), argAsPtr.bufferOffset);
return ptrToPatch;
}
cl_int Kernel::setArgSvm(uint32_t argIndex, size_t svmAllocSize, void *svmPtr, GraphicsAllocation *svmAlloc, cl_mem_flags svmFlags) {
const auto &argAsPtr = getKernelInfo().kernelDescriptor.payloadMappings.explicitArgs[argIndex].as<ArgDescPointer>();
auto patchLocation = ptrOffset(getCrossThreadData(), argAsPtr.stateless);
patchWithRequiredSize(patchLocation, argAsPtr.pointerSize, reinterpret_cast<uintptr_t>(svmPtr));
void *ptrToPatch = patchBufferOffset(argAsPtr, svmPtr, svmAlloc);
if (isValidOffset(argAsPtr.bindful)) {
auto surfaceState = ptrOffset(getSurfaceStateHeap(), argAsPtr.bindful);
Buffer::setSurfaceState(&getDevice().getDevice(), surfaceState, false, false, svmAllocSize + ptrDiff(svmPtr, ptrToPatch), ptrToPatch, 0, svmAlloc, svmFlags, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
storeKernelArg(argIndex, SVM_OBJ, nullptr, svmPtr, sizeof(void *), svmAlloc, svmFlags);
if (!kernelArguments[argIndex].isPatched) {
patchedArgumentsNum++;
kernelArguments[argIndex].isPatched = true;
}
addAllocationToCacheFlushVector(argIndex, svmAlloc);
return CL_SUCCESS;
}
cl_int Kernel::setArgSvmAlloc(uint32_t argIndex, void *svmPtr, GraphicsAllocation *svmAlloc) {
DBG_LOG_INPUTS("setArgBuffer svm_alloc", svmAlloc);
const auto &argAsPtr = getKernelInfo().kernelDescriptor.payloadMappings.explicitArgs[argIndex].as<ArgDescPointer>();
auto patchLocation = ptrOffset(getCrossThreadData(), argAsPtr.stateless);
patchWithRequiredSize(patchLocation, argAsPtr.pointerSize, reinterpret_cast<uintptr_t>(svmPtr));
bool disableL3 = false;
bool forceNonAuxMode = false;
bool isAuxTranslationKernel = (AuxTranslationDirection::None != auxTranslationDirection);
auto &hwInfo = getDevice().getHardwareInfo();
auto &clHwHelper = ClHwHelper::get(hwInfo.platform.eRenderCoreFamily);
if (isAuxTranslationKernel) {
if (((AuxTranslationDirection::AuxToNonAux == auxTranslationDirection) && argIndex == 1) ||
((AuxTranslationDirection::NonAuxToAux == auxTranslationDirection) && argIndex == 0)) {
forceNonAuxMode = true;
}
disableL3 = (argIndex == 0);
} else if (svmAlloc && svmAlloc->getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_COMPRESSED &&
clHwHelper.requiresNonAuxMode(argAsPtr)) {
forceNonAuxMode = true;
}
bool argWasUncacheable = kernelArguments[argIndex].isStatelessUncacheable;
bool argIsUncacheable = svmAlloc ? svmAlloc->isUncacheable() : false;
statelessUncacheableArgsCount += (argIsUncacheable ? 1 : 0) - (argWasUncacheable ? 1 : 0);
void *ptrToPatch = patchBufferOffset(argAsPtr, svmPtr, svmAlloc);
if (isValidOffset(argAsPtr.bindful)) {
auto surfaceState = ptrOffset(getSurfaceStateHeap(), argAsPtr.bindful);
size_t allocSize = 0;
size_t offset = 0;
if (svmAlloc != nullptr) {
allocSize = svmAlloc->getUnderlyingBufferSize();
offset = ptrDiff(ptrToPatch, svmAlloc->getGpuAddressToPatch());
allocSize -= offset;
}
Buffer::setSurfaceState(&getDevice().getDevice(), surfaceState, forceNonAuxMode, disableL3, allocSize, ptrToPatch, offset, svmAlloc, 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
storeKernelArg(argIndex, SVM_ALLOC_OBJ, svmAlloc, svmPtr, sizeof(uintptr_t));
if (!kernelArguments[argIndex].isPatched) {
patchedArgumentsNum++;
kernelArguments[argIndex].isPatched = true;
}
addAllocationToCacheFlushVector(argIndex, svmAlloc);
return CL_SUCCESS;
}
void Kernel::storeKernelArg(uint32_t argIndex, kernelArgType argType, void *argObject,
const void *argValue, size_t argSize,
GraphicsAllocation *argSvmAlloc, cl_mem_flags argSvmFlags) {
kernelArguments[argIndex].type = argType;
kernelArguments[argIndex].object = argObject;
kernelArguments[argIndex].value = argValue;
kernelArguments[argIndex].size = argSize;
kernelArguments[argIndex].pSvmAlloc = argSvmAlloc;
kernelArguments[argIndex].svmFlags = argSvmFlags;
}
const void *Kernel::getKernelArg(uint32_t argIndex) const {
return kernelArguments[argIndex].object;
}
const Kernel::SimpleKernelArgInfo &Kernel::getKernelArgInfo(uint32_t argIndex) const {
return kernelArguments[argIndex];
}
void Kernel::setSvmKernelExecInfo(GraphicsAllocation *argValue) {
kernelSvmGfxAllocations.push_back(argValue);
if (allocationForCacheFlush(argValue)) {
svmAllocationsRequireCacheFlush = true;
}
}
void Kernel::clearSvmKernelExecInfo() {
kernelSvmGfxAllocations.clear();
svmAllocationsRequireCacheFlush = false;
}
void Kernel::setUnifiedMemoryProperty(cl_kernel_exec_info infoType, bool infoValue) {
if (infoType == CL_KERNEL_EXEC_INFO_INDIRECT_DEVICE_ACCESS_INTEL) {
this->unifiedMemoryControls.indirectDeviceAllocationsAllowed = infoValue;
return;
}
if (infoType == CL_KERNEL_EXEC_INFO_INDIRECT_HOST_ACCESS_INTEL) {
this->unifiedMemoryControls.indirectHostAllocationsAllowed = infoValue;
return;
}
if (infoType == CL_KERNEL_EXEC_INFO_INDIRECT_SHARED_ACCESS_INTEL) {
this->unifiedMemoryControls.indirectSharedAllocationsAllowed = infoValue;
return;
}
}
void Kernel::setUnifiedMemoryExecInfo(GraphicsAllocation *unifiedMemoryAllocation) {
kernelUnifiedMemoryGfxAllocations.push_back(unifiedMemoryAllocation);
}
void Kernel::clearUnifiedMemoryExecInfo() {
kernelUnifiedMemoryGfxAllocations.clear();
}
cl_int Kernel::setKernelExecutionType(cl_execution_info_kernel_type_intel executionType) {
switch (executionType) {
case CL_KERNEL_EXEC_INFO_DEFAULT_TYPE_INTEL:
this->executionType = KernelExecutionType::Default;
break;
case CL_KERNEL_EXEC_INFO_CONCURRENT_TYPE_INTEL:
this->executionType = KernelExecutionType::Concurrent;
break;
default: {
return CL_INVALID_VALUE;
}
}
return CL_SUCCESS;
}
void Kernel::getSuggestedLocalWorkSize(const cl_uint workDim, const size_t *globalWorkSize, const size_t *globalWorkOffset,
size_t *localWorkSize) {
UNRECOVERABLE_IF((workDim == 0) || (workDim > 3));
UNRECOVERABLE_IF(globalWorkSize == nullptr);
Vec3<size_t> elws{0, 0, 0};
Vec3<size_t> gws{
globalWorkSize[0],
(workDim > 1) ? globalWorkSize[1] : 1,
(workDim > 2) ? globalWorkSize[2] : 1};
Vec3<size_t> offset{0, 0, 0};
if (globalWorkOffset) {
offset.x = globalWorkOffset[0];
if (workDim > 1) {
offset.y = globalWorkOffset[1];
if (workDim > 2) {
offset.z = globalWorkOffset[2];
}
}
}
Vec3<size_t> suggestedLws{0, 0, 0};
if (kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[0] != 0) {
suggestedLws.x = kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[0];
suggestedLws.y = kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[1];
suggestedLws.z = kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[2];
} else {
uint32_t dispatchWorkDim = std::max(1U, std::max(gws.getSimplifiedDim(), offset.getSimplifiedDim()));
const DispatchInfo dispatchInfo{&clDevice, this, dispatchWorkDim, gws, elws, offset};
suggestedLws = computeWorkgroupSize(dispatchInfo);
}
localWorkSize[0] = suggestedLws.x;
if (workDim > 1)
localWorkSize[1] = suggestedLws.y;
if (workDim > 2)
localWorkSize[2] = suggestedLws.z;
}
uint32_t Kernel::getMaxWorkGroupCount(const cl_uint workDim, const size_t *localWorkSize, const CommandQueue *commandQueue) const {
auto &hardwareInfo = getHardwareInfo();
auto &hwHelper = HwHelper::get(hardwareInfo.platform.eRenderCoreFamily);
auto engineGroupType = hwHelper.getEngineGroupType(commandQueue->getGpgpuEngine().getEngineType(), hardwareInfo);
if (!hwHelper.isCooperativeDispatchSupported(engineGroupType, hardwareInfo.platform.eProductFamily)) {
return 0;
}
const auto &kernelDescriptor = kernelInfo.kernelDescriptor;
auto dssCount = hardwareInfo.gtSystemInfo.DualSubSliceCount;
if (dssCount == 0) {
dssCount = hardwareInfo.gtSystemInfo.SubSliceCount;
}
auto availableThreadCount = hwHelper.calculateAvailableThreadCount(
hardwareInfo.platform.eProductFamily,
kernelDescriptor.kernelAttributes.numGrfRequired,
hardwareInfo.gtSystemInfo.EUCount, hardwareInfo.gtSystemInfo.ThreadCount / hardwareInfo.gtSystemInfo.EUCount);
auto barrierCount = kernelDescriptor.kernelAttributes.barrierCount;
return KernelHelper::getMaxWorkGroupCount(kernelInfo.getMaxSimdSize(),
availableThreadCount,
dssCount,
dssCount * KB * hardwareInfo.capabilityTable.slmSize,
hwHelper.alignSlmSize(slmTotalSize),
static_cast<uint32_t>(hwHelper.getMaxBarrierRegisterPerSlice()),
hwHelper.getBarriersCountFromHasBarriers(barrierCount),
workDim,
localWorkSize);
}
inline void Kernel::makeArgsResident(CommandStreamReceiver &commandStreamReceiver) {
auto numArgs = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs.size();
for (decltype(numArgs) argIndex = 0; argIndex < numArgs; argIndex++) {
if (kernelArguments[argIndex].object) {
if (kernelArguments[argIndex].type == SVM_ALLOC_OBJ) {
auto pSVMAlloc = (GraphicsAllocation *)kernelArguments[argIndex].object;
auto pageFaultManager = executionEnvironment.memoryManager->getPageFaultManager();
if (pageFaultManager &&
this->isUnifiedMemorySyncRequired) {
pageFaultManager->moveAllocationToGpuDomain(reinterpret_cast<void *>(pSVMAlloc->getGpuAddress()));
}
commandStreamReceiver.makeResident(*pSVMAlloc);
} else if (Kernel::isMemObj(kernelArguments[argIndex].type)) {
auto clMem = const_cast<cl_mem>(static_cast<const _cl_mem *>(kernelArguments[argIndex].object));
auto memObj = castToObjectOrAbort<MemObj>(clMem);
auto image = castToObject<Image>(clMem);
if (image && image->isImageFromImage()) {
commandStreamReceiver.setSamplerCacheFlushRequired(CommandStreamReceiver::SamplerCacheFlushState::samplerCacheFlushBefore);
}
commandStreamReceiver.makeResident(*memObj->getGraphicsAllocation(commandStreamReceiver.getRootDeviceIndex()));
if (memObj->getMcsAllocation()) {
commandStreamReceiver.makeResident(*memObj->getMcsAllocation());
}
}
}
}
}
void Kernel::performKernelTuning(CommandStreamReceiver &commandStreamReceiver, const Vec3<size_t> &lws, const Vec3<size_t> &gws, const Vec3<size_t> &offsets, TimestampPacketContainer *timestampContainer) {
auto performTunning = TunningType::DISABLED;
if (DebugManager.flags.EnableKernelTunning.get() != -1) {
performTunning = static_cast<TunningType>(DebugManager.flags.EnableKernelTunning.get());
}
if (performTunning == TunningType::SIMPLE) {
this->singleSubdevicePreferedInCurrentEnqueue = !this->kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics;
} else if (performTunning == TunningType::FULL) {
KernelConfig config{gws, lws, offsets};
auto submissionDataIt = this->kernelSubmissionMap.find(config);
if (submissionDataIt == this->kernelSubmissionMap.end()) {
KernelSubmissionData submissionData;
submissionData.kernelStandardTimestamps = std::make_unique<TimestampPacketContainer>();
submissionData.kernelSubdeviceTimestamps = std::make_unique<TimestampPacketContainer>();
submissionData.status = TunningStatus::STANDARD_TUNNING_IN_PROGRESS;
submissionData.kernelStandardTimestamps->assignAndIncrementNodesRefCounts(*timestampContainer);
this->kernelSubmissionMap[config] = std::move(submissionData);
this->singleSubdevicePreferedInCurrentEnqueue = false;
return;
}
auto &submissionData = submissionDataIt->second;
if (submissionData.status == TunningStatus::TUNNING_DONE) {
this->singleSubdevicePreferedInCurrentEnqueue = submissionData.singleSubdevicePrefered;
}
if (submissionData.status == TunningStatus::SUBDEVICE_TUNNING_IN_PROGRESS) {
if (this->hasTunningFinished(submissionData)) {
submissionData.status = TunningStatus::TUNNING_DONE;
submissionData.kernelStandardTimestamps.reset();
submissionData.kernelSubdeviceTimestamps.reset();
this->singleSubdevicePreferedInCurrentEnqueue = submissionData.singleSubdevicePrefered;
} else {
this->singleSubdevicePreferedInCurrentEnqueue = false;
}
}
if (submissionData.status == TunningStatus::STANDARD_TUNNING_IN_PROGRESS) {
submissionData.status = TunningStatus::SUBDEVICE_TUNNING_IN_PROGRESS;
submissionData.kernelSubdeviceTimestamps->assignAndIncrementNodesRefCounts(*timestampContainer);
this->singleSubdevicePreferedInCurrentEnqueue = true;
}
}
}
bool Kernel::hasTunningFinished(KernelSubmissionData &submissionData) {
if (!this->hasRunFinished(submissionData.kernelStandardTimestamps.get()) ||
!this->hasRunFinished(submissionData.kernelSubdeviceTimestamps.get())) {
return false;
}
uint64_t globalStartTS = 0u;
uint64_t globalEndTS = 0u;
Event::getBoundaryTimestampValues(submissionData.kernelStandardTimestamps.get(), globalStartTS, globalEndTS);
auto standardTSDiff = globalEndTS - globalStartTS;
Event::getBoundaryTimestampValues(submissionData.kernelSubdeviceTimestamps.get(), globalStartTS, globalEndTS);
auto subdeviceTSDiff = globalEndTS - globalStartTS;
submissionData.singleSubdevicePrefered = standardTSDiff > subdeviceTSDiff;
return true;
}
bool Kernel::hasRunFinished(TimestampPacketContainer *timestampContainer) {
for (const auto &node : timestampContainer->peekNodes()) {
for (uint32_t i = 0; i < node->getPacketsUsed(); i++) {
if (node->getContextEndValue(i) == 1) {
return false;
}
}
}
return true;
}
bool Kernel::isSingleSubdevicePreferred() const {
return this->singleSubdevicePreferedInCurrentEnqueue;
}
void Kernel::makeResident(CommandStreamReceiver &commandStreamReceiver) {
auto rootDeviceIndex = commandStreamReceiver.getRootDeviceIndex();
if (privateSurface) {
commandStreamReceiver.makeResident(*privateSurface);
}
if (program->getConstantSurface(rootDeviceIndex)) {
commandStreamReceiver.makeResident(*(program->getConstantSurface(rootDeviceIndex)));
}
if (program->getGlobalSurface(rootDeviceIndex)) {
commandStreamReceiver.makeResident(*(program->getGlobalSurface(rootDeviceIndex)));
}
if (program->getExportedFunctionsSurface(rootDeviceIndex)) {
commandStreamReceiver.makeResident(*(program->getExportedFunctionsSurface(rootDeviceIndex)));
}
for (auto gfxAlloc : kernelSvmGfxAllocations) {
commandStreamReceiver.makeResident(*gfxAlloc);
}
auto pageFaultManager = program->peekExecutionEnvironment().memoryManager->getPageFaultManager();
for (auto gfxAlloc : kernelUnifiedMemoryGfxAllocations) {
commandStreamReceiver.makeResident(*gfxAlloc);
if (pageFaultManager) {
pageFaultManager->moveAllocationToGpuDomain(reinterpret_cast<void *>(gfxAlloc->getGpuAddress()));
}
}
if (unifiedMemoryControls.indirectSharedAllocationsAllowed && pageFaultManager) {
pageFaultManager->moveAllocationsWithinUMAllocsManagerToGpuDomain(this->getContext().getSVMAllocsManager());
}
makeArgsResident(commandStreamReceiver);
auto kernelIsaAllocation = this->kernelInfo.kernelAllocation;
if (kernelIsaAllocation) {
commandStreamReceiver.makeResident(*kernelIsaAllocation);
}
gtpinNotifyMakeResident(this, &commandStreamReceiver);
if (unifiedMemoryControls.indirectDeviceAllocationsAllowed ||
unifiedMemoryControls.indirectHostAllocationsAllowed ||
unifiedMemoryControls.indirectSharedAllocationsAllowed) {
this->getContext().getSVMAllocsManager()->makeInternalAllocationsResident(commandStreamReceiver, unifiedMemoryControls.generateMask());
}
}
void Kernel::getResidency(std::vector<Surface *> &dst) {
if (privateSurface) {
GeneralSurface *surface = new GeneralSurface(privateSurface);
dst.push_back(surface);
}
auto rootDeviceIndex = getDevice().getRootDeviceIndex();
if (program->getConstantSurface(rootDeviceIndex)) {
GeneralSurface *surface = new GeneralSurface(program->getConstantSurface(rootDeviceIndex));
dst.push_back(surface);
}
if (program->getGlobalSurface(rootDeviceIndex)) {
GeneralSurface *surface = new GeneralSurface(program->getGlobalSurface(rootDeviceIndex));
dst.push_back(surface);
}
if (program->getExportedFunctionsSurface(rootDeviceIndex)) {
GeneralSurface *surface = new GeneralSurface(program->getExportedFunctionsSurface(rootDeviceIndex));
dst.push_back(surface);
}
for (auto gfxAlloc : kernelSvmGfxAllocations) {
GeneralSurface *surface = new GeneralSurface(gfxAlloc);
dst.push_back(surface);
}
auto numArgs = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs.size();
for (decltype(numArgs) argIndex = 0; argIndex < numArgs; argIndex++) {
if (kernelArguments[argIndex].object) {
if (kernelArguments[argIndex].type == SVM_ALLOC_OBJ) {
auto pSVMAlloc = (GraphicsAllocation *)kernelArguments[argIndex].object;
dst.push_back(new GeneralSurface(pSVMAlloc));
} else if (Kernel::isMemObj(kernelArguments[argIndex].type)) {
auto clMem = const_cast<cl_mem>(static_cast<const _cl_mem *>(kernelArguments[argIndex].object));
auto memObj = castToObject<MemObj>(clMem);
DEBUG_BREAK_IF(memObj == nullptr);
dst.push_back(new MemObjSurface(memObj));
}
}
}
auto kernelIsaAllocation = this->kernelInfo.kernelAllocation;
if (kernelIsaAllocation) {
GeneralSurface *surface = new GeneralSurface(kernelIsaAllocation);
dst.push_back(surface);
}
gtpinNotifyUpdateResidencyList(this, &dst);
}
bool Kernel::requiresCoherency() {
auto numArgs = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs.size();
for (decltype(numArgs) argIndex = 0; argIndex < numArgs; argIndex++) {
if (kernelArguments[argIndex].object) {
if (kernelArguments[argIndex].type == SVM_ALLOC_OBJ) {
auto pSVMAlloc = (GraphicsAllocation *)kernelArguments[argIndex].object;
if (pSVMAlloc->isCoherent()) {
return true;
}
}
if (Kernel::isMemObj(kernelArguments[argIndex].type)) {
auto clMem = const_cast<cl_mem>(static_cast<const _cl_mem *>(kernelArguments[argIndex].object));
auto memObj = castToObjectOrAbort<MemObj>(clMem);
if (memObj->getMultiGraphicsAllocation().isCoherent()) {
return true;
}
}
}
}
return false;
}
cl_int Kernel::setArgLocal(uint32_t argIndexIn,
size_t argSize,
const void *argVal) {
storeKernelArg(argIndexIn, SLM_OBJ, nullptr, argVal, argSize);
uint32_t *crossThreadData = reinterpret_cast<uint32_t *>(this->crossThreadData);
uint32_t argIndex = argIndexIn;
const auto &args = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs;
const auto &currArg = args[argIndex];
UNRECOVERABLE_IF(currArg.getTraits().getAddressQualifier() != KernelArgMetadata::AddrLocal);
slmSizes[argIndex] = static_cast<uint32_t>(argSize);
UNRECOVERABLE_IF(isUndefinedOffset(currArg.as<NEO::ArgDescPointer>().slmOffset));
auto slmOffset = *ptrOffset(crossThreadData, currArg.as<ArgDescPointer>().slmOffset);
slmOffset += static_cast<uint32_t>(argSize);
++argIndex;
while (argIndex < slmSizes.size()) {
if (args[argIndex].getTraits().getAddressQualifier() != KernelArgMetadata::AddrLocal) {
++argIndex;
continue;
}
const auto &nextArg = args[argIndex].as<ArgDescPointer>();
UNRECOVERABLE_IF(0 == nextArg.requiredSlmAlignment);
slmOffset = alignUp<uint32_t>(slmOffset, nextArg.requiredSlmAlignment);
auto patchLocation = ptrOffset(crossThreadData, nextArg.slmOffset);
*patchLocation = slmOffset;
slmOffset += static_cast<uint32_t>(slmSizes[argIndex]);
++argIndex;
}
slmTotalSize = kernelInfo.kernelDescriptor.kernelAttributes.slmInlineSize + alignUp(slmOffset, KB);
return CL_SUCCESS;
}
cl_int Kernel::setArgBuffer(uint32_t argIndex,
size_t argSize,
const void *argVal) {
if (argSize != sizeof(cl_mem *)) {
return CL_INVALID_ARG_SIZE;
}
auto clMem = reinterpret_cast<const cl_mem *>(argVal);
auto pClDevice = &getDevice();
auto rootDeviceIndex = pClDevice->getRootDeviceIndex();
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex];
const auto &argAsPtr = arg.as<ArgDescPointer>();
if (clMem && *clMem) {
auto clMemObj = *clMem;
DBG_LOG_INPUTS("setArgBuffer cl_mem", clMemObj);
storeKernelArg(argIndex, BUFFER_OBJ, clMemObj, argVal, argSize);
auto buffer = castToObject<Buffer>(clMemObj);
if (!buffer)
return CL_INVALID_MEM_OBJECT;
if (buffer->peekSharingHandler()) {
usingSharedObjArgs = true;
}
patchBufferOffset(argAsPtr, nullptr, nullptr);
if (isValidOffset(argAsPtr.stateless)) {
auto patchLocation = ptrOffset(crossThreadData, argAsPtr.stateless);
uint64_t addressToPatch = buffer->setArgStateless(patchLocation, argAsPtr.pointerSize, rootDeviceIndex, !this->isBuiltIn);
if (DebugManager.flags.AddPatchInfoCommentsForAUBDump.get()) {
PatchInfoData patchInfoData(addressToPatch - buffer->getOffset(), static_cast<uint64_t>(buffer->getOffset()),
PatchInfoAllocationType::KernelArg, reinterpret_cast<uint64_t>(crossThreadData),
static_cast<uint64_t>(argAsPtr.stateless),
PatchInfoAllocationType::IndirectObjectHeap, argAsPtr.pointerSize);
this->patchInfoDataList.push_back(patchInfoData);
}
}
bool disableL3 = false;
bool forceNonAuxMode = false;
bool isAuxTranslationKernel = (AuxTranslationDirection::None != auxTranslationDirection);
auto graphicsAllocation = buffer->getGraphicsAllocation(rootDeviceIndex);
auto &hwInfo = pClDevice->getHardwareInfo();
auto &clHwHelper = ClHwHelper::get(hwInfo.platform.eRenderCoreFamily);
if (isAuxTranslationKernel) {
if (((AuxTranslationDirection::AuxToNonAux == auxTranslationDirection) && argIndex == 1) ||
((AuxTranslationDirection::NonAuxToAux == auxTranslationDirection) && argIndex == 0)) {
forceNonAuxMode = true;
}
disableL3 = (argIndex == 0);
} else if (graphicsAllocation->getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_COMPRESSED &&
clHwHelper.requiresNonAuxMode(argAsPtr)) {
forceNonAuxMode = true;
}
if (isValidOffset(argAsPtr.bindful)) {
buffer->setArgStateful(ptrOffset(getSurfaceStateHeap(), argAsPtr.bindful), forceNonAuxMode,
disableL3, isAuxTranslationKernel, arg.isReadOnly(), pClDevice->getDevice(),
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
} else if (isValidOffset(argAsPtr.bindless)) {
buffer->setArgStateful(patchBindlessSurfaceState(graphicsAllocation, argAsPtr.bindless), forceNonAuxMode,
disableL3, isAuxTranslationKernel, arg.isReadOnly(), pClDevice->getDevice(),
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
kernelArguments[argIndex].isStatelessUncacheable = argAsPtr.isPureStateful() ? false : buffer->isMemObjUncacheable();
auto allocationForCacheFlush = graphicsAllocation;
//if we make object uncacheable for surface state and there are not stateless accessess , then ther is no need to flush caches
if (buffer->isMemObjUncacheableForSurfaceState() && argAsPtr.isPureStateful()) {
allocationForCacheFlush = nullptr;
}
addAllocationToCacheFlushVector(argIndex, allocationForCacheFlush);
return CL_SUCCESS;
} else {
storeKernelArg(argIndex, BUFFER_OBJ, nullptr, argVal, argSize);
if (isValidOffset(argAsPtr.stateless)) {
auto patchLocation = ptrOffset(getCrossThreadData(), argAsPtr.stateless);
patchWithRequiredSize(patchLocation, argAsPtr.pointerSize, 0u);
}
if (isValidOffset(argAsPtr.bindful)) {
auto surfaceState = ptrOffset(getSurfaceStateHeap(), argAsPtr.bindful);
Buffer::setSurfaceState(&pClDevice->getDevice(), surfaceState, false, false, 0, nullptr, 0, nullptr, 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
return CL_SUCCESS;
}
}
cl_int Kernel::setArgPipe(uint32_t argIndex,
size_t argSize,
const void *argVal) {
if (argSize != sizeof(cl_mem *)) {
return CL_INVALID_ARG_SIZE;
}
auto clMem = reinterpret_cast<const cl_mem *>(argVal);
if (clMem && *clMem) {
auto clMemObj = *clMem;
DBG_LOG_INPUTS("setArgPipe cl_mem", clMemObj);
storeKernelArg(argIndex, PIPE_OBJ, clMemObj, argVal, argSize);
auto memObj = castToObject<MemObj>(clMemObj);
if (!memObj) {
return CL_INVALID_MEM_OBJECT;
}
auto pipe = castToObject<Pipe>(clMemObj);
if (!pipe) {
return CL_INVALID_ARG_VALUE;
}
if (memObj->getContext() != &(this->getContext())) {
return CL_INVALID_MEM_OBJECT;
}
auto rootDeviceIndex = getDevice().getRootDeviceIndex();
const auto &argAsPtr = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex].as<ArgDescPointer>();
auto patchLocation = ptrOffset(getCrossThreadData(), argAsPtr.stateless);
pipe->setPipeArg(patchLocation, argAsPtr.pointerSize, rootDeviceIndex);
if (isValidOffset(argAsPtr.bindful)) {
auto graphicsAllocation = pipe->getGraphicsAllocation(rootDeviceIndex);
auto surfaceState = ptrOffset(getSurfaceStateHeap(), argAsPtr.bindful);
Buffer::setSurfaceState(&getDevice().getDevice(), surfaceState, false, false,
pipe->getSize(), pipe->getCpuAddress(), 0,
graphicsAllocation, 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
return CL_SUCCESS;
} else {
return CL_INVALID_MEM_OBJECT;
}
}
cl_int Kernel::setArgImage(uint32_t argIndex,
size_t argSize,
const void *argVal) {
return setArgImageWithMipLevel(argIndex, argSize, argVal, 0u);
}
cl_int Kernel::setArgImageWithMipLevel(uint32_t argIndex,
size_t argSize,
const void *argVal, uint32_t mipLevel) {
auto retVal = CL_INVALID_ARG_VALUE;
auto rootDeviceIndex = getDevice().getRootDeviceIndex();
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex];
const auto &argAsImg = arg.as<ArgDescImage>();
uint32_t *crossThreadData = reinterpret_cast<uint32_t *>(this->crossThreadData);
auto clMemObj = *(static_cast<const cl_mem *>(argVal));
auto pImage = castToObject<Image>(clMemObj);
if (pImage && argSize == sizeof(cl_mem *)) {
if (pImage->peekSharingHandler()) {
usingSharedObjArgs = true;
}
DBG_LOG_INPUTS("setArgImage cl_mem", clMemObj);
storeKernelArg(argIndex, IMAGE_OBJ, clMemObj, argVal, argSize);
void *surfaceState = nullptr;
if (isValidOffset(argAsImg.bindless)) {
surfaceState = patchBindlessSurfaceState(pImage->getGraphicsAllocation(rootDeviceIndex), argAsImg.bindless);
} else {
DEBUG_BREAK_IF(isUndefinedOffset(argAsImg.bindful));
surfaceState = ptrOffset(getSurfaceStateHeap(), argAsImg.bindful);
}
// Sets SS structure
if (arg.getExtendedTypeInfo().isMediaImage) {
DEBUG_BREAK_IF(!kernelInfo.kernelDescriptor.kernelAttributes.flags.usesVme);
pImage->setMediaImageArg(surfaceState, rootDeviceIndex);
} else {
pImage->setImageArg(surfaceState, arg.getExtendedTypeInfo().isMediaBlockImage, mipLevel, rootDeviceIndex,
getKernelInfo().kernelDescriptor.kernelAttributes.flags.useGlobalAtomics);
}
auto &imageDesc = pImage->getImageDesc();
auto &imageFormat = pImage->getImageFormat();
auto graphicsAllocation = pImage->getGraphicsAllocation(rootDeviceIndex);
if (imageDesc.image_type == CL_MEM_OBJECT_IMAGE3D) {
imageTransformer->registerImage3d(argIndex);
}
patch<uint32_t, cl_uint>(imageDesc.num_samples, crossThreadData, argAsImg.metadataPayload.numSamples);
patch<uint32_t, cl_uint>(imageDesc.num_mip_levels, crossThreadData, argAsImg.metadataPayload.numMipLevels);
patch<uint32_t, uint64_t>(imageDesc.image_width, crossThreadData, argAsImg.metadataPayload.imgWidth);
patch<uint32_t, uint64_t>(imageDesc.image_height, crossThreadData, argAsImg.metadataPayload.imgHeight);
patch<uint32_t, uint64_t>(imageDesc.image_depth, crossThreadData, argAsImg.metadataPayload.imgDepth);
patch<uint32_t, uint64_t>(imageDesc.image_array_size, crossThreadData, argAsImg.metadataPayload.arraySize);
patch<uint32_t, cl_channel_type>(imageFormat.image_channel_data_type, crossThreadData, argAsImg.metadataPayload.channelDataType);
patch<uint32_t, cl_channel_order>(imageFormat.image_channel_order, crossThreadData, argAsImg.metadataPayload.channelOrder);
if (arg.getExtendedTypeInfo().hasDeviceSideEnqueueExtendedDescriptor) {
const auto &explicitArgsExtendedDescriptors = kernelInfo.kernelDescriptor.payloadMappings.explicitArgsExtendedDescriptors;
UNRECOVERABLE_IF(argIndex >= explicitArgsExtendedDescriptors.size());
auto deviceSideEnqueueDescriptor = static_cast<ArgDescriptorDeviceSideEnqueue *>(explicitArgsExtendedDescriptors[argIndex].get());
patch<uint32_t, uint32_t>(argAsImg.bindful, crossThreadData, deviceSideEnqueueDescriptor->objectId);
}
auto pixelSize = pImage->getSurfaceFormatInfo().surfaceFormat.ImageElementSizeInBytes;
patch<uint64_t, uint64_t>(graphicsAllocation->getGpuAddress(), crossThreadData, argAsImg.metadataPayload.flatBaseOffset);
patch<uint32_t, uint64_t>((imageDesc.image_width * pixelSize) - 1, crossThreadData, argAsImg.metadataPayload.flatWidth);
patch<uint32_t, uint64_t>((imageDesc.image_height * pixelSize) - 1, crossThreadData, argAsImg.metadataPayload.flatHeight);
patch<uint32_t, uint64_t>(imageDesc.image_row_pitch - 1, crossThreadData, argAsImg.metadataPayload.flatPitch);
addAllocationToCacheFlushVector(argIndex, graphicsAllocation);
retVal = CL_SUCCESS;
}
return retVal;
}
cl_int Kernel::setArgImmediate(uint32_t argIndex,
size_t argSize,
const void *argVal) {
auto retVal = CL_INVALID_ARG_VALUE;
if (argVal) {
storeKernelArg(argIndex, NONE_OBJ, nullptr, nullptr, argSize);
auto crossThreadDataEnd = ptrOffset(crossThreadData, crossThreadDataSize);
const auto &argAsVal = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex].as<ArgDescValue>();
for (const auto &element : argAsVal.elements) {
DEBUG_BREAK_IF(element.size <= 0);
auto pDst = ptrOffset(crossThreadData, element.offset);
auto pSrc = ptrOffset(argVal, element.sourceOffset);
DEBUG_BREAK_IF(!(ptrOffset(pDst, element.size) <= crossThreadDataEnd));
UNUSED_VARIABLE(crossThreadDataEnd);
if (element.sourceOffset < argSize) {
size_t maxBytesToCopy = argSize - element.sourceOffset;
size_t bytesToCopy = std::min(static_cast<size_t>(element.size), maxBytesToCopy);
memcpy_s(pDst, element.size, pSrc, bytesToCopy);
}
}
retVal = CL_SUCCESS;
}
return retVal;
}
cl_int Kernel::setArgSampler(uint32_t argIndex,
size_t argSize,
const void *argVal) {
auto retVal = CL_INVALID_SAMPLER;
if (!argVal) {
return retVal;
}
uint32_t *crossThreadData = reinterpret_cast<uint32_t *>(this->crossThreadData);
auto clSamplerObj = *(static_cast<const cl_sampler *>(argVal));
auto pSampler = castToObject<Sampler>(clSamplerObj);
if (pSampler) {
pSampler->incRefInternal();
}
if (kernelArguments.at(argIndex).object) {
auto oldSampler = castToObject<Sampler>(kernelArguments.at(argIndex).object);
UNRECOVERABLE_IF(!oldSampler);
oldSampler->decRefInternal();
}
if (pSampler && argSize == sizeof(cl_sampler *)) {
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex];
const auto &argAsSmp = arg.as<ArgDescSampler>();
storeKernelArg(argIndex, SAMPLER_OBJ, clSamplerObj, argVal, argSize);
auto dsh = getDynamicStateHeap();
auto samplerState = ptrOffset(dsh, argAsSmp.bindful);
pSampler->setArg(const_cast<void *>(samplerState), clDevice.getHardwareInfo());
patch<uint32_t, uint32_t>(pSampler->getSnapWaValue(), crossThreadData, argAsSmp.metadataPayload.samplerSnapWa);
patch<uint32_t, uint32_t>(GetAddrModeEnum(pSampler->addressingMode), crossThreadData, argAsSmp.metadataPayload.samplerAddressingMode);
patch<uint32_t, uint32_t>(GetNormCoordsEnum(pSampler->normalizedCoordinates), crossThreadData, argAsSmp.metadataPayload.samplerNormalizedCoords);
if (arg.getExtendedTypeInfo().hasDeviceSideEnqueueExtendedDescriptor) {
const auto &explicitArgsExtendedDescriptors = kernelInfo.kernelDescriptor.payloadMappings.explicitArgsExtendedDescriptors;
UNRECOVERABLE_IF(argIndex >= explicitArgsExtendedDescriptors.size());
auto deviceSideEnqueueDescriptor = static_cast<ArgDescriptorDeviceSideEnqueue *>(explicitArgsExtendedDescriptors[argIndex].get());
patch<uint32_t, uint32_t>(SAMPLER_OBJECT_ID_SHIFT + argAsSmp.bindful, crossThreadData, deviceSideEnqueueDescriptor->objectId);
}
retVal = CL_SUCCESS;
}
return retVal;
}
cl_int Kernel::setArgAccelerator(uint32_t argIndex,
size_t argSize,
const void *argVal) {
auto retVal = CL_INVALID_ARG_VALUE;
if (argSize != sizeof(cl_accelerator_intel)) {
return CL_INVALID_ARG_SIZE;
}
if (!argVal) {
return retVal;
}
auto clAcceleratorObj = *(static_cast<const cl_accelerator_intel *>(argVal));
DBG_LOG_INPUTS("setArgAccelerator cl_mem", clAcceleratorObj);
const auto pAccelerator = castToObject<IntelAccelerator>(clAcceleratorObj);
if (pAccelerator) {
storeKernelArg(argIndex, ACCELERATOR_OBJ, clAcceleratorObj, argVal, argSize);
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex];
const auto &argAsSmp = arg.as<ArgDescSampler>();
if (argAsSmp.samplerType == iOpenCL::SAMPLER_OBJECT_VME) {
const auto pVmeAccelerator = castToObjectOrAbort<VmeAccelerator>(pAccelerator);
auto pDesc = static_cast<const cl_motion_estimation_desc_intel *>(pVmeAccelerator->getDescriptor());
DEBUG_BREAK_IF(!pDesc);
if (arg.getExtendedTypeInfo().hasVmeExtendedDescriptor) {
const auto &explicitArgsExtendedDescriptors = kernelInfo.kernelDescriptor.payloadMappings.explicitArgsExtendedDescriptors;
UNRECOVERABLE_IF(argIndex >= explicitArgsExtendedDescriptors.size());
auto vmeDescriptor = static_cast<ArgDescVme *>(explicitArgsExtendedDescriptors[argIndex].get());
auto pVmeMbBlockTypeDst = reinterpret_cast<cl_uint *>(ptrOffset(crossThreadData, vmeDescriptor->mbBlockType));
*pVmeMbBlockTypeDst = pDesc->mb_block_type;
auto pVmeSubpixelMode = reinterpret_cast<cl_uint *>(ptrOffset(crossThreadData, vmeDescriptor->subpixelMode));
*pVmeSubpixelMode = pDesc->subpixel_mode;
auto pVmeSadAdjustMode = reinterpret_cast<cl_uint *>(ptrOffset(crossThreadData, vmeDescriptor->sadAdjustMode));
*pVmeSadAdjustMode = pDesc->sad_adjust_mode;
auto pVmeSearchPathType = reinterpret_cast<cl_uint *>(ptrOffset(crossThreadData, vmeDescriptor->searchPathType));
*pVmeSearchPathType = pDesc->search_path_type;
}
retVal = CL_SUCCESS;
} else if (argAsSmp.samplerType == iOpenCL::SAMPLER_OBJECT_VE) {
retVal = CL_SUCCESS;
}
}
return retVal;
}
cl_int Kernel::setArgDevQueue(uint32_t argIndex,
size_t argSize,
const void *argVal) {
if (argVal == nullptr) {
return CL_INVALID_ARG_VALUE;
}
if (argSize != sizeof(cl_command_queue)) {
return CL_INVALID_ARG_SIZE;
}
auto clDeviceQueue = *(static_cast<const device_queue *>(argVal));
auto pDeviceQueue = castToObject<DeviceQueue>(clDeviceQueue);
if (pDeviceQueue == nullptr) {
return CL_INVALID_DEVICE_QUEUE;
}
storeKernelArg(argIndex, DEVICE_QUEUE_OBJ, clDeviceQueue, argVal, argSize);
const auto &argAsPtr = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex].as<ArgDescPointer>();
auto patchLocation = ptrOffset(reinterpret_cast<uint32_t *>(crossThreadData), argAsPtr.stateless);
patchWithRequiredSize(patchLocation, argAsPtr.pointerSize,
static_cast<uintptr_t>(pDeviceQueue->getQueueBuffer()->getGpuAddressToPatch()));
return CL_SUCCESS;
}
void Kernel::setKernelArgHandler(uint32_t argIndex, KernelArgHandler handler) {
if (kernelArgHandlers.size() <= argIndex) {
kernelArgHandlers.resize(argIndex + 1);
}
kernelArgHandlers[argIndex] = handler;
}
void Kernel::unsetArg(uint32_t argIndex) {
if (kernelArguments[argIndex].isPatched) {
patchedArgumentsNum--;
kernelArguments[argIndex].isPatched = false;
if (kernelArguments[argIndex].isStatelessUncacheable) {
statelessUncacheableArgsCount--;
kernelArguments[argIndex].isStatelessUncacheable = false;
}
}
}
void Kernel::createReflectionSurface() {
auto pClDevice = &clDevice;
if (this->isParentKernel && kernelReflectionSurface == nullptr) {
auto &hwInfo = pClDevice->getHardwareInfo();
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
BlockKernelManager *blockManager = program->getBlockKernelManager();
uint32_t blockCount = static_cast<uint32_t>(blockManager->getCount());
ObjectCounts objectCount;
getParentObjectCounts(objectCount);
uint32_t parentImageCount = objectCount.imageCount;
uint32_t parentSamplerCount = objectCount.samplerCount;
size_t maxConstantBufferSize = 0;
std::vector<IGIL_KernelCurbeParams> *curbeParamsForBlocks = new std::vector<IGIL_KernelCurbeParams>[blockCount];
uint64_t *tokenMask = new uint64_t[blockCount];
uint32_t *sshTokenOffsetsFromKernelData = new uint32_t[blockCount];
size_t kernelReflectionSize = alignUp(sizeof(IGIL_KernelDataHeader) + blockCount * sizeof(IGIL_KernelAddressData), sizeof(void *));
uint32_t kernelDataOffset = static_cast<uint32_t>(kernelReflectionSize);
uint32_t parentSSHAlignedSize = alignUp(this->kernelInfo.heapInfo.SurfaceStateHeapSize, hwHelper.getBindingTableStateAlignement());
uint32_t btOffset = parentSSHAlignedSize;
for (uint32_t i = 0; i < blockCount; i++) {
const KernelInfo *pBlockInfo = blockManager->getBlockKernelInfo(i);
size_t samplerStateAndBorderColorSize = 0;
uint32_t firstSSHTokenIndex = 0;
ReflectionSurfaceHelper::getCurbeParams(curbeParamsForBlocks[i], tokenMask[i], firstSSHTokenIndex, *pBlockInfo, hwInfo);
maxConstantBufferSize = std::max(maxConstantBufferSize, static_cast<size_t>(pBlockInfo->kernelDescriptor.kernelAttributes.crossThreadDataSize));
samplerStateAndBorderColorSize = pBlockInfo->getSamplerStateArraySize(hwInfo);
samplerStateAndBorderColorSize = alignUp(samplerStateAndBorderColorSize, Sampler::samplerStateArrayAlignment);
samplerStateAndBorderColorSize += pBlockInfo->getBorderColorStateSize();
samplerStateAndBorderColorSize = alignUp(samplerStateAndBorderColorSize, sizeof(void *));
sshTokenOffsetsFromKernelData[i] = offsetof(IGIL_KernelData, m_data) + sizeof(IGIL_KernelCurbeParams) * firstSSHTokenIndex;
kernelReflectionSize += alignUp(sizeof(IGIL_KernelData) + sizeof(IGIL_KernelCurbeParams) * curbeParamsForBlocks[i].size(), sizeof(void *));
kernelReflectionSize += parentSamplerCount * sizeof(IGIL_SamplerParams) + samplerStateAndBorderColorSize;
}
maxConstantBufferSize = alignUp(maxConstantBufferSize, sizeof(void *));
kernelReflectionSize += blockCount * alignUp(maxConstantBufferSize, sizeof(void *));
kernelReflectionSize += parentImageCount * sizeof(IGIL_ImageParamters);
kernelReflectionSize += parentSamplerCount * sizeof(IGIL_ParentSamplerParams);
kernelReflectionSurface = executionEnvironment.memoryManager->allocateGraphicsMemoryWithProperties(
{pClDevice->getRootDeviceIndex(), kernelReflectionSize,
GraphicsAllocation::AllocationType::DEVICE_QUEUE_BUFFER,
pClDevice->getDeviceBitfield()});
for (uint32_t i = 0; i < blockCount; i++) {
const KernelInfo *pBlockInfo = blockManager->getBlockKernelInfo(i);
uint32_t newKernelDataOffset = ReflectionSurfaceHelper::setKernelData(kernelReflectionSurface->getUnderlyingBuffer(),
kernelDataOffset,
curbeParamsForBlocks[i],
tokenMask[i],
maxConstantBufferSize,
parentSamplerCount,
*pBlockInfo,
hwInfo);
uint32_t offset = static_cast<uint32_t>(offsetof(IGIL_KernelDataHeader, m_data) + sizeof(IGIL_KernelAddressData) * i);
uint32_t samplerHeapOffset = static_cast<uint32_t>(alignUp(kernelDataOffset + sizeof(IGIL_KernelData) + curbeParamsForBlocks[i].size() * sizeof(IGIL_KernelCurbeParams), sizeof(void *)));
uint32_t samplerHeapSize = static_cast<uint32_t>(alignUp(pBlockInfo->getSamplerStateArraySize(hwInfo), Sampler::samplerStateArrayAlignment) + pBlockInfo->getBorderColorStateSize());
uint32_t constantBufferOffset = alignUp(samplerHeapOffset + samplerHeapSize, sizeof(void *));
uint32_t samplerParamsOffset = 0;
if (parentSamplerCount) {
samplerParamsOffset = newKernelDataOffset - sizeof(IGIL_SamplerParams) * parentSamplerCount;
IGIL_SamplerParams *pSamplerParams = (IGIL_SamplerParams *)ptrOffset(kernelReflectionSurface->getUnderlyingBuffer(), samplerParamsOffset);
uint32_t sampler = 0;
const auto &args = pBlockInfo->kernelDescriptor.payloadMappings.explicitArgs;
for (uint32_t argID = 0; argID < args.size(); argID++) {
if (args[argID].is<ArgDescriptor::ArgTSampler>()) {
pSamplerParams[sampler].m_ArgID = argID;
pSamplerParams[sampler].m_SamplerStateOffset = args[argID].as<ArgDescSampler>().bindful;
sampler++;
}
}
}
ReflectionSurfaceHelper::setKernelAddressData(kernelReflectionSurface->getUnderlyingBuffer(),
offset,
kernelDataOffset,
samplerHeapOffset,
constantBufferOffset,
samplerParamsOffset,
sshTokenOffsetsFromKernelData[i] + kernelDataOffset,
btOffset,
*pBlockInfo,
hwInfo);
if (samplerHeapSize > 0) {
void *pDst = ptrOffset(kernelReflectionSurface->getUnderlyingBuffer(), samplerHeapOffset);
const void *pSrc = ptrOffset(pBlockInfo->heapInfo.pDsh, pBlockInfo->getBorderColorOffset());
memcpy_s(pDst, samplerHeapSize, pSrc, samplerHeapSize);
}
void *pDst = ptrOffset(kernelReflectionSurface->getUnderlyingBuffer(), constantBufferOffset);
const char *pSrc = pBlockInfo->crossThreadData;
memcpy_s(pDst, pBlockInfo->getConstantBufferSize(), pSrc, pBlockInfo->getConstantBufferSize());
btOffset += pBlockInfo->kernelDescriptor.payloadMappings.bindingTable.tableOffset;
kernelDataOffset = newKernelDataOffset;
}
uint32_t samplerOffset = 0;
if (parentSamplerCount) {
samplerOffset = kernelDataOffset + parentImageCount * sizeof(IGIL_ImageParamters);
}
ReflectionSurfaceHelper::setKernelDataHeader(kernelReflectionSurface->getUnderlyingBuffer(), blockCount, parentImageCount, parentSamplerCount, kernelDataOffset, samplerOffset);
delete[] curbeParamsForBlocks;
delete[] tokenMask;
delete[] sshTokenOffsetsFromKernelData;
// Patch constant values once after reflection surface creation
patchBlocksCurbeWithConstantValues();
}
if (DebugManager.flags.ForceDispatchScheduler.get()) {
if (this->isSchedulerKernel && kernelReflectionSurface == nullptr) {
kernelReflectionSurface = executionEnvironment.memoryManager->allocateGraphicsMemoryWithProperties(
{pClDevice->getRootDeviceIndex(), MemoryConstants::pageSize,
GraphicsAllocation::AllocationType::DEVICE_QUEUE_BUFFER,
pClDevice->getDeviceBitfield()});
}
}
}
void Kernel::getParentObjectCounts(ObjectCounts &objectCount) {
objectCount.imageCount = 0;
objectCount.samplerCount = 0;
DEBUG_BREAK_IF(!isParentKernel);
for (const auto &arg : this->kernelArguments) {
if (arg.type == SAMPLER_OBJ) {
objectCount.samplerCount++;
} else if (arg.type == IMAGE_OBJ) {
objectCount.imageCount++;
}
}
}
bool Kernel::hasPrintfOutput() const {
return kernelInfo.kernelDescriptor.kernelAttributes.flags.usesPrintf;
}
size_t Kernel::getInstructionHeapSizeForExecutionModel() const {
BlockKernelManager *blockManager = program->getBlockKernelManager();
uint32_t blockCount = static_cast<uint32_t>(blockManager->getCount());
size_t totalSize = 0;
if (isParentKernel) {
totalSize = kernelBinaryAlignement - 1; // for initial alignment
for (uint32_t i = 0; i < blockCount; i++) {
const KernelInfo *pBlockInfo = blockManager->getBlockKernelInfo(i);
totalSize += pBlockInfo->heapInfo.KernelHeapSize;
totalSize = alignUp(totalSize, kernelBinaryAlignement);
}
}
return totalSize;
}
void Kernel::patchBlocksCurbeWithConstantValues() {
auto rootDeviceIndex = clDevice.getRootDeviceIndex();
BlockKernelManager *blockManager = program->getBlockKernelManager();
uint32_t blockCount = static_cast<uint32_t>(blockManager->getCount());
uint64_t globalMemoryGpuAddress = program->getGlobalSurface(rootDeviceIndex) != nullptr ? program->getGlobalSurface(rootDeviceIndex)->getGpuAddressToPatch() : 0;
uint64_t constantMemoryGpuAddress = program->getConstantSurface(rootDeviceIndex) != nullptr ? program->getConstantSurface(rootDeviceIndex)->getGpuAddressToPatch() : 0;
for (uint32_t blockID = 0; blockID < blockCount; blockID++) {
const KernelInfo *pBlockInfo = blockManager->getBlockKernelInfo(blockID);
uint64_t globalMemoryCurbeOffset = ReflectionSurfaceHelper::undefinedOffset;
uint32_t globalMemoryPatchSize = 0;
uint64_t constantMemoryCurbeOffset = ReflectionSurfaceHelper::undefinedOffset;
uint32_t constantMemoryPatchSize = 0;
if (isValidOffset(pBlockInfo->kernelDescriptor.payloadMappings.implicitArgs.globalVariablesSurfaceAddress.stateless)) {
globalMemoryCurbeOffset = pBlockInfo->kernelDescriptor.payloadMappings.implicitArgs.globalVariablesSurfaceAddress.stateless;
globalMemoryPatchSize = pBlockInfo->kernelDescriptor.payloadMappings.implicitArgs.globalVariablesSurfaceAddress.pointerSize;
}
if (isValidOffset(pBlockInfo->kernelDescriptor.payloadMappings.implicitArgs.globalConstantsSurfaceAddress.stateless)) {
constantMemoryCurbeOffset = pBlockInfo->kernelDescriptor.payloadMappings.implicitArgs.globalConstantsSurfaceAddress.stateless;
constantMemoryPatchSize = pBlockInfo->kernelDescriptor.payloadMappings.implicitArgs.globalConstantsSurfaceAddress.pointerSize;
}
ReflectionSurfaceHelper::patchBlocksCurbeWithConstantValues(kernelReflectionSurface->getUnderlyingBuffer(), blockID,
globalMemoryCurbeOffset, globalMemoryPatchSize, globalMemoryGpuAddress,
constantMemoryCurbeOffset, constantMemoryPatchSize, constantMemoryGpuAddress,
ReflectionSurfaceHelper::undefinedOffset, 0, 0);
}
}
void Kernel::ReflectionSurfaceHelper::getCurbeParams(std::vector<IGIL_KernelCurbeParams> &curbeParamsOut, uint64_t &tokenMaskOut, uint32_t &firstSSHTokenIndex, const KernelInfo &kernelInfo, const HardwareInfo &hwInfo) {
const auto &args = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs;
const auto gpuPointerSize = kernelInfo.kernelDescriptor.kernelAttributes.gpuPointerSize;
uint32_t bindingTableIndex = 253;
uint64_t tokenMask = 0;
for (size_t argNum = 0; argNum < args.size(); argNum++) {
const auto &arg = args[argNum];
auto sizeOfKernelArgForSSH = gpuPointerSize;
bindingTableIndex = 253;
if (arg.is<ArgDescriptor::ArgTPointer>()) {
const auto &argAsPtr = arg.as<ArgDescPointer>();
if (argAsPtr.requiredSlmAlignment) {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{DATA_PARAMETER_SUM_OF_LOCAL_MEMORY_OBJECT_ARGUMENT_SIZES, 0, argAsPtr.slmOffset, argAsPtr.requiredSlmAlignment});
tokenMask |= shiftLeftBy(DATA_PARAMETER_SUM_OF_LOCAL_MEMORY_OBJECT_ARGUMENT_SIZES);
} else {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{COMPILER_DATA_PARAMETER_GLOBAL_SURFACE, gpuPointerSize, argAsPtr.stateless, static_cast<uint>(argNum)});
tokenMask |= shiftLeftBy(63);
}
} else if (arg.is<ArgDescriptor::ArgTImage>()) {
const auto &argAsImg = arg.as<ArgDescImage>();
auto emplaceIfValidOffset = [&](uint parameterType, NEO::CrossThreadDataOffset offset) {
if (isValidOffset(offset)) {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{parameterType + 50, sizeof(uint32_t), offset, static_cast<uint>(argNum)});
}
};
emplaceIfValidOffset(DATA_PARAMETER_IMAGE_WIDTH, argAsImg.metadataPayload.imgWidth);
emplaceIfValidOffset(DATA_PARAMETER_IMAGE_HEIGHT, argAsImg.metadataPayload.imgHeight);
emplaceIfValidOffset(DATA_PARAMETER_IMAGE_DEPTH, argAsImg.metadataPayload.imgDepth);
emplaceIfValidOffset(DATA_PARAMETER_IMAGE_CHANNEL_DATA_TYPE, argAsImg.metadataPayload.channelDataType);
emplaceIfValidOffset(DATA_PARAMETER_IMAGE_CHANNEL_ORDER, argAsImg.metadataPayload.channelOrder);
emplaceIfValidOffset(DATA_PARAMETER_IMAGE_ARRAY_SIZE, argAsImg.metadataPayload.arraySize);
if (arg.getExtendedTypeInfo().hasDeviceSideEnqueueExtendedDescriptor) {
const auto &argsExtDescriptors = kernelInfo.kernelDescriptor.payloadMappings.explicitArgsExtendedDescriptors;
UNRECOVERABLE_IF(argNum >= argsExtDescriptors.size());
const auto &deviceSideEnqueueDescriptor = static_cast<ArgDescriptorDeviceSideEnqueue *>(argsExtDescriptors[argNum].get());
emplaceIfValidOffset(DATA_PARAMETER_OBJECT_ID, deviceSideEnqueueDescriptor->objectId);
}
const auto &bindingTable = kernelInfo.kernelDescriptor.payloadMappings.bindingTable;
if (isValidOffset(bindingTable.tableOffset)) {
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
const auto ssh = static_cast<const char *>(kernelInfo.heapInfo.pSsh) + bindingTable.tableOffset;
for (uint8_t i = 0; i < bindingTable.numEntries; i++) {
const auto pointer = static_cast<NEO::SurfaceStateHeapOffset>(hwHelper.getBindingTableStateSurfaceStatePointer(ssh, i));
if (pointer == argAsImg.bindful) {
bindingTableIndex = i;
break;
}
}
DEBUG_BREAK_IF(bindingTableIndex == 253);
}
tokenMask |= shiftLeftBy(50);
} else if (arg.is<ArgDescriptor::ArgTSampler>()) {
const auto &argAsSmp = arg.as<ArgDescSampler>();
auto emplaceIfValidOffset = [&](uint parameterType, NEO::CrossThreadDataOffset offset) {
if (isValidOffset(offset)) {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{parameterType + 100, sizeof(uint32_t), offset, static_cast<uint>(argNum)});
}
};
emplaceIfValidOffset(DATA_PARAMETER_SAMPLER_COORDINATE_SNAP_WA_REQUIRED, argAsSmp.metadataPayload.samplerSnapWa);
emplaceIfValidOffset(DATA_PARAMETER_SAMPLER_ADDRESS_MODE, argAsSmp.metadataPayload.samplerAddressingMode);
emplaceIfValidOffset(DATA_PARAMETER_SAMPLER_NORMALIZED_COORDS, argAsSmp.metadataPayload.samplerNormalizedCoords);
if (arg.getExtendedTypeInfo().hasDeviceSideEnqueueExtendedDescriptor) {
const auto &argsExtDescriptors = kernelInfo.kernelDescriptor.payloadMappings.explicitArgsExtendedDescriptors;
UNRECOVERABLE_IF(argNum >= argsExtDescriptors.size());
const auto &deviceSideEnqueueDescriptor = static_cast<ArgDescriptorDeviceSideEnqueue *>(argsExtDescriptors[argNum].get());
emplaceIfValidOffset(DATA_PARAMETER_OBJECT_ID, deviceSideEnqueueDescriptor->objectId);
}
tokenMask |= shiftLeftBy(51);
} else {
bindingTableIndex = 0;
sizeOfKernelArgForSSH = 0;
}
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{1024, sizeOfKernelArgForSSH, bindingTableIndex, static_cast<uint>(argNum)});
}
for (const auto &param : kernelInfo.kernelDescriptor.kernelMetadata.allByValueKernelArguments) {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{DATA_PARAMETER_KERNEL_ARGUMENT, param.byValueElement.size, param.byValueElement.offset, param.argNum});
tokenMask |= shiftLeftBy(DATA_PARAMETER_KERNEL_ARGUMENT);
}
const auto &dispatchTraits = kernelInfo.kernelDescriptor.payloadMappings.dispatchTraits;
for (uint32_t i = 0; i < 3U; i++) {
auto emplaceIfValidOffsetAndSetTokenMask = [&](uint parameterType, NEO::CrossThreadDataOffset offset) {
constexpr uint paramSize = sizeof(uint32_t);
if (isValidOffset(offset)) {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{parameterType, paramSize, offset, static_cast<uint>(i * paramSize)});
tokenMask |= shiftLeftBy(parameterType);
}
};
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_LOCAL_WORK_SIZE, dispatchTraits.localWorkSize[i]);
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_LOCAL_WORK_SIZE, dispatchTraits.localWorkSize2[i]);
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_GLOBAL_WORK_OFFSET, dispatchTraits.globalWorkOffset[i]);
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_ENQUEUED_LOCAL_WORK_SIZE, dispatchTraits.enqueuedLocalWorkSize[i]);
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_GLOBAL_WORK_SIZE, dispatchTraits.globalWorkSize[i]);
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_NUM_WORK_GROUPS, dispatchTraits.numWorkGroups[i]);
}
{
const auto &payloadMappings = kernelInfo.kernelDescriptor.payloadMappings;
auto emplaceIfValidOffsetAndSetTokenMask = [&](uint parameterType, NEO::CrossThreadDataOffset offset) {
if (isValidOffset(offset)) {
curbeParamsOut.emplace_back(IGIL_KernelCurbeParams{parameterType, sizeof(uint32_t), offset, 0});
tokenMask |= shiftLeftBy(parameterType);
}
};
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_PARENT_EVENT, payloadMappings.implicitArgs.deviceSideEnqueueParentEvent);
emplaceIfValidOffsetAndSetTokenMask(DATA_PARAMETER_WORK_DIMENSIONS, payloadMappings.dispatchTraits.workDim);
}
std::sort(curbeParamsOut.begin(), curbeParamsOut.end(), compareFunction);
tokenMaskOut = tokenMask;
firstSSHTokenIndex = static_cast<uint32_t>(curbeParamsOut.size() - args.size());
}
uint32_t Kernel::ReflectionSurfaceHelper::setKernelData(void *reflectionSurface, uint32_t offset,
std::vector<IGIL_KernelCurbeParams> &curbeParamsIn, uint64_t tokenMaskIn,
size_t maxConstantBufferSize, size_t samplerCount, const KernelInfo &kernelInfo, const HardwareInfo &hwInfo) {
uint32_t offsetToEnd = 0;
IGIL_KernelData *kernelData = reinterpret_cast<IGIL_KernelData *>(ptrOffset(reflectionSurface, offset));
size_t samplerHeapSize = alignUp(kernelInfo.getSamplerStateArraySize(hwInfo), Sampler::samplerStateArrayAlignment) + kernelInfo.getBorderColorStateSize();
kernelData->m_numberOfCurbeParams = static_cast<uint32_t>(curbeParamsIn.size()); // number of paramters to patch
kernelData->m_numberOfCurbeTokens = static_cast<uint32_t>(curbeParamsIn.size() - kernelInfo.kernelDescriptor.payloadMappings.explicitArgs.size());
kernelData->m_numberOfSamplerStates = static_cast<uint32_t>(kernelInfo.getSamplerStateArrayCount());
kernelData->m_SizeOfSamplerHeap = static_cast<uint32_t>(samplerHeapSize);
kernelData->m_SamplerBorderColorStateOffsetOnDSH = isValidOffset(kernelInfo.kernelDescriptor.payloadMappings.samplerTable.borderColor) ? kernelInfo.kernelDescriptor.payloadMappings.samplerTable.borderColor : 0;
kernelData->m_SamplerStateArrayOffsetOnDSH = isValidOffset(kernelInfo.kernelDescriptor.payloadMappings.samplerTable.tableOffset) ? kernelInfo.kernelDescriptor.payloadMappings.samplerTable.tableOffset : -1;
kernelData->m_sizeOfConstantBuffer = kernelInfo.getConstantBufferSize();
kernelData->m_PatchTokensMask = tokenMaskIn;
kernelData->m_ScratchSpacePatchValue = 0;
kernelData->m_SIMDSize = kernelInfo.getMaxSimdSize();
kernelData->m_HasBarriers = kernelInfo.kernelDescriptor.kernelAttributes.barrierCount;
kernelData->m_RequiredWkgSizes[0] = kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[0];
kernelData->m_RequiredWkgSizes[1] = kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[1];
kernelData->m_RequiredWkgSizes[2] = kernelInfo.kernelDescriptor.kernelAttributes.requiredWorkgroupSize[2];
kernelData->m_InilineSLMSize = kernelInfo.kernelDescriptor.kernelAttributes.slmInlineSize;
bool localIdRequired = false;
if (kernelInfo.kernelDescriptor.kernelAttributes.flags.usesFlattenedLocalIds || (kernelInfo.kernelDescriptor.kernelAttributes.numLocalIdChannels > 0)) {
localIdRequired = true;
}
kernelData->m_PayloadSize = PerThreadDataHelper::getThreadPayloadSize(kernelInfo.kernelDescriptor, hwInfo.capabilityTable.grfSize);
kernelData->m_NeedLocalIDS = localIdRequired ? 1 : 0;
kernelData->m_DisablePreemption = 0u;
bool concurrentExecAllowed = true;
if (kernelInfo.kernelDescriptor.kernelAttributes.perHwThreadPrivateMemorySize > 0) {
concurrentExecAllowed = false;
}
kernelData->m_CanRunConcurently = concurrentExecAllowed ? 1 : 0;
if (DebugManager.flags.DisableConcurrentBlockExecution.get()) {
kernelData->m_CanRunConcurently = false;
}
IGIL_KernelCurbeParams *kernelCurbeParams = kernelData->m_data;
for (uint32_t i = 0; i < curbeParamsIn.size(); i++) {
kernelCurbeParams[i] = curbeParamsIn[i];
}
offsetToEnd = static_cast<uint32_t>(offset +
alignUp(sizeof(IGIL_KernelData) + sizeof(IGIL_KernelCurbeParams) * curbeParamsIn.size(), sizeof(void *)) +
alignUp(samplerHeapSize, sizeof(void *)) +
alignUp(maxConstantBufferSize, sizeof(void *)) +
sizeof(IGIL_SamplerParams) * samplerCount);
return offsetToEnd;
}
void Kernel::ReflectionSurfaceHelper::setKernelAddressDataBtOffset(void *reflectionSurface, uint32_t blockID, uint32_t btOffset) {
uint32_t offset = static_cast<uint32_t>(offsetof(IGIL_KernelDataHeader, m_data) + sizeof(IGIL_KernelAddressData) * blockID);
IGIL_KernelAddressData *kernelAddressData = reinterpret_cast<IGIL_KernelAddressData *>(ptrOffset(reflectionSurface, offset));
kernelAddressData->m_BTSoffset = btOffset;
}
void Kernel::ReflectionSurfaceHelper::setKernelAddressData(void *reflectionSurface, uint32_t offset, uint32_t kernelDataOffset, uint32_t samplerHeapOffset,
uint32_t constantBufferOffset, uint32_t samplerParamsOffset,
uint32_t sshTokensOffset, uint32_t btOffset, const KernelInfo &kernelInfo, const HardwareInfo &hwInfo) {
IGIL_KernelAddressData *kernelAddressData = reinterpret_cast<IGIL_KernelAddressData *>(ptrOffset(reflectionSurface, offset));
auto &hwHelper = HwHelper::get(hwInfo.platform.eRenderCoreFamily);
kernelAddressData->m_KernelDataOffset = kernelDataOffset;
kernelAddressData->m_SamplerHeapOffset = samplerHeapOffset;
kernelAddressData->m_SamplerParamsOffset = samplerParamsOffset;
kernelAddressData->m_ConstantBufferOffset = constantBufferOffset;
kernelAddressData->m_SSHTokensOffset = sshTokensOffset;
kernelAddressData->m_BTSoffset = btOffset;
kernelAddressData->m_BTSize = static_cast<uint32_t>(kernelInfo.kernelDescriptor.payloadMappings.bindingTable.numEntries * hwHelper.getBindingTableStateSize());
}
template <>
void Kernel::ReflectionSurfaceHelper::patchBlocksCurbe<false>(void *reflectionSurface, uint32_t blockID,
uint64_t defaultDeviceQueueCurbeOffset, uint32_t patchSizeDefaultQueue, uint64_t defaultDeviceQueueGpuAddress,
uint64_t eventPoolCurbeOffset, uint32_t patchSizeEventPool, uint64_t eventPoolGpuAddress,
uint64_t deviceQueueCurbeOffset, uint32_t patchSizeDeviceQueue, uint64_t deviceQueueGpuAddress,
uint64_t printfBufferOffset, uint32_t patchSizePrintfBuffer, uint64_t printfBufferGpuAddress,
uint64_t privateSurfaceOffset, uint32_t privateSurfaceSize, uint64_t privateSurfaceGpuAddress) {
IGIL_KernelDataHeader *pKernelHeader = reinterpret_cast<IGIL_KernelDataHeader *>(reflectionSurface);
// Reflection surface must be initialized prior to patching blocks curbe on KRS
DEBUG_BREAK_IF(blockID >= pKernelHeader->m_numberOfKernels);
IGIL_KernelAddressData *addressData = pKernelHeader->m_data;
// const buffer offsets must be set
DEBUG_BREAK_IF(addressData[blockID].m_ConstantBufferOffset == 0);
void *pCurbe = ptrOffset(reflectionSurface, addressData[blockID].m_ConstantBufferOffset);
if (defaultDeviceQueueCurbeOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)defaultDeviceQueueCurbeOffset);
patchWithRequiredSize(patchedPointer, patchSizeDefaultQueue, (uintptr_t)defaultDeviceQueueGpuAddress);
}
if (eventPoolCurbeOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)eventPoolCurbeOffset);
patchWithRequiredSize(patchedPointer, patchSizeEventPool, (uintptr_t)eventPoolGpuAddress);
}
if (deviceQueueCurbeOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)deviceQueueCurbeOffset);
patchWithRequiredSize(patchedPointer, patchSizeDeviceQueue, (uintptr_t)deviceQueueGpuAddress);
}
if (printfBufferOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)printfBufferOffset);
patchWithRequiredSize(patchedPointer, patchSizePrintfBuffer, (uintptr_t)printfBufferGpuAddress);
}
if (privateSurfaceOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)privateSurfaceOffset);
patchWithRequiredSize(patchedPointer, privateSurfaceSize, (uintptr_t)privateSurfaceGpuAddress);
}
}
void Kernel::ReflectionSurfaceHelper::patchBlocksCurbeWithConstantValues(void *reflectionSurface, uint32_t blockID,
uint64_t globalMemoryCurbeOffset, uint32_t globalMemoryPatchSize, uint64_t globalMemoryGpuAddress,
uint64_t constantMemoryCurbeOffset, uint32_t constantMemoryPatchSize, uint64_t constantMemoryGpuAddress,
uint64_t privateMemoryCurbeOffset, uint32_t privateMemoryPatchSize, uint64_t privateMemoryGpuAddress) {
IGIL_KernelDataHeader *pKernelHeader = reinterpret_cast<IGIL_KernelDataHeader *>(reflectionSurface);
// Reflection surface must be initialized prior to patching blocks curbe on KRS
DEBUG_BREAK_IF(blockID >= pKernelHeader->m_numberOfKernels);
IGIL_KernelAddressData *addressData = pKernelHeader->m_data;
// const buffer offsets must be set
DEBUG_BREAK_IF(addressData[blockID].m_ConstantBufferOffset == 0);
void *pCurbe = ptrOffset(reflectionSurface, addressData[blockID].m_ConstantBufferOffset);
if (globalMemoryCurbeOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)globalMemoryCurbeOffset);
patchWithRequiredSize(patchedPointer, globalMemoryPatchSize, (uintptr_t)globalMemoryGpuAddress);
}
if (constantMemoryCurbeOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)constantMemoryCurbeOffset);
patchWithRequiredSize(patchedPointer, constantMemoryPatchSize, (uintptr_t)constantMemoryGpuAddress);
}
if (privateMemoryCurbeOffset != undefinedOffset) {
auto *patchedPointer = ptrOffset(pCurbe, (size_t)privateMemoryCurbeOffset);
patchWithRequiredSize(patchedPointer, privateMemoryPatchSize, (uintptr_t)privateMemoryGpuAddress);
}
}
void Kernel::ReflectionSurfaceHelper::setParentImageParams(void *reflectionSurface, std::vector<Kernel::SimpleKernelArgInfo> &parentArguments, const KernelInfo &parentKernelInfo) {
IGIL_KernelDataHeader *pKernelHeader = reinterpret_cast<IGIL_KernelDataHeader *>(reflectionSurface);
IGIL_ImageParamters *pImageParameters = reinterpret_cast<IGIL_ImageParamters *>(ptrOffset(pKernelHeader, (size_t)pKernelHeader->m_ParentImageDataOffset));
uint32_t numArgs = (uint32_t)parentArguments.size();
for (uint32_t i = 0; i < numArgs; i++) {
if (parentArguments[i].type == Kernel::kernelArgType::IMAGE_OBJ) {
const Image *image = castToObject<Image>((cl_mem)parentArguments[i].object);
if (image) {
pImageParameters->m_ArraySize = (uint32_t)image->getImageDesc().image_array_size;
pImageParameters->m_Depth = (uint32_t)image->getImageDesc().image_depth;
pImageParameters->m_Height = (uint32_t)image->getImageDesc().image_height;
pImageParameters->m_Width = (uint32_t)image->getImageDesc().image_width;
pImageParameters->m_NumMipLevels = (uint32_t)image->getImageDesc().num_mip_levels;
pImageParameters->m_NumSamples = (uint32_t)image->getImageDesc().num_samples;
pImageParameters->m_ChannelDataType = (uint32_t)image->getImageFormat().image_channel_data_type;
pImageParameters->m_ChannelOrder = (uint32_t)image->getImageFormat().image_channel_data_type;
pImageParameters->m_ObjectID = (uint32_t)parentKernelInfo.kernelDescriptor.payloadMappings.explicitArgs[i].as<ArgDescImage>().bindful;
pImageParameters++;
}
}
}
}
void Kernel::ReflectionSurfaceHelper::setParentSamplerParams(void *reflectionSurface, std::vector<Kernel::SimpleKernelArgInfo> &parentArguments, const KernelInfo &parentKernelInfo) {
IGIL_KernelDataHeader *pKernelHeader = reinterpret_cast<IGIL_KernelDataHeader *>(reflectionSurface);
IGIL_ParentSamplerParams *pParentSamplerParams = reinterpret_cast<IGIL_ParentSamplerParams *>(ptrOffset(pKernelHeader, (size_t)pKernelHeader->m_ParentSamplerParamsOffset));
uint32_t numArgs = (uint32_t)parentArguments.size();
for (uint32_t i = 0; i < numArgs; i++) {
if (parentArguments[i].type == Kernel::kernelArgType::SAMPLER_OBJ) {
const Sampler *sampler = castToObject<Sampler>((cl_sampler)parentArguments[i].object);
if (sampler) {
pParentSamplerParams->CoordinateSnapRequired = (uint32_t)sampler->getSnapWaValue();
pParentSamplerParams->m_AddressingMode = (uint32_t)sampler->addressingMode;
pParentSamplerParams->NormalizedCoords = (uint32_t)sampler->normalizedCoordinates;
pParentSamplerParams->m_ObjectID = OCLRT_ARG_OFFSET_TO_SAMPLER_OBJECT_ID((uint32_t)parentKernelInfo.kernelDescriptor.payloadMappings.explicitArgs[i].as<ArgDescSampler>().bindful);
pParentSamplerParams++;
}
}
}
}
void Kernel::resetSharedObjectsPatchAddresses() {
for (size_t i = 0; i < getKernelArgsNumber(); i++) {
auto clMem = (cl_mem)kernelArguments[i].object;
auto memObj = castToObject<MemObj>(clMem);
if (memObj && memObj->peekSharingHandler()) {
setArg((uint32_t)i, sizeof(cl_mem), &clMem);
}
}
}
void Kernel::provideInitializationHints() {
Context *context = program->getContextPtr();
if (context == nullptr || !context->isProvidingPerformanceHints())
return;
auto pClDevice = &getDevice();
if (privateSurfaceSize) {
context->providePerformanceHint(CL_CONTEXT_DIAGNOSTICS_LEVEL_BAD_INTEL, PRIVATE_MEMORY_USAGE_TOO_HIGH,
kernelInfo.kernelDescriptor.kernelMetadata.kernelName.c_str(),
privateSurfaceSize);
}
auto scratchSize = kernelInfo.kernelDescriptor.kernelAttributes.perThreadScratchSize[0] *
pClDevice->getSharedDeviceInfo().computeUnitsUsedForScratch * kernelInfo.getMaxSimdSize();
if (scratchSize > 0) {
context->providePerformanceHint(CL_CONTEXT_DIAGNOSTICS_LEVEL_BAD_INTEL, REGISTER_PRESSURE_TOO_HIGH,
kernelInfo.kernelDescriptor.kernelMetadata.kernelName.c_str(), scratchSize);
}
}
void Kernel::patchDefaultDeviceQueue(DeviceQueue *devQueue) {
const auto &defaultQueueSurfaceAddress = kernelInfo.kernelDescriptor.payloadMappings.implicitArgs.deviceSideEnqueueDefaultQueueSurfaceAddress;
if (isValidOffset(defaultQueueSurfaceAddress.stateless) && crossThreadData) {
auto patchLocation = ptrOffset(reinterpret_cast<uint32_t *>(crossThreadData), defaultQueueSurfaceAddress.stateless);
patchWithRequiredSize(patchLocation, defaultQueueSurfaceAddress.pointerSize,
static_cast<uintptr_t>(devQueue->getQueueBuffer()->getGpuAddressToPatch()));
}
if (isValidOffset(defaultQueueSurfaceAddress.bindful)) {
auto surfaceState = ptrOffset(reinterpret_cast<uintptr_t *>(getSurfaceStateHeap()), defaultQueueSurfaceAddress.bindful);
Buffer::setSurfaceState(&devQueue->getDevice(), surfaceState, false, false, devQueue->getQueueBuffer()->getUnderlyingBufferSize(),
(void *)devQueue->getQueueBuffer()->getGpuAddress(), 0, devQueue->getQueueBuffer(), 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
}
void Kernel::patchEventPool(DeviceQueue *devQueue) {
const auto &eventPoolSurfaceAddress = kernelInfo.kernelDescriptor.payloadMappings.implicitArgs.deviceSideEnqueueEventPoolSurfaceAddress;
if (isValidOffset(eventPoolSurfaceAddress.stateless) && crossThreadData) {
auto patchLocation = ptrOffset(reinterpret_cast<uint32_t *>(crossThreadData), eventPoolSurfaceAddress.stateless);
patchWithRequiredSize(patchLocation, eventPoolSurfaceAddress.pointerSize,
static_cast<uintptr_t>(devQueue->getEventPoolBuffer()->getGpuAddressToPatch()));
}
if (isValidOffset(eventPoolSurfaceAddress.bindful)) {
auto surfaceState = ptrOffset(reinterpret_cast<uintptr_t *>(getSurfaceStateHeap()), eventPoolSurfaceAddress.bindful);
auto eventPoolBuffer = devQueue->getEventPoolBuffer();
Buffer::setSurfaceState(&devQueue->getDevice(), surfaceState, false, false, eventPoolBuffer->getUnderlyingBufferSize(),
(void *)eventPoolBuffer->getGpuAddress(), 0, eventPoolBuffer, 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
}
void Kernel::patchBlocksSimdSize() {
BlockKernelManager *blockManager = program->getBlockKernelManager();
for (auto &idOffset : kernelInfo.childrenKernelsIdOffset) {
DEBUG_BREAK_IF(!(idOffset.first < static_cast<uint32_t>(blockManager->getCount())));
const KernelInfo *blockInfo = blockManager->getBlockKernelInfo(idOffset.first);
uint32_t *simdSize = reinterpret_cast<uint32_t *>(&crossThreadData[idOffset.second]);
*simdSize = blockInfo->getMaxSimdSize();
}
}
bool Kernel::usesSyncBuffer() {
return kernelInfo.kernelDescriptor.kernelAttributes.flags.usesSyncBuffer;
}
void Kernel::patchSyncBuffer(GraphicsAllocation *gfxAllocation, size_t bufferOffset) {
const auto &syncBuffer = kernelInfo.kernelDescriptor.payloadMappings.implicitArgs.syncBufferAddress;
auto bufferPatchAddress = ptrOffset(crossThreadData, syncBuffer.stateless);
patchWithRequiredSize(bufferPatchAddress, syncBuffer.pointerSize,
ptrOffset(gfxAllocation->getGpuAddressToPatch(), bufferOffset));
if (isValidOffset(syncBuffer.bindful)) {
auto surfaceState = ptrOffset(reinterpret_cast<uintptr_t *>(getSurfaceStateHeap()), syncBuffer.bindful);
auto addressToPatch = gfxAllocation->getUnderlyingBuffer();
auto sizeToPatch = gfxAllocation->getUnderlyingBufferSize();
Buffer::setSurfaceState(&clDevice.getDevice(), surfaceState, false, false, sizeToPatch, addressToPatch, 0, gfxAllocation, 0, 0,
kernelInfo.kernelDescriptor.kernelAttributes.flags.useGlobalAtomics, areMultipleSubDevicesInContext());
}
}
template void Kernel::patchReflectionSurface<false>(DeviceQueue *, PrintfHandler *);
bool Kernel::isPatched() const {
return patchedArgumentsNum == kernelInfo.kernelDescriptor.kernelAttributes.numArgsToPatch;
}
cl_int Kernel::checkCorrectImageAccessQualifier(cl_uint argIndex,
size_t argSize,
const void *argValue) const {
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[argIndex];
if (arg.is<ArgDescriptor::ArgTImage>()) {
cl_mem mem = *(static_cast<const cl_mem *>(argValue));
MemObj *pMemObj = nullptr;
WithCastToInternal(mem, &pMemObj);
if (pMemObj) {
auto accessQualifier = arg.getTraits().accessQualifier;
cl_mem_flags flags = pMemObj->getFlags();
if ((accessQualifier == KernelArgMetadata::AccessReadOnly && ((flags | CL_MEM_WRITE_ONLY) == flags)) ||
(accessQualifier == KernelArgMetadata::AccessWriteOnly && ((flags | CL_MEM_READ_ONLY) == flags))) {
return CL_INVALID_ARG_VALUE;
}
} else {
return CL_INVALID_ARG_VALUE;
}
}
return CL_SUCCESS;
}
void Kernel::resolveArgs() {
if (!Kernel::isPatched() || !imageTransformer->hasRegisteredImages3d() || !canTransformImages())
return;
bool canTransformImageTo2dArray = true;
const auto &args = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs;
for (uint32_t i = 0; i < patchedArgumentsNum; i++) {
if (args[i].is<ArgDescriptor::ArgTSampler>()) {
auto sampler = castToObject<Sampler>(kernelArguments.at(i).object);
if (sampler->isTransformable()) {
canTransformImageTo2dArray = true;
} else {
canTransformImageTo2dArray = false;
break;
}
}
}
if (canTransformImageTo2dArray) {
imageTransformer->transformImagesTo2dArray(kernelInfo, kernelArguments, getSurfaceStateHeap());
} else if (imageTransformer->didTransform()) {
imageTransformer->transformImagesTo3d(kernelInfo, kernelArguments, getSurfaceStateHeap());
}
}
bool Kernel::canTransformImages() const {
auto renderCoreFamily = clDevice.getHardwareInfo().platform.eRenderCoreFamily;
return renderCoreFamily >= IGFX_GEN9_CORE && renderCoreFamily <= IGFX_GEN11LP_CORE && !isBuiltIn;
}
void Kernel::fillWithKernelObjsForAuxTranslation(KernelObjsForAuxTranslation &kernelObjsForAuxTranslation) {
kernelObjsForAuxTranslation.reserve(getKernelArgsNumber());
for (uint32_t i = 0; i < getKernelArgsNumber(); i++) {
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[i];
if (BUFFER_OBJ == kernelArguments.at(i).type && !arg.as<ArgDescPointer>().isPureStateful()) {
auto buffer = castToObject<Buffer>(getKernelArg(i));
if (buffer && buffer->getMultiGraphicsAllocation().getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_COMPRESSED) {
kernelObjsForAuxTranslation.insert({KernelObjForAuxTranslation::Type::MEM_OBJ, buffer});
auto &context = this->program->getContext();
if (context.isProvidingPerformanceHints()) {
const auto &argExtMeta = kernelInfo.kernelDescriptor.explicitArgsExtendedMetadata[i];
context.providePerformanceHint(CL_CONTEXT_DIAGNOSTICS_LEVEL_BAD_INTEL, KERNEL_ARGUMENT_AUX_TRANSLATION,
kernelInfo.kernelDescriptor.kernelMetadata.kernelName.c_str(), i, argExtMeta.argName.c_str());
}
}
}
if (SVM_ALLOC_OBJ == getKernelArguments().at(i).type && !arg.as<ArgDescPointer>().isPureStateful()) {
auto svmAlloc = reinterpret_cast<GraphicsAllocation *>(const_cast<void *>(getKernelArg(i)));
if (svmAlloc && svmAlloc->getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_COMPRESSED) {
kernelObjsForAuxTranslation.insert({KernelObjForAuxTranslation::Type::GFX_ALLOC, svmAlloc});
auto &context = this->program->getContext();
if (context.isProvidingPerformanceHints()) {
const auto &argExtMeta = kernelInfo.kernelDescriptor.explicitArgsExtendedMetadata[i];
context.providePerformanceHint(CL_CONTEXT_DIAGNOSTICS_LEVEL_BAD_INTEL, KERNEL_ARGUMENT_AUX_TRANSLATION,
kernelInfo.kernelDescriptor.kernelMetadata.kernelName.c_str(), i, argExtMeta.argName.c_str());
}
}
}
}
}
bool Kernel::hasDirectStatelessAccessToHostMemory() const {
for (uint32_t i = 0; i < getKernelArgsNumber(); i++) {
const auto &arg = kernelInfo.kernelDescriptor.payloadMappings.explicitArgs[i];
if (BUFFER_OBJ == kernelArguments.at(i).type && !arg.as<ArgDescPointer>().isPureStateful()) {
auto buffer = castToObject<Buffer>(getKernelArg(i));
if (buffer && buffer->getMultiGraphicsAllocation().getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_HOST_MEMORY) {
return true;
}
}
if (SVM_ALLOC_OBJ == kernelArguments.at(i).type && !arg.as<ArgDescPointer>().isPureStateful()) {
auto svmAlloc = reinterpret_cast<const GraphicsAllocation *>(getKernelArg(i));
if (svmAlloc && svmAlloc->getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_HOST_MEMORY) {
return true;
}
}
}
return false;
}
bool Kernel::hasIndirectStatelessAccessToHostMemory() const {
if (!kernelInfo.hasIndirectStatelessAccess) {
return false;
}
for (auto gfxAllocation : kernelUnifiedMemoryGfxAllocations) {
if (gfxAllocation->getAllocationType() == GraphicsAllocation::AllocationType::BUFFER_HOST_MEMORY) {
return true;
}
}
if (unifiedMemoryControls.indirectHostAllocationsAllowed) {
return getContext().getSVMAllocsManager()->hasHostAllocations();
}
return false;
}
void Kernel::getAllocationsForCacheFlush(CacheFlushAllocationsVec &out) const {
if (false == HwHelper::cacheFlushAfterWalkerSupported(getHardwareInfo())) {
return;
}
for (GraphicsAllocation *alloc : this->kernelArgRequiresCacheFlush) {
if (nullptr == alloc) {
continue;
}
out.push_back(alloc);
}
auto rootDeviceIndex = getDevice().getRootDeviceIndex();
auto global = getProgram()->getGlobalSurface(rootDeviceIndex);
if (global != nullptr) {
out.push_back(global);
}
if (svmAllocationsRequireCacheFlush) {
for (GraphicsAllocation *alloc : kernelSvmGfxAllocations) {
if (allocationForCacheFlush(alloc)) {
out.push_back(alloc);
}
}
}
}
bool Kernel::allocationForCacheFlush(GraphicsAllocation *argAllocation) const {
return argAllocation->isFlushL3Required();
}
void Kernel::addAllocationToCacheFlushVector(uint32_t argIndex, GraphicsAllocation *argAllocation) {
if (argAllocation == nullptr) {
kernelArgRequiresCacheFlush[argIndex] = nullptr;
} else {
if (allocationForCacheFlush(argAllocation)) {
kernelArgRequiresCacheFlush[argIndex] = argAllocation;
} else {
kernelArgRequiresCacheFlush[argIndex] = nullptr;
}
}
}
void Kernel::setReflectionSurfaceBlockBtOffset(uint32_t blockID, uint32_t offset) {
DEBUG_BREAK_IF(blockID >= program->getBlockKernelManager()->getCount());
ReflectionSurfaceHelper::setKernelAddressDataBtOffset(getKernelReflectionSurface()->getUnderlyingBuffer(), blockID, offset);
}
bool Kernel::checkIfIsParentKernelAndBlocksUsesPrintf() {
return isParentKernel && getProgram()->getBlockKernelManager()->getIfBlockUsesPrintf();
}
uint64_t Kernel::getKernelStartOffset(
const bool localIdsGenerationByRuntime,
const bool kernelUsesLocalIds,
const bool isCssUsed) const {
uint64_t kernelStartOffset = 0;
if (kernelInfo.getGraphicsAllocation()) {
kernelStartOffset = kernelInfo.getGraphicsAllocation()->getGpuAddressToPatch();
if (localIdsGenerationByRuntime == false && kernelUsesLocalIds == true) {
kernelStartOffset += kernelInfo.kernelDescriptor.entryPoints.skipPerThreadDataLoad;
}
}
kernelStartOffset += getStartOffset();
auto &hardwareInfo = getHardwareInfo();
auto &hwHelper = HwHelper::get(hardwareInfo.platform.eRenderCoreFamily);
if (isCssUsed && hwHelper.isOffsetToSkipSetFFIDGPWARequired(hardwareInfo)) {
kernelStartOffset += kernelInfo.kernelDescriptor.entryPoints.skipSetFFIDGP;
}
return kernelStartOffset;
}
void *Kernel::patchBindlessSurfaceState(NEO::GraphicsAllocation *alloc, uint32_t bindless) {
auto &hwHelper = HwHelper::get(getDevice().getHardwareInfo().platform.eRenderCoreFamily);
auto surfaceStateSize = hwHelper.getRenderSurfaceStateSize();
NEO::BindlessHeapsHelper *bindlessHeapsHelper = getDevice().getDevice().getBindlessHeapsHelper();
auto ssInHeap = bindlessHeapsHelper->allocateSSInHeap(surfaceStateSize, alloc, NEO::BindlessHeapsHelper::GLOBAL_SSH);
auto patchLocation = ptrOffset(getCrossThreadData(), bindless);
auto patchValue = hwHelper.getBindlessSurfaceExtendedMessageDescriptorValue(static_cast<uint32_t>(ssInHeap.surfaceStateOffset));
patchWithRequiredSize(patchLocation, sizeof(patchValue), patchValue);
return ssInHeap.ssPtr;
}
void Kernel::setAdditionalKernelExecInfo(uint32_t additionalKernelExecInfo) {
this->additionalKernelExecInfo = additionalKernelExecInfo;
}
uint32_t Kernel::getAdditionalKernelExecInfo() const {
return this->additionalKernelExecInfo;
}
bool Kernel::requiresWaDisableRccRhwoOptimization() const {
auto &hardwareInfo = getHardwareInfo();
auto &hwHelper = HwHelper::get(hardwareInfo.platform.eRenderCoreFamily);
auto rootDeviceIndex = getDevice().getRootDeviceIndex();
if (hwHelper.isWaDisableRccRhwoOptimizationRequired() && isUsingSharedObjArgs()) {
for (auto &arg : getKernelArguments()) {
auto clMemObj = static_cast<cl_mem>(arg.object);
auto memObj = castToObject<MemObj>(clMemObj);
if (memObj && memObj->peekSharingHandler()) {
auto allocation = memObj->getGraphicsAllocation(rootDeviceIndex);
for (uint32_t handleId = 0u; handleId < allocation->getNumGmms(); handleId++) {
if (allocation->getGmm(handleId)->gmmResourceInfo->getResourceFlags()->Info.MediaCompressed) {
return true;
}
}
}
}
}
return false;
}
const HardwareInfo &Kernel::getHardwareInfo() const {
return getDevice().getHardwareInfo();
}
void Kernel::setWorkDim(uint32_t workDim) {
patchNonPointer(getCrossThreadDataRef(), getDescriptor().payloadMappings.dispatchTraits.workDim, workDim);
}
void Kernel::setGlobalWorkOffsetValues(uint32_t globalWorkOffsetX, uint32_t globalWorkOffsetY, uint32_t globalWorkOffsetZ) {
patchVecNonPointer(getCrossThreadDataRef(),
getDescriptor().payloadMappings.dispatchTraits.globalWorkOffset,
{globalWorkOffsetX, globalWorkOffsetY, globalWorkOffsetZ});
}
void Kernel::setGlobalWorkSizeValues(uint32_t globalWorkSizeX, uint32_t globalWorkSizeY, uint32_t globalWorkSizeZ) {
patchVecNonPointer(getCrossThreadDataRef(),
getDescriptor().payloadMappings.dispatchTraits.globalWorkSize,
{globalWorkSizeX, globalWorkSizeY, globalWorkSizeZ});
}
void Kernel::setLocalWorkSizeValues(uint32_t localWorkSizeX, uint32_t localWorkSizeY, uint32_t localWorkSizeZ) {
patchVecNonPointer(getCrossThreadDataRef(),
getDescriptor().payloadMappings.dispatchTraits.localWorkSize,
{localWorkSizeX, localWorkSizeY, localWorkSizeZ});
}
void Kernel::setLocalWorkSize2Values(uint32_t localWorkSizeX, uint32_t localWorkSizeY, uint32_t localWorkSizeZ) {
patchVecNonPointer(getCrossThreadDataRef(),
getDescriptor().payloadMappings.dispatchTraits.localWorkSize2,
{localWorkSizeX, localWorkSizeY, localWorkSizeZ});
}
void Kernel::setEnqueuedLocalWorkSizeValues(uint32_t localWorkSizeX, uint32_t localWorkSizeY, uint32_t localWorkSizeZ) {
patchVecNonPointer(getCrossThreadDataRef(),
getDescriptor().payloadMappings.dispatchTraits.enqueuedLocalWorkSize,
{localWorkSizeX, localWorkSizeY, localWorkSizeZ});
}
void Kernel::setNumWorkGroupsValues(uint32_t numWorkGroupsX, uint32_t numWorkGroupsY, uint32_t numWorkGroupsZ) {
patchVecNonPointer(getCrossThreadDataRef(),
getDescriptor().payloadMappings.dispatchTraits.numWorkGroups,
{numWorkGroupsX, numWorkGroupsY, numWorkGroupsZ});
}
bool Kernel::isLocalWorkSize2Patchable() {
const auto &localWorkSize2 = getDescriptor().payloadMappings.dispatchTraits.localWorkSize2;
return isValidOffset(localWorkSize2[0]) && isValidOffset(localWorkSize2[1]) && isValidOffset(localWorkSize2[2]);
}
uint32_t Kernel::getMaxKernelWorkGroupSize() const {
return maxKernelWorkGroupSize;
}
uint32_t Kernel::getSlmTotalSize() const {
return slmTotalSize;
}
bool Kernel::areMultipleSubDevicesInContext() const {
auto context = program->getContextPtr();
return context ? context->containsMultipleSubDevices(clDevice.getRootDeviceIndex()) : false;
}
} // namespace NEO
Reference in New Issue View Git Blame Copy Permalink