intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-25 13:33:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

Store single KernelInfo in Kernel

Browse Source 
remove root device index from Kernel's methods

Related-To: NEO-5001
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
This commit is contained in:
Mateusz Jablonski
2021-03-22 15:26:03 +00:00
committed by Compute-Runtime-Automation
parent ecceddcab6
commit 7098e9c5f2
136 changed files with 1043 additions and 1192 deletions
Download Patch File Download Diff File Expand all files Collapse all files

85
opencl/source/kernel/kernel.h
View File

@@ -88,11 +88,11 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
const void *argVal);
template <typename kernel_t = Kernel, typename program_t = Program>
static kernel_t *create(program_t *program, const KernelInfoContainer &kernelInfos, ClDevice &clDevice, cl_int *errcodeRet) {
static kernel_t *create(program_t *program, const KernelInfo &kernelInfo, ClDevice &clDevice, cl_int *errcodeRet) {
cl_int retVal;
kernel_t *pKernel = nullptr;
pKernel = new kernel_t(program, kernelInfos, clDevice);
pKernel = new kernel_t(program, kernelInfo, clDevice);
retVal = pKernel->initialize();
if (retVal != CL_SUCCESS) {
@@ -107,7 +107,7 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
if (FileLoggerInstance().enabled()) {
std::string source;
program->getSource(source);
FileLoggerInstance().dumpKernel(kernelInfos[program->getDevices()[0]->getRootDeviceIndex()]->kernelDescriptor.kernelMetadata.kernelName, source);
FileLoggerInstance().dumpKernel(kernelInfo.kernelDescriptor.kernelMetadata.kernelName, source);
}
return pKernel;
@@ -152,7 +152,7 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
cl_int getInfo(cl_kernel_info paramName, size_t paramValueSize,
void *paramValue, size_t *paramValueSizeRet) const;
void getAdditionalInfo(cl_kernel_info paramName, const void *&paramValue, size_t &paramValueSizeRet) const;
void getAdditionalWorkGroupInfo(cl_kernel_work_group_info paramName, const void *&paramValue, size_t &paramValueSizeRet, uint32_t rootDeviceIndex) const;
void getAdditionalWorkGroupInfo(cl_kernel_work_group_info paramName, const void *&paramValue, size_t &paramValueSizeRet) const;
cl_int getArgInfo(cl_uint argIndx, cl_kernel_arg_info paramName,
size_t paramValueSize, void *paramValue, size_t *paramValueSizeRet) const;
@@ -165,13 +165,13 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
size_t paramValueSize, void *paramValue,
size_t *paramValueSizeRet) const;
const void *getKernelHeap(uint32_t rootDeviceIndex) const;
void *getSurfaceStateHeap(uint32_t rootDeviceIndex) const;
const void *getDynamicStateHeap(uint32_t rootDeviceIndex) const;
const void *getKernelHeap() const;
void *getSurfaceStateHeap() const;
const void *getDynamicStateHeap() const;
size_t getKernelHeapSize(uint32_t rootDeviceIndex) const;
size_t getSurfaceStateHeapSize(uint32_t rootDeviceIndex) const;
size_t getDynamicStateHeapSize(uint32_t rootDeviceIndex) const;
size_t getKernelHeapSize() const;
size_t getSurfaceStateHeapSize() const;
size_t getDynamicStateHeapSize() const;
size_t getNumberOfBindingTableStates() const;
size_t getBindingTableOffset() const {
return localBindingTableOffset;
@@ -180,9 +180,9 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
void resizeSurfaceStateHeap(void *pNewSsh, size_t newSshSize, size_t newBindingTableCount, size_t newBindingTableOffset);
void substituteKernelHeap(const Device &device, void *newKernelHeap, size_t newKernelHeapSize);
bool isKernelHeapSubstituted(uint32_t rootDeviceIndex) const;
uint64_t getKernelId(uint32_t rootDeviceIndex) const;
void setKernelId(uint32_t rootDeviceIndex, uint64_t newKernelId);
bool isKernelHeapSubstituted() const;
uint64_t getKernelId() const;
void setKernelId(uint64_t newKernelId);
uint32_t getStartOffset() const;
void setStartOffset(uint32_t offset);
@@ -194,15 +194,12 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
return kernelArguments.size();
}
bool requiresSshForBuffers(uint32_t rootDeviceIndex) const {
return getKernelInfo(rootDeviceIndex).requiresSshForBuffers;
bool requiresSshForBuffers() const {
return kernelInfo.requiresSshForBuffers;
}
const KernelInfo &getKernelInfo(uint32_t rootDeviceIndex) const {
return *kernelInfos[rootDeviceIndex];
}
const KernelInfoContainer &getKernelInfos() const {
return kernelInfos;
const KernelInfo &getKernelInfo() const {
return kernelInfo;
}
Context &getContext() const {
@@ -211,12 +208,12 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
Program *getProgram() const { return program; }
uint32_t getScratchSize(uint32_t rootDeviceIndex) {
return getKernelInfo(rootDeviceIndex).kernelDescriptor.kernelAttributes.perThreadScratchSize[0];
uint32_t getScratchSize() {
return kernelInfo.kernelDescriptor.kernelAttributes.perThreadScratchSize[0];
}
uint32_t getPrivateScratchSize(uint32_t rootDeviceIndex) {
return getKernelInfo(rootDeviceIndex).kernelDescriptor.kernelAttributes.perThreadScratchSize[1];
uint32_t getPrivateScratchSize() {
return kernelInfo.kernelDescriptor.kernelAttributes.perThreadScratchSize[1];
}
void createReflectionSurface();
@@ -225,8 +222,8 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
void patchDefaultDeviceQueue(DeviceQueue *devQueue);
void patchEventPool(DeviceQueue *devQueue);
void patchBlocksSimdSize(uint32_t rootDeviceIndex);
bool usesSyncBuffer(uint32_t rootDeviceIndex);
void patchBlocksSimdSize();
bool usesSyncBuffer();
void patchSyncBuffer(Device &device, GraphicsAllocation *gfxAllocation, size_t bufferOffset);
void patchBindlessSurfaceStateOffsets(const Device &device, const size_t sshOffset);
@@ -295,7 +292,7 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
const SimpleKernelArgInfo &getKernelArgInfo(uint32_t argIndex) const;
bool getAllowNonUniform() const { return program->getAllowNonUniform(); }
bool isVmeKernel() const { return getDefaultKernelInfo().isVmeWorkload; }
bool isVmeKernel() const { return kernelInfo.isVmeWorkload; }
bool requiresSpecialPipelineSelectMode() const { return specialPipelineSelectMode; }
void performKernelTunning(CommandStreamReceiver &commandStreamReceiver, const Vec3<size_t> &lws, const Vec3<size_t> &gws, const Vec3<size_t> &offsets, TimestampPacketContainer *timestampContainer);
@@ -309,7 +306,7 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
bool isUsingSharedObjArgs() const { return usingSharedObjArgs; }
bool hasUncacheableStatelessArgs() const { return statelessUncacheableArgsCount > 0; }
bool hasPrintfOutput(uint32_t rootDeviceIndex) const;
bool hasPrintfOutput() const;
void setReflectionSurfaceBlockBtOffset(uint32_t blockID, uint32_t offset);
@@ -334,12 +331,12 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
bool checkIfIsParentKernelAndBlocksUsesPrintf();
bool is32Bit(uint32_t rootDeviceIndex) const {
return getKernelInfo(rootDeviceIndex).gpuPointerSize == 4;
bool is32Bit() const {
return kernelInfo.gpuPointerSize == 4;
}
size_t getPerThreadSystemThreadSurfaceSize(uint32_t rootDeviceIndex) const {
return getKernelInfo(rootDeviceIndex).kernelDescriptor.kernelAttributes.perThreadSystemThreadSurfaceSize;
size_t getPerThreadSystemThreadSurfaceSize() const {
return kernelInfo.kernelDescriptor.kernelAttributes.perThreadSystemThreadSurfaceSize;
}
std::vector<PatchInfoData> &getPatchInfoDataList() { return patchInfoDataList; };
@@ -347,12 +344,12 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
return usingImagesOnly;
}
void fillWithKernelObjsForAuxTranslation(KernelObjsForAuxTranslation &kernelObjsForAuxTranslation, uint32_t rootDeviceIndex);
void fillWithKernelObjsForAuxTranslation(KernelObjsForAuxTranslation &kernelObjsForAuxTranslation);
MOCKABLE_VIRTUAL bool requiresCacheFlushCommand(const CommandQueue &commandQueue) const;
using CacheFlushAllocationsVec = StackVec<GraphicsAllocation *, 32>;
void getAllocationsForCacheFlush(CacheFlushAllocationsVec &out, uint32_t rootDeviceIndex) const;
void getAllocationsForCacheFlush(CacheFlushAllocationsVec &out) const;
void setAuxTranslationDirection(AuxTranslationDirection auxTranslationDirection) {
this->auxTranslationDirection = auxTranslationDirection;
@@ -377,20 +374,18 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
uint64_t getKernelStartOffset(
const bool localIdsGenerationByRuntime,
const bool kernelUsesLocalIds,
const bool isCssUsed,
uint32_t rootDeviceIndex) const;
const bool isCssUsed) const;
bool requiresPerDssBackedBuffer(uint32_t rootDeviceIndex) const;
bool requiresLimitedWorkgroupSize(uint32_t rootDeviceIndex) const;
bool requiresPerDssBackedBuffer() const;
bool requiresLimitedWorkgroupSize() const;
bool isKernelDebugEnabled() const { return debugEnabled; }
int32_t setAdditionalKernelExecInfoWithParam(uint32_t paramName, size_t paramValueSize, const void *paramValue);
void setAdditionalKernelExecInfo(uint32_t additionalKernelExecInfo);
uint32_t getAdditionalKernelExecInfo() const;
MOCKABLE_VIRTUAL bool requiresWaDisableRccRhwoOptimization(uint32_t rootDeviceIndex) const;
MOCKABLE_VIRTUAL bool requiresWaDisableRccRhwoOptimization() const;
const ClDeviceVector &getDevices() const {
return program->getDevices();
}
const KernelInfo &getDefaultKernelInfo() const;
void setGlobalWorkOffsetValues(uint32_t globalWorkOffsetX, uint32_t globalWorkOffsetY, uint32_t globalWorkOffsetZ);
void setGlobalWorkSizeValues(uint32_t globalWorkSizeX, uint32_t globalWorkSizeY, uint32_t globalWorkSizeZ);
@@ -482,7 +477,7 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
void
makeArgsResident(CommandStreamReceiver &commandStreamReceiver);
void *patchBufferOffset(const KernelArgInfo &argInfo, void *svmPtr, GraphicsAllocation *svmAlloc, uint32_t rootDeviceIndex);
void *patchBufferOffset(const KernelArgInfo &argInfo, void *svmPtr, GraphicsAllocation *svmAlloc);
void patchWithImplicitSurface(void *ptrToPatchInCrossThreadData, GraphicsAllocation &allocation, const Device &device, const ArgDescPointer &arg);
// Sets-up both crossThreadData and ssh for given implicit (private/constant, etc.) allocation
@@ -490,21 +485,21 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
void patchWithImplicitSurface(void *ptrToPatchInCrossThreadData, GraphicsAllocation &allocation, const Device &device, const PatchTokenT &patch);
void getParentObjectCounts(ObjectCounts &objectCount);
Kernel(Program *programArg, const KernelInfoContainer &kernelInfsoArg, ClDevice &clDevice, bool schedulerKernel = false);
Kernel(Program *programArg, const KernelInfo &kernelInfo, ClDevice &clDevice, bool schedulerKernel = false);
void provideInitializationHints();
void patchBlocksCurbeWithConstantValues();
void resolveArgs();
void reconfigureKernel(uint32_t rootDeviceIndex);
void reconfigureKernel();
bool hasDirectStatelessAccessToHostMemory() const;
bool hasIndirectStatelessAccessToHostMemory() const;
void addAllocationToCacheFlushVector(uint32_t argIndex, GraphicsAllocation *argAllocation);
bool allocationForCacheFlush(GraphicsAllocation *argAllocation) const;
const HardwareInfo &getHardwareInfo(uint32_t rootDeviceIndex) const;
const HardwareInfo &getHardwareInfo() const;
const ClDevice &getDevice() const {
return clDevice;
@@ -514,7 +509,7 @@ class Kernel : public ReferenceTrackedObject<Kernel> {
Program *program;
ClDevice &clDevice;
const ClDeviceVector &deviceVector;
const KernelInfoContainer kernelInfos;
const KernelInfo &kernelInfo;
std::vector<SimpleKernelArgInfo> kernelArguments;
std::vector<KernelArgHandler> kernelArgHandlers;