compute-runtime/runtime/scheduler/scheduler_kernel.cpp

77 lines
4.1 KiB
C++

/*
* Copyright (c) 2017 - 2018, Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included
* in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*/
#include "runtime/helpers/hw_helper.h"
#include "runtime/scheduler/scheduler_kernel.h"
namespace OCLRT {
void SchedulerKernel::setArgs(GraphicsAllocation *queue,
GraphicsAllocation *commandsStack,
GraphicsAllocation *eventsPool,
GraphicsAllocation *secondaryBatchBuffer,
GraphicsAllocation *dsh,
GraphicsAllocation *reflectionSurface,
GraphicsAllocation *queueStorageBuffer,
GraphicsAllocation *ssh,
GraphicsAllocation *debugQueue) {
setArgSvmAlloc(0, queue->getUnderlyingBuffer(), queue);
setArgSvmAlloc(1, commandsStack->getUnderlyingBuffer(), commandsStack);
setArgSvmAlloc(2, eventsPool->getUnderlyingBuffer(), eventsPool);
setArgSvmAlloc(3, secondaryBatchBuffer->getUnderlyingBuffer(), secondaryBatchBuffer);
setArgSvmAlloc(4, dsh->getUnderlyingBuffer(), dsh);
setArgSvmAlloc(5, reflectionSurface->getUnderlyingBuffer(), reflectionSurface);
setArgSvmAlloc(6, queueStorageBuffer->getUnderlyingBuffer(), queueStorageBuffer);
setArgSvmAlloc(7, ssh->getUnderlyingBuffer(), ssh);
if (debugQueue)
setArgSvmAlloc(8, debugQueue->getUnderlyingBuffer(), debugQueue);
DBG_LOG(PrintEMDebugInformation,
"Scheduler Surfaces: \nqueue=", queue->getUnderlyingBuffer(), " \nstack=", commandsStack->getUnderlyingBuffer(),
" \nevents=", eventsPool->getUnderlyingBuffer(), " \nslb=", secondaryBatchBuffer->getUnderlyingBuffer(), "\ndsh=", dsh->getUnderlyingBuffer(),
" \nkrs=", reflectionSurface->getUnderlyingBuffer(), " \nstorage=", queueStorageBuffer->getUnderlyingBuffer(), "\nssh=", ssh->getUnderlyingBuffer());
}
void SchedulerKernel::computeGws() {
auto &devInfo = device.getDeviceInfo();
auto &hwInfo = device.getHardwareInfo();
auto &helper = HwHelper::get(hwInfo.pPlatform->eRenderCoreFamily);
size_t hWThreadsPerSubSlice = devInfo.maxComputUnits / hwInfo.pSysInfo->SubSliceCount;
size_t wkgsPerSubSlice = hWThreadsPerSubSlice / PARALLEL_SCHEDULER_HWTHREADS_IN_HW_GROUP20;
wkgsPerSubSlice = std::min(wkgsPerSubSlice, helper.getMaxBarrierRegisterPerSlice());
gws = wkgsPerSubSlice * hwInfo.pSysInfo->SubSliceCount * PARALLEL_SCHEDULER_HWTHREADS_IN_HW_GROUP20 * PARALLEL_SCHEDULER_COMPILATION_SIZE_20;
if (device.isSimulation()) {
gws = PARALLEL_SCHEDULER_HWTHREADS_IN_HW_GROUP20 * PARALLEL_SCHEDULER_COMPILATION_SIZE_20;
}
if (DebugManager.flags.SchedulerGWS.get() != 0) {
DEBUG_BREAK_IF(DebugManager.flags.SchedulerGWS.get() % (PARALLEL_SCHEDULER_HWTHREADS_IN_HW_GROUP20 * PARALLEL_SCHEDULER_COMPILATION_SIZE_20) != 0);
gws = DebugManager.flags.SchedulerGWS.get();
}
DBG_LOG(PrintEMDebugInformation, "Scheduler GWS: ", gws);
printDebugString(DebugManager.flags.PrintDebugMessages.get(), stderr, "Scheduler GWS: ", gws);
}
} // namespace OCLRT