P4 to Git Change 1370387 by wchau@wchau_OCL_boltzmann on 2017/02/07 17:00:35

SWDEV-103424 - [ROCm CQE][OCL] OCLRuntime - OCLCreateBuffer tests are failing.  The failure is due to AQL cannot support global size > 32bit range.  Adding dispatch split support for ROCm, similar to that of GSL (CL#1159349), to resolve the issue.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#13 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#56 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#8 edit
... //depot/stg/opencl/drivers/opencl/tests/ocltst/module/runtime/OCLCreateBuffer.cpp#6 edit
This commit is contained in:
foreman
2017-02-07 17:13:15 -05:00
parent ef61b25945
commit dadd21fced
4 changed files with 305 additions and 216 deletions
@@ -1080,6 +1080,7 @@ HSAILProgram::setKernels_LC(amd::option::Options *options, void* binary, size_t
return false;
}
aKernel->setUniformWorkGroupSize(options->oVariables->UniformWorkGroupSize);
aKernel->setInternalKernelFlag(compileOptions_.find("-cl-internal-kernel") != std::string::npos);
kernels()[kernelName] = aKernel;
}
@@ -1336,6 +1337,7 @@ HSAILProgram::linkImpl(amd::option::Options *options)
return false;
}
aKernel->setUniformWorkGroupSize(options->oVariables->UniformWorkGroupSize);
aKernel->setInternalKernelFlag(compileOptions_.find("-cl-internal-kernel") != std::string::npos);
kernels()[kernelName] = aKernel;
}
saveBinaryAndSetType(TYPE_EXECUTABLE);