[Depend-intel#333] Convert a subset of GPU dialect ops to the GPU OpenCL runtime calls

Author: AndreyPavlenko

include/gc/ExecutionEngine/GPURuntime/GpuOclRuntime.h

@@ -0,0 +1,29 @@
+//===-- GpuOclRuntime.h - GPU OpenCL runtime --------------------*- C++ -*-===//
+//
+// This file is licensed under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef GC_GPUOCLRUNTIME_H
+#define GC_GPUOCLRUNTIME_H
+
+namespace mlir::gc::gpu {
+constexpr char GPU_OCL_MALLOC[] = "gcGpuOclMalloc";
+constexpr char GPU_OCL_DEALLOC[] = "gcGpuOclDealloc";
+constexpr char GPU_OCL_MEMCPY[] = "gcGpuOclMemcpy";
+constexpr char GPU_OCL_KERNEL_CREATE[] = "gcGpuOclKernelCreate";
+constexpr char GPU_OCL_KERNEL_DESTROY[] = "gcGpuOclKernelDestroy";
+constexpr char GPU_OCL_KERNEL_LAUNCH[] = "gcGpuOclKernelLaunch";
+constexpr char GPU_OCL_MOD_DESTRUCTOR[] = "gcGpuOclModuleDestructor";
+} // namespace mlir::gc::gpu
+
+#ifndef GC_GPU_OCL_CONST_ONLY
+
+// TBD
+
+#else
+#undef GC_GPU_OCL_CONST_ONLY
+#endif
+#endif

include/gc/Transforms/Passes.td

@@ -93,6 +93,13 @@ def LinalgToXeGPU : Pass<"linalg-to-xegpu", "func::FuncOp"> {
                "DPAS register block sizes MxNxK">,
   ];
 }
+
+def GpuToGpuOcl : Pass<"gpu-to-gpuocl", "ModuleOp"> {
+  let summary = "Convert the GPU operations to GpuOclRuntime calls.";
+  let description = [{
+    Convert the gpu alloc, dealloc, memcpy and launch operations to GpuOclRuntime calls.
+  }];
+}
 #endif // GC_USE_IMEX
 
 def IterativeTilingAndFusion : Pass<"iterative-tiling-and-fusion",

lib/gc/Transforms/GPU/CMakeLists.txt

@@ -1,4 +1,5 @@
 gc_add_mlir_library(GcGpuPasses
+  GpuToGpuOcl.cpp
   LinalgToXeGPU.cpp
   Pipeline.cpp