[SYCLomatic] Add query-api-mapping for 8 Runtime APIs:

clang/examples/DPCT/Runtime/cudaLaunchHostFunc.cu

@@ -0,0 +1,5 @@
+void test(cudaStream_t stream, cudaHostFn_t fn, void*userData) {
+  // Start
+  cudaLaunchHostFunc(stream/*cudaStream_t*/, fn/*cudaHostFn_t*/, userData/*void**/);
+  // End
+}

clang/examples/DPCT/Runtime/cudaMemcpy3DPeer.cu

@@ -0,0 +1,5 @@
+void test(const cudaMemcpy3DPeerParms*p) {
+  // Start
+  cudaMemcpy3DPeer(p/*const cudaMemcpy3DPeerParms**/);
+  // End
+}

clang/examples/DPCT/Runtime/cudaMemcpy3DPeerAsync.cu

@@ -0,0 +1,5 @@
+void test(const cudaMemcpy3DPeerParms*p, cudaStream_t stream) {
+  // Start
+  cudaMemcpy3DPeerAsync(p/*const cudaMemcpy3DPeerParms**/, stream/*cudaStream_t*/);
+  // End
+}

clang/examples/DPCT/Runtime/cudaMemcpyPeer.cu

@@ -0,0 +1,5 @@
+void test(void*dst, int dstDevice, const void*src, int srcDevice, size_t count) {
+  // Start
+  cudaMemcpyPeer(dst/*void**/, dstDevice/*int*/, src/*const void**/, srcDevice/*int*/, count/*size_t*/);
+  // End
+}

clang/examples/DPCT/Runtime/cudaMemcpyPeerAsync.cu

@@ -0,0 +1,5 @@
+void test(void*dst, int dstDevice, const void*src, int srcDevice, size_t count, cudaStream_t stream) {
+  // Start
+  cudaMemcpyPeerAsync(dst/*void**/, dstDevice/*int*/, src/*const void**/, srcDevice/*int*/, count/*size_t*/, stream/*cudaStream_t*/);
+  // End
+}

clang/examples/DPCT/Runtime/cudaProfilerStart.cu

@@ -0,0 +1,7 @@
+#include "cuda_profiler_api.h"
+
+void test() {
+  // Start
+  cudaProfilerStart();
+  // End
+}

clang/examples/DPCT/Runtime/cudaProfilerStop.cu

@@ -0,0 +1,7 @@
+#include "cuda_profiler_api.h"
+
+void test() {
+  // Start
+  cudaProfilerStop();
+  // End
+}

clang/test/dpct/query_api_mapping/Runtime/test-after10.cu

@@ -21,3 +21,15 @@
 // CUDASTREAMISCAPTURING-NEXT:                         ps /* enum cudaStreamCaptureStatus **/);
 // CUDASTREAMISCAPTURING-NEXT: Is migrated to (with the option --use-experimental-features=graph):
 // CUDASTREAMISCAPTURING-NEXT: *ps = s->ext_oneapi_get_state();
+
+
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaLaunchHostFunc | FileCheck %s -check-prefix=CUDALAUNCHHOSTFUNC
+// CUDALAUNCHHOSTFUNC: CUDA API:
+// CUDALAUNCHHOSTFUNC-NEXT:   cudaLaunchHostFunc(stream/*cudaStream_t*/, fn/*cudaHostFn_t*/, userData/*void**/);
+// CUDALAUNCHHOSTFUNC-NEXT: Is migrated to: 
+// CUDALAUNCHHOSTFUNC-NEXT:   stream->submit([&](sycl::handler &cgh) {
+// CUDALAUNCHHOSTFUNC-NEXT:     cgh.host_task([=](){
+// CUDALAUNCHHOSTFUNC-NEXT:       fn(userData);
+// CUDALAUNCHHOSTFUNC-NEXT:     });
+// CUDALAUNCHHOSTFUNC-NEXT:   });

clang/test/dpct/query_api_mapping/Runtime/test.cu

@@ -519,3 +519,41 @@
 // CUDARUNTIMEGETVERSION-NEXT:   cudaRuntimeGetVersion(pi /*int **/);
 // CUDARUNTIMEGETVERSION-NEXT: Is migrated to:
 // CUDARUNTIMEGETVERSION-NEXT:   *pi = dpct::get_major_version(dpct::get_current_device());
+
+
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaMemcpy3DPeer | FileCheck %s -check-prefix=CUDAMEMCPY3DPEER
+// CUDAMEMCPY3DPEER: CUDA API:
+// CUDAMEMCPY3DPEER-NEXT:   cudaMemcpy3DPeer(p/*const cudaMemcpy3DPeerParms**/);
+// CUDAMEMCPY3DPEER-NEXT: Is migrated to: 
+// CUDAMEMCPY3DPEER-NEXT:   dpct::dpct_memcpy(*p/*const cudaMemcpy3DPeerParms**/);
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaMemcpy3DPeerAsync | FileCheck %s -check-prefix=CUDAMEMCPY3DPEERASYNC
+// CUDAMEMCPY3DPEERASYNC: CUDA API:
+// CUDAMEMCPY3DPEERASYNC-NEXT:   cudaMemcpy3DPeerAsync(p/*const cudaMemcpy3DPeerParms**/, stream/*cudaStream_t*/);
+// CUDAMEMCPY3DPEERASYNC-NEXT: Is migrated to: 
+// CUDAMEMCPY3DPEERASYNC-NEXT:   dpct::async_dpct_memcpy(*p/*const cudaMemcpy3DPeerParms**/, *stream/*cudaStream_t*/);
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaMemcpyPeer | FileCheck %s -check-prefix=CUDAMEMCPYPEER
+// CUDAMEMCPYPEER: CUDA API:
+// CUDAMEMCPYPEER-NEXT:   cudaMemcpyPeer(dst/*void**/, dstDevice/*int*/, src/*const void**/, srcDevice/*int*/, count/*size_t*/);
+// CUDAMEMCPYPEER-NEXT: Is migrated to: 
+// CUDAMEMCPYPEER-NEXT:   dpct::dpct_memcpy(dst/*void**/, dstDevice/*int*/, src/*const void**/, srcDevice/*int*/, count/*size_t*/);
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaMemcpyPeerAsync | FileCheck %s -check-prefix=CUDAMEMCPYPEERASYNC
+// CUDAMEMCPYPEERASYNC: CUDA API:
+// CUDAMEMCPYPEERASYNC-NEXT:   cudaMemcpyPeerAsync(dst/*void**/, dstDevice/*int*/, src/*const void**/, srcDevice/*int*/, count/*size_t*/, stream/*cudaStream_t*/);
+// CUDAMEMCPYPEERASYNC-NEXT: Is migrated to: 
+// CUDAMEMCPYPEERASYNC-NEXT:   dpct::async_dpct_memcpy(dst/*void**/, dstDevice/*int*/, src/*const void**/, srcDevice/*int*/, count/*size_t*/, *stream/*cudaStream_t*/);
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaProfilerStart | FileCheck %s -check-prefix=CUDAPROFILERSTART
+// CUDAPROFILERSTART: CUDA API:
+// CUDAPROFILERSTART-NEXT:   cudaProfilerStart();
+// CUDAPROFILERSTART-NEXT: The API is Removed.
+// CUDAPROFILERSTART-EMPTY: 
+
+// RUN: dpct --cuda-include-path="%cuda-path/include" -query-api-mapping=cudaProfilerStop | FileCheck %s -check-prefix=CUDAPROFILERSTOP
+// CUDAPROFILERSTOP: CUDA API:
+// CUDAPROFILERSTOP-NEXT:   cudaProfilerStop();
+// CUDAPROFILERSTOP-NEXT: The API is Removed.
+// CUDAPROFILERSTOP-EMPTY: 

clang/test/dpct/query_api_mapping/test_all.cu

@@ -1485,6 +1485,7 @@
 // CHECK-NEXT: cudaIpcGetMemHandle
 // CHECK-NEXT: cudaIpcOpenMemHandle
 // CHECK-NEXT: cudaLaunchCooperativeKernel
+// CHECK-NEXT: cudaLaunchHostFunc
 // CHECK-NEXT: cudaLaunchKernel
 // CHECK-NEXT: cudaMalloc
 // CHECK-NEXT: cudaMalloc3D
@@ -1507,12 +1508,16 @@
 // CHECK-NEXT: cudaMemcpy2DToArrayAsync
 // CHECK-NEXT: cudaMemcpy3D
 // CHECK-NEXT: cudaMemcpy3DAsync
+// CHECK-NEXT: cudaMemcpy3DPeer
+// CHECK-NEXT: cudaMemcpy3DPeerAsync
 // CHECK-NEXT: cudaMemcpyArrayToArray
 // CHECK-NEXT: cudaMemcpyAsync
 // CHECK-NEXT: cudaMemcpyFromArray
 // CHECK-NEXT: cudaMemcpyFromArrayAsync
 // CHECK-NEXT: cudaMemcpyFromSymbol
 // CHECK-NEXT: cudaMemcpyFromSymbolAsync
+// CHECK-NEXT: cudaMemcpyPeer
+// CHECK-NEXT: cudaMemcpyPeerAsync
 // CHECK-NEXT: cudaMemcpyToArray
 // CHECK-NEXT: cudaMemcpyToArrayAsync
 // CHECK-NEXT: cudaMemcpyToSymbol
@@ -1526,6 +1531,8 @@
 // CHECK-NEXT: cudaOccupancyMaxActiveBlocksPerMultiprocessor
 // CHECK-NEXT: cudaPeekAtLastError
 // CHECK-NEXT: cudaPointerGetAttributes
+// CHECK-NEXT: cudaProfilerStart
+// CHECK-NEXT: cudaProfilerStop
 // CHECK-NEXT: cudaRuntimeGetVersion
 // CHECK-NEXT: cudaSetDevice
 // CHECK-NEXT: cudaSetDeviceFlags