[SYCLomatic #2634] Fix analysis-mode test error due to feature changes (#872)

Signed-off-by: Ziran Zhang <[email protected]>

Author: ziranzha

behavior_tests/src/analysis-mode/do_test.py

@@ -32,18 +32,13 @@ def migrate_test():
                         ["APIs/Types - Medium manual effort for code fixing."],
                         ["lines of code", "will not be automatically migrated."],
                         ["APIs/Types - High manual effort for code fixing."],
-                        ["Total Project:"],
-                        ["lines of code", "will be automatically migrated."],
-                        ["APIs/Types - No manual effort."],
-                        ["APIs/Types - Low manual effort for checking and code fixing."],
-                        ["APIs/Types - Medium manual effort for code fixing."],
-                        ["lines of code", "will not be automatically migrated."],
-                        ["APIs/Types - High manual effort for code fixing."],
                         ["Library Dependencies of SYCL Project:"],
-                        ["oneAPI Math Kernel Library (oneMKL)"],
-                        ["oneAPI DPC++ Library (oneDPL)"],
-                        ["oneAPI Deep Neural Network Library (oneDNN)"],
-                        ["oneAPI Collective Communications Library (oneCCL)"]]
+                        ["The Intel oneAPI Math Kernel Library (oneMKL)"],
+                        ["The Intel oneAPI DPC++ Library (oneDPL)"],
+                        ["The Intel oneAPI Deep Neural Network Library (oneDNN)"],
+                        ["The Intel oneAPI Collective Communications Library (oneCCL)"],
+                        ["Unsupported APIs and Types:"],
+                        ["cudaDeviceGetPCIBusId Occurrences 1 times"]]
 
     ret_lines = test_config.command_output.splitlines()
     res = True

behavior_tests/src/analysis-mode/test.cu

@@ -21,6 +21,7 @@ __global__ void kernel(int *a) {
 void foo() {
     int *a;
     size_t b, c;
+    cudaDeviceGetPCIBusId(nullptr, 0, 0);
     cudaMemGetInfo(&b, &c);
     cudaMalloc(&a, sizeof(int));
     cudaFree(a);