[OMPT] Fix cast and printf of wait_id in lock test

[Hahnfeld]

Summary

This didn't work on 32 bit platforms.

[protze.joachim]

I would have expected, that

printf("%" PRIu64 ": &lock: %lu\n", ompt_get_thread_data()->value, (unsigned long) &lock);

would also fix the issue.

But (uint64_t) &lock actually extends the 32-bit address to 64-bit signed integer: fffef9c8 -> fffffffffffef9c8
This is the same value as provided by the callback function.

The spec only requires a unique identifier here, so this behavior is fine.

[Hahnfeld]

In D41853#971265, @protze.joachim wrote:

But (uint64_t) &lock actually extends the 32-bit address to 64-bit signed integer: fffef9c8 -> fffffffffffef9c8
This is the same value as provided by the callback function.

Yes, that's the "problem": For FileCheck to succeed, the value needs to be the same.

openmp/trunk/runtime/test/ompt/synchronization/lock.c

 // RUN: %libomp-compile-and-run | FileCheck %s
 // REQUIRES: ompt
 #include "callback.h"
 #include <omp.h>
 
 int main()
 {
   //need to use an OpenMP construct so that OMPT will be initalized
   #pragma omp parallel num_threads(1)
     print_ids(0);
 
   omp_lock_t lock;
-  printf("%" PRIu64 ": &lock: %lli\n", ompt_get_thread_data()->value, (long long) &lock);
+  printf("%" PRIu64 ": &lock: %" PRIu64 "\n", ompt_get_thread_data()->value, (uint64_t) &lock);
   omp_init_lock(&lock);
   print_current_address(1);
   omp_set_lock(&lock);
   print_current_address(2);
   omp_unset_lock(&lock);
   print_current_address(3);
   omp_destroy_lock(&lock);
   print_current_address(4);