perf tests: Don't tail call optimize in unwind test
[linux-2.6-microblaze.git] / tools / perf / tests / openat-syscall-all-cpus.c
index 93c1765..71f85e2 100644 (file)
@@ -44,7 +44,7 @@ int test__openat_syscall_event_on_all_cpus(struct test *test __maybe_unused, int
 
        CPU_ZERO(&cpu_set);
 
-       evsel = perf_evsel__newtp("syscalls", "sys_enter_openat");
+       evsel = evsel__newtp("syscalls", "sys_enter_openat");
        if (IS_ERR(evsel)) {
                tracing_path__strerror_open_tp(errno, errbuf, sizeof(errbuf), "syscalls", "sys_enter_openat");
                pr_debug("%s\n", errbuf);
@@ -90,8 +90,8 @@ int test__openat_syscall_event_on_all_cpus(struct test *test __maybe_unused, int
         * we use the auto allocation it will allocate just for 1 cpu,
         * as we start by cpu 0.
         */
-       if (perf_evsel__alloc_counts(evsel, cpus->nr, 1) < 0) {
-               pr_debug("perf_evsel__alloc_counts(ncpus=%d)\n", cpus->nr);
+       if (evsel__alloc_counts(evsel, cpus->nr, 1) < 0) {
+               pr_debug("evsel__alloc_counts(ncpus=%d)\n", cpus->nr);
                goto out_close_fd;
        }
 
@@ -103,21 +103,21 @@ int test__openat_syscall_event_on_all_cpus(struct test *test __maybe_unused, int
                if (cpus->map[cpu] >= CPU_SETSIZE)
                        continue;
 
-               if (perf_evsel__read_on_cpu(evsel, cpu, 0) < 0) {
-                       pr_debug("perf_evsel__read_on_cpu\n");
+               if (evsel__read_on_cpu(evsel, cpu, 0) < 0) {
+                       pr_debug("evsel__read_on_cpu\n");
                        err = -1;
                        break;
                }
 
                expected = nr_openat_calls + cpu;
                if (perf_counts(evsel->counts, cpu, 0)->val != expected) {
-                       pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls on cpu %d, got %" PRIu64 "\n",
+                       pr_debug("evsel__read_on_cpu: expected to intercept %d calls on cpu %d, got %" PRIu64 "\n",
                                 expected, cpus->map[cpu], perf_counts(evsel->counts, cpu, 0)->val);
                        err = -1;
                }
        }
 
-       perf_evsel__free_counts(evsel);
+       evsel__free_counts(evsel);
 out_close_fd:
        perf_evsel__close_fd(&evsel->core);
 out_evsel_delete: