tracing: Disable trace_printk() on post poned tests
authorSteven Rostedt (VMware) <rostedt@goodmis.org>
Thu, 20 Feb 2020 20:38:01 +0000 (15:38 -0500)
committerSteven Rostedt (VMware) <rostedt@goodmis.org>
Thu, 20 Feb 2020 22:43:58 +0000 (17:43 -0500)
The tracing seftests checks various aspects of the tracing infrastructure,
and one is filtering. If trace_printk() is active during a self test, it can
cause the filtering to fail, which will disable that part of the trace.

To keep the selftests from failing because of trace_printk() calls,
trace_printk() checks the variable tracing_selftest_running, and if set, it
does not write to the tracing buffer.

As some tracers were registered earlier in boot, the selftest they triggered
would fail because not all the infrastructure was set up for the full
selftest. Thus, some of the tests were post poned to when their
infrastructure was ready (namely file system code). The postpone code did
not set the tracing_seftest_running variable, and could fail if a
trace_printk() was added and executed during their run.

Cc: stable@vger.kernel.org
Fixes: 9afecfbb95198 ("tracing: Postpone tracer start-up tests till the system is more robust")
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
kernel/trace/trace.c

index 183b031a38280ce5ab582a7da07ed4bc0ea026af..a89c562ffb8fc58a8e0ba0a9c069087fb2502b51 100644 (file)
@@ -1837,6 +1837,7 @@ static __init int init_trace_selftests(void)
 
        pr_info("Running postponed tracer tests:\n");
 
+       tracing_selftest_running = true;
        list_for_each_entry_safe(p, n, &postponed_selftests, list) {
                /* This loop can take minutes when sanitizers are enabled, so
                 * lets make sure we allow RCU processing.
@@ -1859,6 +1860,7 @@ static __init int init_trace_selftests(void)
                list_del(&p->list);
                kfree(p);
        }
+       tracing_selftest_running = false;
 
  out:
        mutex_unlock(&trace_types_lock);