Re: [PATCH] tracing/timerlat: Check tlat_var for NULL in timerlat_fd_release

[Steven Rostedt]

On Thu, 22 Aug 2024 08:20:52 -0300
"Luis Claudio R. Goncalves" <lgonc...@redhat.com> wrote:

> You mean the 
> 
> +     if (!tlat_var->kthread) {
> +             /* the fd has been closed already */
> 
> bit or the kthread handling in rtla itself?
> 
> As Tomas already said, thank you for testing and reviewing the suggested fix!

The close only happens once, but the kthread does get freed elsewhere.

I used my new persistent ring buffer to debug this! I'll be posting some
more patches that allow trace_printk() and friends to use the persistent
ring buffer (or any instance) soon.

For this bug, I would recommend instead doing:

diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c
index 66a871553d4a..53de719f35cb 100644
--- a/kernel/trace/trace_osnoise.c
+++ b/kernel/trace/trace_osnoise.c
@@ -2579,7 +2587,8 @@ static int timerlat_fd_release(struct inode *inode, 
struct file *file)
        osn_var = per_cpu_ptr(&per_cpu_osnoise_var, cpu);
        tlat_var = per_cpu_ptr(&per_cpu_timerlat_var, cpu);
 
-       hrtimer_cancel(&tlat_var->timer);
+       if (tlat_var->kthread)
+               hrtimer_cancel(&tlat_var->timer);
        memset(tlat_var, 0, sizeof(*tlat_var));
 
        osn_var->sampling = 0;

-- Steve