Re: [REGRESSION] Random oops on SG2042 with Linux 6.16-rc and dynamic ftrace

[Steven Rostedt <rostedt@goodmis.org>]

On Tue, 1 Jul 2025 14:27:32 +0200
Alexandre Ghiti <alex@ghiti.fr> wrote:

> @@ -149,16 +157,17 @@ int ftrace_init_nop(struct module *mod, struct 
> dyn_ftrace *rec)
>          unsigned int nops[2], offset;
>          int ret;
> 
> +       mutex_lock(&text_mutex);

This looks like a prime candidate for:

	guard(mutex)(&text_mutex);

>          ret = ftrace_rec_set_nop_ops(rec);
>          if (ret)
> -               return ret;
> +               goto end;

Then this change wouldn't be needed.

> 
>          offset = (unsigned long) &ftrace_caller - pc;
>          nops[0] = to_auipc_t0(offset);
>          nops[1] = RISCV_INSN_NOP4;
> 
> -       mutex_lock(&text_mutex);
>          ret = patch_insn_write((void *)pc, nops, 2 * MCOUNT_INSN_SIZE);
> +end:
>          mutex_unlock(&text_mutex);

and "end:" as well as the mutex_unlock() can be removed.

> 
>          return ret;
> 
> 
> Andy is also taking a look, I'll let him confirm the above fix is correct.
> 
> Thanks for the thorough report!

Yes, thank you Yao Zi for the very detailed bug report! That made
things so much easier.

-- Steve