Re: [PATCH 1/3] stacktrace: Remove superfluous WARN_ONCE() from save_stack_trace_tsk_reliable()

[Miroslav Benes]

On Fri, 31 May 2019, Petr Mladek wrote:

> WARN_ONCE() in the generic save_stack_trace_tsk_reliable() is superfluous.
> 
> The information is passed also via the return value. The only current
> user klp_check_stack() writes its own warning when the reliable stack
> traces are not supported. Other eventual users might want its own error
> handling as well.
> 
> Signed-off-by: Petr Mladek <pmladek@xxxxxxxx>
> Acked-by: Miroslav Benes <mbenes@xxxxxxx>
> Reviewed-by: Kamalesh Babulal <kamalesh@xxxxxxxxxxxxxxxxxx>
> ---
>  kernel/stacktrace.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/kernel/stacktrace.c b/kernel/stacktrace.c
> index 5667f1da3ede..8d088408928d 100644
> --- a/kernel/stacktrace.c
> +++ b/kernel/stacktrace.c
> @@ -259,7 +259,6 @@ __weak int
>  save_stack_trace_tsk_reliable(struct task_struct *tsk,
>  			      struct stack_trace *trace)
>  {
> -	WARN_ONCE(1, KERN_INFO "save_stack_tsk_reliable() not implemented yet.\n");
>  	return -ENOSYS;
>  }

Do we even need the weak function now after Thomas' changes to 
kernel/stacktrace.c?

- livepatch is the only user and it calls stack_trace_save_tsk_reliable()
- x86 defines CONFIG_ARCH_STACKWALK and CONFIG_HAVE_RELIABLE_STACKTRACE, 
  so it has stack_trace_save_tsk_reliable() implemented and it calls 
  arch_stack_walk_reliable()
- powerpc defines CONFIG_HAVE_RELIABLE_STACKTRACE and does not have 
  CONFIG_ARCH_STACKWALK. It also has stack_trace_save_tsk_reliable() 
  implemented and it calls save_stack_trace_tsk_reliable(), which is 
  implemented in arch/powerpc/
- all other archs do not have CONFIG_HAVE_RELIABLE_STACKTRACE and there is 
  stack_trace_save_tsk_reliable() returning ENOSYS for these cases in 
  include/linux/stacktrace.c

Miroslav